case study vs correlational study vs experiment

Want to create or adapt books like this? Learn more about how Pressbooks supports open publishing practices.

Chapter 3. Psychological Science

3.2 Psychologists Use Descriptive, Correlational, and Experimental Research Designs to Understand Behaviour

Learning objectives.

• Differentiate the goals of descriptive, correlational, and experimental research designs and explain the advantages and disadvantages of each.
• Explain the goals of descriptive research and the statistical techniques used to interpret it.
• Summarize the uses of correlational research and describe why correlational research cannot be used to infer causality.
• Review the procedures of experimental research and explain how it can be used to draw causal inferences.

Psychologists agree that if their ideas and theories about human behaviour are to be taken seriously, they must be backed up by data. However, the research of different psychologists is designed with different goals in mind, and the different goals require different approaches. These varying approaches, summarized in Table 3.2, are known as research designs . A research design  is the specific method a researcher uses to collect, analyze, and interpret data . Psychologists use three major types of research designs in their research, and each provides an essential avenue for scientific investigation. Descriptive research  is research designed to provide a snapshot of the current state of affairs . Correlational research  is research designed to discover relationships among variables and to allow the prediction of future events from present knowledge . Experimental research  is research in which initial equivalence among research participants in more than one group is created, followed by a manipulation of a given experience for these groups and a measurement of the influence of the manipulation . Each of the three research designs varies according to its strengths and limitations, and it is important to understand how each differs.

Descriptive Research: Assessing the Current State of Affairs

Descriptive research is designed to create a snapshot of the current thoughts, feelings, or behaviour of individuals. This section reviews three types of descriptive research : case studies , surveys , and naturalistic observation (Figure 3.4).

Sometimes the data in a descriptive research project are based on only a small set of individuals, often only one person or a single small group. These research designs are known as case studies — descriptive records of one or more individual’s experiences and behaviour . Sometimes case studies involve ordinary individuals, as when developmental psychologist Jean Piaget used his observation of his own children to develop his stage theory of cognitive development. More frequently, case studies are conducted on individuals who have unusual or abnormal experiences or characteristics or who find themselves in particularly difficult or stressful situations. The assumption is that by carefully studying individuals who are socially marginal, who are experiencing unusual situations, or who are going through a difficult phase in their lives, we can learn something about human nature.

Sigmund Freud was a master of using the psychological difficulties of individuals to draw conclusions about basic psychological processes. Freud wrote case studies of some of his most interesting patients and used these careful examinations to develop his important theories of personality. One classic example is Freud’s description of “Little Hans,” a child whose fear of horses the psychoanalyst interpreted in terms of repressed sexual impulses and the Oedipus complex (Freud, 1909/1964).

Another well-known case study is Phineas Gage, a man whose thoughts and emotions were extensively studied by cognitive psychologists after a railroad spike was blasted through his skull in an accident. Although there are questions about the interpretation of this case study (Kotowicz, 2007), it did provide early evidence that the brain’s frontal lobe is involved in emotion and morality (Damasio et al., 2005). An interesting example of a case study in clinical psychology is described by Rokeach (1964), who investigated in detail the beliefs of and interactions among three patients with schizophrenia, all of whom were convinced they were Jesus Christ.

In other cases the data from descriptive research projects come in the form of a survey — a measure administered through either an interview or a written questionnaire to get a picture of the beliefs or behaviours of a sample of people of interest . The people chosen to participate in the research (known as the sample) are selected to be representative of all the people that the researcher wishes to know about (the population). In election polls, for instance, a sample is taken from the population of all “likely voters” in the upcoming elections.

The results of surveys may sometimes be rather mundane, such as “Nine out of 10 doctors prefer Tymenocin” or “The median income in the city of Hamilton is $46,712.” Yet other times (particularly in discussions of social behaviour), the results can be shocking: “More than 40,000 people are killed by gunfire in the United States every year” or “More than 60% of women between the ages of 50 and 60 suffer from depression.” Descriptive research is frequently used by psychologists to get an estimate of the prevalence (or incidence ) of psychological disorders.

A final type of descriptive research — known as naturalistic observation — is research based on the observation of everyday events . For instance, a developmental psychologist who watches children on a playground and describes what they say to each other while they play is conducting descriptive research, as is a biopsychologist who observes animals in their natural habitats. One example of observational research involves a systematic procedure known as the strange situation , used to get a picture of how adults and young children interact. The data that are collected in the strange situation are systematically coded in a coding sheet such as that shown in Table 3.3.

The results of descriptive research projects are analyzed using descriptive statistics — numbers that summarize the distribution of scores on a measured variable . Most variables have distributions similar to that shown in Figure 3.5 where most of the scores are located near the centre of the distribution, and the distribution is symmetrical and bell-shaped. A data distribution that is shaped like a bell is known as a normal distribution .

A distribution can be described in terms of its central tendency — that is, the point in the distribution around which the data are centred — and its dispersion, or spread . The arithmetic average, or arithmetic mean , symbolized by the letter M , is the most commonly used measure of central tendency . It is computed by calculating the sum of all the scores of the variable and dividing this sum by the number of participants in the distribution (denoted by the letter N ). In the data presented in Figure 3.5 the mean height of the students is 67.12 inches (170.5 cm). The sample mean is usually indicated by the letter M .

In some cases, however, the data distribution is not symmetrical. This occurs when there are one or more extreme scores (known as outliers ) at one end of the distribution. Consider, for instance, the variable of family income (see Figure 3.6), which includes an outlier (a value of $3,800,000). In this case the mean is not a good measure of central tendency. Although it appears from Figure 3.6 that the central tendency of the family income variable should be around $70,000, the mean family income is actually $223,960. The single very extreme income has a disproportionate impact on the mean, resulting in a value that does not well represent the central tendency.

The median is used as an alternative measure of central tendency when distributions are not symmetrical. The median  is the score in the center of the distribution, meaning that 50% of the scores are greater than the median and 50% of the scores are less than the median . In our case, the median household income ($73,000) is a much better indication of central tendency than is the mean household income ($223,960).

A final measure of central tendency, known as the mode , represents the value that occurs most frequently in the distribution . You can see from Figure 3.6 that the mode for the family income variable is $93,000 (it occurs four times).

In addition to summarizing the central tendency of a distribution, descriptive statistics convey information about how the scores of the variable are spread around the central tendency. Dispersion refers to the extent to which the scores are all tightly clustered around the central tendency , as seen in Figure 3.7.

Or they may be more spread out away from it, as seen in Figure 3.8.

One simple measure of dispersion is to find the largest (the maximum ) and the smallest (the minimum ) observed values of the variable and to compute the range of the variable as the maximum observed score minus the minimum observed score. You can check that the range of the height variable in Figure 3.5 is 72 – 62 = 10. The standard deviation , symbolized as s , is the most commonly used measure of dispersion . Distributions with a larger standard deviation have more spread. The standard deviation of the height variable is s = 2.74, and the standard deviation of the family income variable is s = $745,337.

An advantage of descriptive research is that it attempts to capture the complexity of everyday behaviour. Case studies provide detailed information about a single person or a small group of people, surveys capture the thoughts or reported behaviours of a large population of people, and naturalistic observation objectively records the behaviour of people or animals as it occurs naturally. Thus descriptive research is used to provide a relatively complete understanding of what is currently happening.

Despite these advantages, descriptive research has a distinct disadvantage in that, although it allows us to get an idea of what is currently happening, it is usually limited to static pictures. Although descriptions of particular experiences may be interesting, they are not always transferable to other individuals in other situations, nor do they tell us exactly why specific behaviours or events occurred. For instance, descriptions of individuals who have suffered a stressful event, such as a war or an earthquake, can be used to understand the individuals’ reactions to the event but cannot tell us anything about the long-term effects of the stress. And because there is no comparison group that did not experience the stressful situation, we cannot know what these individuals would be like if they hadn’t had the stressful experience.

Correlational Research: Seeking Relationships among Variables

In contrast to descriptive research, which is designed primarily to provide static pictures, correlational research involves the measurement of two or more relevant variables and an assessment of the relationship between or among those variables. For instance, the variables of height and weight are systematically related (correlated) because taller people generally weigh more than shorter people. In the same way, study time and memory errors are also related, because the more time a person is given to study a list of words, the fewer errors he or she will make. When there are two variables in the research design, one of them is called the predictor variable and the other the outcome variable . The research design can be visualized as shown in Figure 3.9, where the curved arrow represents the expected correlation between these two variables.

One way of organizing the data from a correlational study with two variables is to graph the values of each of the measured variables using a scatter plot . As you can see in Figure 3.10 a scatter plot  is a visual image of the relationship between two variables . A point is plotted for each individual at the intersection of his or her scores for the two variables. When the association between the variables on the scatter plot can be easily approximated with a straight line , as in parts (a) and (b) of Figure 3.10 the variables are said to have a linear relationship .

When the straight line indicates that individuals who have above-average values for one variable also tend to have above-average values for the other variable , as in part (a), the relationship is said to be positive linear . Examples of positive linear relationships include those between height and weight, between education and income, and between age and mathematical abilities in children. In each case, people who score higher on one of the variables also tend to score higher on the other variable. Negative linear relationships , in contrast, as shown in part (b), occur when above-average values for one variable tend to be associated with below-average values for the other variable. Examples of negative linear relationships include those between the age of a child and the number of diapers the child uses, and between practice on and errors made on a learning task. In these cases, people who score higher on one of the variables tend to score lower on the other variable.

Relationships between variables that cannot be described with a straight line are known as nonlinear relationships . Part (c) of Figure 3.10 shows a common pattern in which the distribution of the points is essentially random. In this case there is no relationship at all between the two variables, and they are said to be independent . Parts (d) and (e) of Figure 3.10 show patterns of association in which, although there is an association, the points are not well described by a single straight line. For instance, part (d) shows the type of relationship that frequently occurs between anxiety and performance. Increases in anxiety from low to moderate levels are associated with performance increases, whereas increases in anxiety from moderate to high levels are associated with decreases in performance. Relationships that change in direction and thus are not described by a single straight line are called curvilinear relationships .

The most common statistical measure of the strength of linear relationships among variables is the Pearson correlation coefficient , which is symbolized by the letter r . The value of the correlation coefficient ranges from r = –1.00 to r = +1.00. The direction of the linear relationship is indicated by the sign of the correlation coefficient. Positive values of r (such as r = .54 or r = .67) indicate that the relationship is positive linear (i.e., the pattern of the dots on the scatter plot runs from the lower left to the upper right), whereas negative values of r (such as r = –.30 or r = –.72) indicate negative linear relationships (i.e., the dots run from the upper left to the lower right). The strength of the linear relationship is indexed by the distance of the correlation coefficient from zero (its absolute value). For instance, r = –.54 is a stronger relationship than r = .30, and r = .72 is a stronger relationship than r = –.57. Because the Pearson correlation coefficient only measures linear relationships, variables that have curvilinear relationships are not well described by r , and the observed correlation will be close to zero.

It is also possible to study relationships among more than two measures at the same time. A research design in which more than one predictor variable is used to predict a single outcome variable is analyzed through multiple regression (Aiken & West, 1991).  Multiple regression  is a statistical technique, based on correlation coefficients among variables, that allows predicting a single outcome variable from more than one predictor variable . For instance, Figure 3.11 shows a multiple regression analysis in which three predictor variables (Salary, job satisfaction, and years employed) are used to predict a single outcome (job performance). The use of multiple regression analysis shows an important advantage of correlational research designs — they can be used to make predictions about a person’s likely score on an outcome variable (e.g., job performance) based on knowledge of other variables.

An important limitation of correlational research designs is that they cannot be used to draw conclusions about the causal relationships among the measured variables. Consider, for instance, a researcher who has hypothesized that viewing violent behaviour will cause increased aggressive play in children. He has collected, from a sample of Grade 4 children, a measure of how many violent television shows each child views during the week, as well as a measure of how aggressively each child plays on the school playground. From his collected data, the researcher discovers a positive correlation between the two measured variables.

Although this positive correlation appears to support the researcher’s hypothesis, it cannot be taken to indicate that viewing violent television causes aggressive behaviour. Although the researcher is tempted to assume that viewing violent television causes aggressive play, there are other possibilities. One alternative possibility is that the causal direction is exactly opposite from what has been hypothesized. Perhaps children who have behaved aggressively at school develop residual excitement that leads them to want to watch violent television shows at home (Figure 3.13):

Although this possibility may seem less likely, there is no way to rule out the possibility of such reverse causation on the basis of this observed correlation. It is also possible that both causal directions are operating and that the two variables cause each other (Figure 3.14).

Still another possible explanation for the observed correlation is that it has been produced by the presence of a common-causal variable (also known as a third variable ). A common-causal variable  is a variable that is not part of the research hypothesis but that causes both the predictor and the outcome variable and thus produces the observed correlation between them . In our example, a potential common-causal variable is the discipline style of the children’s parents. Parents who use a harsh and punitive discipline style may produce children who like to watch violent television and who also behave aggressively in comparison to children whose parents use less harsh discipline (Figure 3.15)

In this case, television viewing and aggressive play would be positively correlated (as indicated by the curved arrow between them), even though neither one caused the other but they were both caused by the discipline style of the parents (the straight arrows). When the predictor and outcome variables are both caused by a common-causal variable, the observed relationship between them is said to be spurious . A spurious relationship  is a relationship between two variables in which a common-causal variable produces and “explains away” the relationship . If effects of the common-causal variable were taken away, or controlled for, the relationship between the predictor and outcome variables would disappear. In the example, the relationship between aggression and television viewing might be spurious because by controlling for the effect of the parents’ disciplining style, the relationship between television viewing and aggressive behaviour might go away.

Common-causal variables in correlational research designs can be thought of as mystery variables because, as they have not been measured, their presence and identity are usually unknown to the researcher. Since it is not possible to measure every variable that could cause both the predictor and outcome variables, the existence of an unknown common-causal variable is always a possibility. For this reason, we are left with the basic limitation of correlational research: correlation does not demonstrate causation. It is important that when you read about correlational research projects, you keep in mind the possibility of spurious relationships, and be sure to interpret the findings appropriately. Although correlational research is sometimes reported as demonstrating causality without any mention being made of the possibility of reverse causation or common-causal variables, informed consumers of research, like you, are aware of these interpretational problems.

In sum, correlational research designs have both strengths and limitations. One strength is that they can be used when experimental research is not possible because the predictor variables cannot be manipulated. Correlational designs also have the advantage of allowing the researcher to study behaviour as it occurs in everyday life. And we can also use correlational designs to make predictions — for instance, to predict from the scores on their battery of tests the success of job trainees during a training session. But we cannot use such correlational information to determine whether the training caused better job performance. For that, researchers rely on experiments.

Experimental Research: Understanding the Causes of Behaviour

The goal of experimental research design is to provide more definitive conclusions about the causal relationships among the variables in the research hypothesis than is available from correlational designs. In an experimental research design, the variables of interest are called the independent variable (or variables ) and the dependent variable . The independent variable  in an experiment is the causing variable that is created (manipulated) by the experimenter . The dependent variable  in an experiment is a measured variable that is expected to be influenced by the experimental manipulation . The research hypothesis suggests that the manipulated independent variable or variables will cause changes in the measured dependent variables. We can diagram the research hypothesis by using an arrow that points in one direction. This demonstrates the expected direction of causality (Figure 3.16):

Research Focus: Video Games and Aggression

Consider an experiment conducted by Anderson and Dill (2000). The study was designed to test the hypothesis that viewing violent video games would increase aggressive behaviour. In this research, male and female undergraduates from Iowa State University were given a chance to play with either a violent video game (Wolfenstein 3D) or a nonviolent video game (Myst). During the experimental session, the participants played their assigned video games for 15 minutes. Then, after the play, each participant played a competitive game with an opponent in which the participant could deliver blasts of white noise through the earphones of the opponent. The operational definition of the dependent variable (aggressive behaviour) was the level and duration of noise delivered to the opponent. The design of the experiment is shown in Figure 3.17

Two advantages of the experimental research design are (a) the assurance that the independent variable (also known as the experimental manipulation ) occurs prior to the measured dependent variable, and (b) the creation of initial equivalence between the conditions of the experiment (in this case by using random assignment to conditions).

Experimental designs have two very nice features. For one, they guarantee that the independent variable occurs prior to the measurement of the dependent variable. This eliminates the possibility of reverse causation. Second, the influence of common-causal variables is controlled, and thus eliminated, by creating initial equivalence among the participants in each of the experimental conditions before the manipulation occurs.

The most common method of creating equivalence among the experimental conditions is through random assignment to conditions, a procedure in which the condition that each participant is assigned to is determined through a random process, such as drawing numbers out of an envelope or using a random number table . Anderson and Dill first randomly assigned about 100 participants to each of their two groups (Group A and Group B). Because they used random assignment to conditions, they could be confident that, before the experimental manipulation occurred, the students in Group A were, on average, equivalent to the students in Group B on every possible variable, including variables that are likely to be related to aggression, such as parental discipline style, peer relationships, hormone levels, diet — and in fact everything else.

Then, after they had created initial equivalence, Anderson and Dill created the experimental manipulation — they had the participants in Group A play the violent game and the participants in Group B play the nonviolent game. Then they compared the dependent variable (the white noise blasts) between the two groups, finding that the students who had viewed the violent video game gave significantly longer noise blasts than did the students who had played the nonviolent game.

Anderson and Dill had from the outset created initial equivalence between the groups. This initial equivalence allowed them to observe differences in the white noise levels between the two groups after the experimental manipulation, leading to the conclusion that it was the independent variable (and not some other variable) that caused these differences. The idea is that the only thing that was different between the students in the two groups was the video game they had played.

Despite the advantage of determining causation, experiments do have limitations. One is that they are often conducted in laboratory situations rather than in the everyday lives of people. Therefore, we do not know whether results that we find in a laboratory setting will necessarily hold up in everyday life. Second, and more important, is that some of the most interesting and key social variables cannot be experimentally manipulated. If we want to study the influence of the size of a mob on the destructiveness of its behaviour, or to compare the personality characteristics of people who join suicide cults with those of people who do not join such cults, these relationships must be assessed using correlational designs, because it is simply not possible to experimentally manipulate these variables.

Key Takeaways

• Descriptive, correlational, and experimental research designs are used to collect and analyze data.
• Descriptive designs include case studies, surveys, and naturalistic observation. The goal of these designs is to get a picture of the current thoughts, feelings, or behaviours in a given group of people. Descriptive research is summarized using descriptive statistics.
• Correlational research designs measure two or more relevant variables and assess a relationship between or among them. The variables may be presented on a scatter plot to visually show the relationships. The Pearson Correlation Coefficient ( r ) is a measure of the strength of linear relationship between two variables.
• Common-causal variables may cause both the predictor and outcome variable in a correlational design, producing a spurious relationship. The possibility of common-causal variables makes it impossible to draw causal conclusions from correlational research designs.
• Experimental research involves the manipulation of an independent variable and the measurement of a dependent variable. Random assignment to conditions is normally used to create initial equivalence between the groups, allowing researchers to draw causal conclusions.

Exercises and Critical Thinking

• There is a negative correlation between the row that a student sits in in a large class (when the rows are numbered from front to back) and his or her final grade in the class. Do you think this represents a causal relationship or a spurious relationship, and why?
• Think of two variables (other than those mentioned in this book) that are likely to be correlated, but in which the correlation is probably spurious. What is the likely common-causal variable that is producing the relationship?
• Imagine a researcher wants to test the hypothesis that participating in psychotherapy will cause a decrease in reported anxiety. Describe the type of research design the investigator might use to draw this conclusion. What would be the independent and dependent variables in the research?

Image Attributions

Figure 3.4: “ Reading newspaper ” by Alaskan Dude (http://commons.wikimedia.org/wiki/File:Reading_newspaper.jpg) is licensed under CC BY 2.0

Aiken, L., & West, S. (1991).  Multiple regression: Testing and interpreting interactions . Newbury Park, CA: Sage.

Ainsworth, M. S., Blehar, M. C., Waters, E., & Wall, S. (1978).  Patterns of attachment: A psychological study of the strange situation . Hillsdale, NJ: Lawrence Erlbaum Associates.

Anderson, C. A., & Dill, K. E. (2000). Video games and aggressive thoughts, feelings, and behavior in the laboratory and in life.  Journal of Personality and Social Psychology, 78 (4), 772–790.

Damasio, H., Grabowski, T., Frank, R., Galaburda, A. M., Damasio, A. R., Cacioppo, J. T., & Berntson, G. G. (2005). The return of Phineas Gage: Clues about the brain from the skull of a famous patient. In  Social neuroscience: Key readings.  (pp. 21–28). New York, NY: Psychology Press.

Freud, S. (1909/1964). Analysis of phobia in a five-year-old boy. In E. A. Southwell & M. Merbaum (Eds.),  Personality: Readings in theory and research  (pp. 3–32). Belmont, CA: Wadsworth. (Original work published 1909).

Kotowicz, Z. (2007). The strange case of Phineas Gage.  History of the Human Sciences, 20 (1), 115–131.

Rokeach, M. (1964).  The three Christs of Ypsilanti: A psychological study . New York, NY: Knopf.

Stangor, C. (2011). Research methods for the behavioural sciences (4th ed.). Mountain View, CA: Cengage.

Long Descriptions

Figure 3.6 long description: There are 25 families. 24 families have an income between $44,000 and $111,000 and one family has an income of $3,800,000. The mean income is $223,960 while the median income is $73,000. [Return to Figure 3.6]

Figure 3.10 long description: Types of scatter plots.

• Positive linear, r=positive .82. The plots on the graph form a rough line that runs from lower left to upper right.
• Negative linear, r=negative .70. The plots on the graph form a rough line that runs from upper left to lower right.
• Independent, r=0.00. The plots on the graph are spread out around the centre.
• Curvilinear, r=0.00. The plots of the graph form a rough line that goes up and then down like a hill.
• Curvilinear, r=0.00. The plots on the graph for a rough line that goes down and then up like a ditch.

[Return to Figure 3.10]

Introduction to Psychology - 1st Canadian Edition Copyright © 2014 by Jennifer Walinga and Charles Stangor is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License , except where otherwise noted.

Share This Book

• Privacy Policy

Home » Correlational Research Vs Experimental Research

Correlational Research Vs Experimental Research

Table of Contents

Correlational research and experimental research are two different research approaches used in social sciences and other fields of research.

Correlational Research

Correlational Research is a research approach that examines the relationship between two or more variables. It involves measuring the degree of association or correlation between the variables without manipulating them. The goal of correlational research is to identify whether there is a relationship between the variables and the strength of that relationship. Correlational research is typically conducted through surveys, observational studies, or secondary data analysis.

Experimental Research

Experimental Research , on the other hand, is a research approach that involves the manipulation of one or more variables to observe the effect on another variable. The goal of experimental research is to establish a cause-and-effect relationship between the variables. Experimental research is typically conducted in a controlled environment and involves random assignment of participants to different groups to ensure that the groups are equivalent. The data is collected through measurements and observations, and statistical analysis is used to test the hypotheses.

Difference Between Correlational Research and Experimental Research

Here’s a comparison table that highlights the differences between correlational research and experimental research:

Also see Research Methods

About the author

Muhammad Hassan

Researcher, Academic Writer, Web developer

You may also like

Research Hypothesis Vs Null Hypothesis

Exploratory Vs Explanatory Research

Essay Vs Research Paper

Descriptive vs Experimental Research

Inductive Vs Deductive Research

Longitudinal Vs Cross-Sectional Research

Frequently asked questions

What’s the difference between correlational and experimental research.

Controlled experiments establish causality, whereas correlational studies only show associations between variables.

• In an experimental design , you manipulate an independent variable and measure its effect on a dependent variable. Other variables are controlled so they can’t impact the results.
• In a correlational design , you measure variables without manipulating any of them. You can test whether your variables change together, but you can’t be sure that one variable caused a change in another.

In general, correlational research is high in external validity while experimental research is high in internal validity .

Frequently asked questions: Methodology

Quantitative observations involve measuring or counting something and expressing the result in numerical form, while qualitative observations involve describing something in non-numerical terms, such as its appearance, texture, or color.

To make quantitative observations , you need to use instruments that are capable of measuring the quantity you want to observe. For example, you might use a ruler to measure the length of an object or a thermometer to measure its temperature.

Scope of research is determined at the beginning of your research process , prior to the data collection stage. Sometimes called “scope of study,” your scope delineates what will and will not be covered in your project. It helps you focus your work and your time, ensuring that you’ll be able to achieve your goals and outcomes.

Defining a scope can be very useful in any research project, from a research proposal to a thesis or dissertation . A scope is needed for all types of research: quantitative , qualitative , and mixed methods .

To define your scope of research, consider the following:

• Budget constraints or any specifics of grant funding
• Your proposed timeline and duration
• Specifics about your population of study, your proposed sample size , and the research methodology you’ll pursue
• Any inclusion and exclusion criteria
• Any anticipated control , extraneous , or confounding variables that could bias your research if not accounted for properly.

Inclusion and exclusion criteria are predominantly used in non-probability sampling . In purposive sampling and snowball sampling , restrictions apply as to who can be included in the sample .

Inclusion and exclusion criteria are typically presented and discussed in the methodology section of your thesis or dissertation .

The purpose of theory-testing mode is to find evidence in order to disprove, refine, or support a theory. As such, generalisability is not the aim of theory-testing mode.

Due to this, the priority of researchers in theory-testing mode is to eliminate alternative causes for relationships between variables . In other words, they prioritise internal validity over external validity , including ecological validity .

Convergent validity shows how much a measure of one construct aligns with other measures of the same or related constructs .

On the other hand, concurrent validity is about how a measure matches up to some known criterion or gold standard, which can be another measure.

Although both types of validity are established by calculating the association or correlation between a test score and another variable , they represent distinct validation methods.

Validity tells you how accurately a method measures what it was designed to measure. There are 4 main types of validity :

• Construct validity : Does the test measure the construct it was designed to measure?
• Face validity : Does the test appear to be suitable for its objectives ?
• Content validity : Does the test cover all relevant parts of the construct it aims to measure.
• Criterion validity : Do the results accurately measure the concrete outcome they are designed to measure?

Criterion validity evaluates how well a test measures the outcome it was designed to measure. An outcome can be, for example, the onset of a disease.

Criterion validity consists of two subtypes depending on the time at which the two measures (the criterion and your test) are obtained:

• Concurrent validity is a validation strategy where the the scores of a test and the criterion are obtained at the same time
• Predictive validity is a validation strategy where the criterion variables are measured after the scores of the test

Attrition refers to participants leaving a study. It always happens to some extent – for example, in randomised control trials for medical research.

Differential attrition occurs when attrition or dropout rates differ systematically between the intervention and the control group . As a result, the characteristics of the participants who drop out differ from the characteristics of those who stay in the study. Because of this, study results may be biased .

Criterion validity and construct validity are both types of measurement validity . In other words, they both show you how accurately a method measures something.

While construct validity is the degree to which a test or other measurement method measures what it claims to measure, criterion validity is the degree to which a test can predictively (in the future) or concurrently (in the present) measure something.

Construct validity is often considered the overarching type of measurement validity . You need to have face validity , content validity , and criterion validity in order to achieve construct validity.

Convergent validity and discriminant validity are both subtypes of construct validity . Together, they help you evaluate whether a test measures the concept it was designed to measure.

• Convergent validity indicates whether a test that is designed to measure a particular construct correlates with other tests that assess the same or similar construct.
• Discriminant validity indicates whether two tests that should not be highly related to each other are indeed not related. This type of validity is also called divergent validity .

You need to assess both in order to demonstrate construct validity. Neither one alone is sufficient for establishing construct validity.

Face validity and content validity are similar in that they both evaluate how suitable the content of a test is. The difference is that face validity is subjective, and assesses content at surface level.

When a test has strong face validity, anyone would agree that the test’s questions appear to measure what they are intended to measure.

For example, looking at a 4th grade math test consisting of problems in which students have to add and multiply, most people would agree that it has strong face validity (i.e., it looks like a math test).

On the other hand, content validity evaluates how well a test represents all the aspects of a topic. Assessing content validity is more systematic and relies on expert evaluation. of each question, analysing whether each one covers the aspects that the test was designed to cover.

A 4th grade math test would have high content validity if it covered all the skills taught in that grade. Experts(in this case, math teachers), would have to evaluate the content validity by comparing the test to the learning objectives.

Content validity shows you how accurately a test or other measurement method taps  into the various aspects of the specific construct you are researching.

In other words, it helps you answer the question: “does the test measure all aspects of the construct I want to measure?” If it does, then the test has high content validity.

The higher the content validity, the more accurate the measurement of the construct.

If the test fails to include parts of the construct, or irrelevant parts are included, the validity of the instrument is threatened, which brings your results into question.

Construct validity refers to how well a test measures the concept (or construct) it was designed to measure. Assessing construct validity is especially important when you’re researching concepts that can’t be quantified and/or are intangible, like introversion. To ensure construct validity your test should be based on known indicators of introversion ( operationalisation ).

On the other hand, content validity assesses how well the test represents all aspects of the construct. If some aspects are missing or irrelevant parts are included, the test has low content validity.

• Discriminant validity indicates whether two tests that should not be highly related to each other are indeed not related

Construct validity has convergent and discriminant subtypes. They assist determine if a test measures the intended notion.

The reproducibility and replicability of a study can be ensured by writing a transparent, detailed method section and using clear, unambiguous language.

Reproducibility and replicability are related terms.

• A successful reproduction shows that the data analyses were conducted in a fair and honest manner.
• A successful replication shows that the reliability of the results is high.
• Reproducing research entails reanalysing the existing data in the same manner.
• Replicating (or repeating ) the research entails reconducting the entire analysis, including the collection of new data . 

Snowball sampling is a non-probability sampling method . Unlike probability sampling (which involves some form of random selection ), the initial individuals selected to be studied are the ones who recruit new participants.

Because not every member of the target population has an equal chance of being recruited into the sample, selection in snowball sampling is non-random.

Snowball sampling is a non-probability sampling method , where there is not an equal chance for every member of the population to be included in the sample .

This means that you cannot use inferential statistics and make generalisations – often the goal of quantitative research . As such, a snowball sample is not representative of the target population, and is usually a better fit for qualitative research .

Snowball sampling relies on the use of referrals. Here, the researcher recruits one or more initial participants, who then recruit the next ones. 

Participants share similar characteristics and/or know each other. Because of this, not every member of the population has an equal chance of being included in the sample, giving rise to sampling bias .

Snowball sampling is best used in the following cases:

• If there is no sampling frame available (e.g., people with a rare disease)
• If the population of interest is hard to access or locate (e.g., people experiencing homelessness)
• If the research focuses on a sensitive topic (e.g., extra-marital affairs)

Stratified sampling and quota sampling both involve dividing the population into subgroups and selecting units from each subgroup. The purpose in both cases is to select a representative sample and/or to allow comparisons between subgroups.

The main difference is that in stratified sampling, you draw a random sample from each subgroup ( probability sampling ). In quota sampling you select a predetermined number or proportion of units, in a non-random manner ( non-probability sampling ).

Random sampling or probability sampling is based on random selection. This means that each unit has an equal chance (i.e., equal probability) of being included in the sample.

On the other hand, convenience sampling involves stopping people at random, which means that not everyone has an equal chance of being selected depending on the place, time, or day you are collecting your data.

Convenience sampling and quota sampling are both non-probability sampling methods. They both use non-random criteria like availability, geographical proximity, or expert knowledge to recruit study participants.

However, in convenience sampling, you continue to sample units or cases until you reach the required sample size.

In quota sampling, you first need to divide your population of interest into subgroups (strata) and estimate their proportions (quota) in the population. Then you can start your data collection , using convenience sampling to recruit participants, until the proportions in each subgroup coincide with the estimated proportions in the population.

A sampling frame is a list of every member in the entire population . It is important that the sampling frame is as complete as possible, so that your sample accurately reflects your population.

Stratified and cluster sampling may look similar, but bear in mind that groups created in cluster sampling are heterogeneous , so the individual characteristics in the cluster vary. In contrast, groups created in stratified sampling are homogeneous , as units share characteristics.

Relatedly, in cluster sampling you randomly select entire groups and include all units of each group in your sample. However, in stratified sampling, you select some units of all groups and include them in your sample. In this way, both methods can ensure that your sample is representative of the target population .

When your population is large in size, geographically dispersed, or difficult to contact, it’s necessary to use a sampling method .

This allows you to gather information from a smaller part of the population, i.e. the sample, and make accurate statements by using statistical analysis. A few sampling methods include simple random sampling , convenience sampling , and snowball sampling .

The two main types of social desirability bias are:

• Self-deceptive enhancement (self-deception): The tendency to see oneself in a favorable light without realizing it.
• Impression managemen t (other-deception): The tendency to inflate one’s abilities or achievement in order to make a good impression on other people.

Response bias refers to conditions or factors that take place during the process of responding to surveys, affecting the responses. One type of response bias is social desirability bias .

Demand characteristics are aspects of experiments that may give away the research objective to participants. Social desirability bias occurs when participants automatically try to respond in ways that make them seem likeable in a study, even if it means misrepresenting how they truly feel.

Participants may use demand characteristics to infer social norms or experimenter expectancies and act in socially desirable ways, so you should try to control for demand characteristics wherever possible.

A systematic review is secondary research because it uses existing research. You don’t collect new data yourself.

Ethical considerations in research are a set of principles that guide your research designs and practices. These principles include voluntary participation, informed consent, anonymity, confidentiality, potential for harm, and results communication.

Scientists and researchers must always adhere to a certain code of conduct when collecting data from others .

These considerations protect the rights of research participants, enhance research validity , and maintain scientific integrity.

Research ethics matter for scientific integrity, human rights and dignity, and collaboration between science and society. These principles make sure that participation in studies is voluntary, informed, and safe.

Research misconduct means making up or falsifying data, manipulating data analyses, or misrepresenting results in research reports. It’s a form of academic fraud.

These actions are committed intentionally and can have serious consequences; research misconduct is not a simple mistake or a point of disagreement but a serious ethical failure.

Anonymity means you don’t know who the participants are, while confidentiality means you know who they are but remove identifying information from your research report. Both are important ethical considerations .

You can only guarantee anonymity by not collecting any personally identifying information – for example, names, phone numbers, email addresses, IP addresses, physical characteristics, photos, or videos.

You can keep data confidential by using aggregate information in your research report, so that you only refer to groups of participants rather than individuals.

Peer review is a process of evaluating submissions to an academic journal. Utilising rigorous criteria, a panel of reviewers in the same subject area decide whether to accept each submission for publication.

For this reason, academic journals are often considered among the most credible sources you can use in a research project – provided that the journal itself is trustworthy and well regarded.

In general, the peer review process follows the following steps:

• First, the author submits the manuscript to the editor.
• Reject the manuscript and send it back to author, or
• Send it onward to the selected peer reviewer(s)
• Next, the peer review process occurs. The reviewer provides feedback, addressing any major or minor issues with the manuscript, and gives their advice regarding what edits should be made.
• Lastly, the edited manuscript is sent back to the author. They input the edits, and resubmit it to the editor for publication.

Peer review can stop obviously problematic, falsified, or otherwise untrustworthy research from being published. It also represents an excellent opportunity to get feedback from renowned experts in your field.

It acts as a first defence, helping you ensure your argument is clear and that there are no gaps, vague terms, or unanswered questions for readers who weren’t involved in the research process.

Peer-reviewed articles are considered a highly credible source due to this stringent process they go through before publication.

Many academic fields use peer review , largely to determine whether a manuscript is suitable for publication. Peer review enhances the credibility of the published manuscript.

However, peer review is also common in non-academic settings. The United Nations, the European Union, and many individual nations use peer review to evaluate grant applications. It is also widely used in medical and health-related fields as a teaching or quality-of-care measure.

Peer assessment is often used in the classroom as a pedagogical tool. Both receiving feedback and providing it are thought to enhance the learning process, helping students think critically and collaboratively.

• In a single-blind study , only the participants are blinded.
• In a double-blind study , both participants and experimenters are blinded.
• In a triple-blind study , the assignment is hidden not only from participants and experimenters, but also from the researchers analysing the data.

Blinding is important to reduce bias (e.g., observer bias , demand characteristics ) and ensure a study’s internal validity .

If participants know whether they are in a control or treatment group , they may adjust their behaviour in ways that affect the outcome that researchers are trying to measure. If the people administering the treatment are aware of group assignment, they may treat participants differently and thus directly or indirectly influence the final results.

Blinding means hiding who is assigned to the treatment group and who is assigned to the control group in an experiment .

Explanatory research is a research method used to investigate how or why something occurs when only a small amount of information is available pertaining to that topic. It can help you increase your understanding of a given topic.

Explanatory research is used to investigate how or why a phenomenon occurs. Therefore, this type of research is often one of the first stages in the research process , serving as a jumping-off point for future research.

Exploratory research is a methodology approach that explores research questions that have not previously been studied in depth. It is often used when the issue you’re studying is new, or the data collection process is challenging in some way.

Exploratory research is often used when the issue you’re studying is new or when the data collection process is challenging for some reason.

You can use exploratory research if you have a general idea or a specific question that you want to study but there is no preexisting knowledge or paradigm with which to study it.

To implement random assignment , assign a unique number to every member of your study’s sample .

Then, you can use a random number generator or a lottery method to randomly assign each number to a control or experimental group. You can also do so manually, by flipping a coin or rolling a die to randomly assign participants to groups.

Random selection, or random sampling , is a way of selecting members of a population for your study’s sample.

In contrast, random assignment is a way of sorting the sample into control and experimental groups.

Random sampling enhances the external validity or generalisability of your results, while random assignment improves the internal validity of your study.

Random assignment is used in experiments with a between-groups or independent measures design. In this research design, there’s usually a control group and one or more experimental groups. Random assignment helps ensure that the groups are comparable.

In general, you should always use random assignment in this type of experimental design when it is ethically possible and makes sense for your study topic.

Clean data are valid, accurate, complete, consistent, unique, and uniform. Dirty data include inconsistencies and errors.

Dirty data can come from any part of the research process, including poor research design , inappropriate measurement materials, or flawed data entry.

Data cleaning takes place between data collection and data analyses. But you can use some methods even before collecting data.

For clean data, you should start by designing measures that collect valid data. Data validation at the time of data entry or collection helps you minimize the amount of data cleaning you’ll need to do.

After data collection, you can use data standardisation and data transformation to clean your data. You’ll also deal with any missing values, outliers, and duplicate values.

Data cleaning involves spotting and resolving potential data inconsistencies or errors to improve your data quality. An error is any value (e.g., recorded weight) that doesn’t reflect the true value (e.g., actual weight) of something that’s being measured.

In this process, you review, analyse, detect, modify, or remove ‘dirty’ data to make your dataset ‘clean’. Data cleaning is also called data cleansing or data scrubbing.

Data cleaning is necessary for valid and appropriate analyses. Dirty data contain inconsistencies or errors , but cleaning your data helps you minimise or resolve these.

Without data cleaning, you could end up with a Type I or II error in your conclusion. These types of erroneous conclusions can be practically significant with important consequences, because they lead to misplaced investments or missed opportunities.

Observer bias occurs when a researcher’s expectations, opinions, or prejudices influence what they perceive or record in a study. It usually affects studies when observers are aware of the research aims or hypotheses. This type of research bias is also called detection bias or ascertainment bias .

The observer-expectancy effect occurs when researchers influence the results of their own study through interactions with participants.

Researchers’ own beliefs and expectations about the study results may unintentionally influence participants through demand characteristics .

You can use several tactics to minimise observer bias .

• Use masking (blinding) to hide the purpose of your study from all observers.
• Triangulate your data with different data collection methods or sources.
• Use multiple observers and ensure inter-rater reliability.
• Train your observers to make sure data is consistently recorded between them.
• Standardise your observation procedures to make sure they are structured and clear.

Naturalistic observation is a valuable tool because of its flexibility, external validity , and suitability for topics that can’t be studied in a lab setting.

The downsides of naturalistic observation include its lack of scientific control , ethical considerations , and potential for bias from observers and subjects.

Naturalistic observation is a qualitative research method where you record the behaviours of your research subjects in real-world settings. You avoid interfering or influencing anything in a naturalistic observation.

You can think of naturalistic observation as ‘people watching’ with a purpose.

Closed-ended, or restricted-choice, questions offer respondents a fixed set of choices to select from. These questions are easier to answer quickly.

Open-ended or long-form questions allow respondents to answer in their own words. Because there are no restrictions on their choices, respondents can answer in ways that researchers may not have otherwise considered.

You can organise the questions logically, with a clear progression from simple to complex, or randomly between respondents. A logical flow helps respondents process the questionnaire easier and quicker, but it may lead to bias. Randomisation can minimise the bias from order effects.

Questionnaires can be self-administered or researcher-administered.

Self-administered questionnaires can be delivered online or in paper-and-pen formats, in person or by post. All questions are standardised so that all respondents receive the same questions with identical wording.

Researcher-administered questionnaires are interviews that take place by phone, in person, or online between researchers and respondents. You can gain deeper insights by clarifying questions for respondents or asking follow-up questions.

In a controlled experiment , all extraneous variables are held constant so that they can’t influence the results. Controlled experiments require:

• A control group that receives a standard treatment, a fake treatment, or no treatment
• Random assignment of participants to ensure the groups are equivalent

Depending on your study topic, there are various other methods of controlling variables .

An experimental group, also known as a treatment group, receives the treatment whose effect researchers wish to study, whereas a control group does not. They should be identical in all other ways.

A true experiment (aka a controlled experiment) always includes at least one control group that doesn’t receive the experimental treatment.

However, some experiments use a within-subjects design to test treatments without a control group. In these designs, you usually compare one group’s outcomes before and after a treatment (instead of comparing outcomes between different groups).

For strong internal validity , it’s usually best to include a control group if possible. Without a control group, it’s harder to be certain that the outcome was caused by the experimental treatment and not by other variables.

A questionnaire is a data collection tool or instrument, while a survey is an overarching research method that involves collecting and analysing data from people using questionnaires.

A Likert scale is a rating scale that quantitatively assesses opinions, attitudes, or behaviours. It is made up of four or more questions that measure a single attitude or trait when response scores are combined.

To use a Likert scale in a survey , you present participants with Likert-type questions or statements, and a continuum of items, usually with five or seven possible responses, to capture their degree of agreement.

Individual Likert-type questions are generally considered ordinal data , because the items have clear rank order, but don’t have an even distribution.

Overall Likert scale scores are sometimes treated as interval data. These scores are considered to have directionality and even spacing between them.

The type of data determines what statistical tests you should use to analyse your data.

A research hypothesis is your proposed answer to your research question. The research hypothesis usually includes an explanation (‘ x affects y because …’).

A statistical hypothesis, on the other hand, is a mathematical statement about a population parameter. Statistical hypotheses always come in pairs: the null and alternative hypotheses. In a well-designed study , the statistical hypotheses correspond logically to the research hypothesis.

A hypothesis states your predictions about what your research will find. It is a tentative answer to your research question that has not yet been tested. For some research projects, you might have to write several hypotheses that address different aspects of your research question.

A hypothesis is not just a guess. It should be based on existing theories and knowledge. It also has to be testable, which means you can support or refute it through scientific research methods (such as experiments, observations, and statistical analysis of data).

Cross-sectional studies are less expensive and time-consuming than many other types of study. They can provide useful insights into a population’s characteristics and identify correlations for further research.

Sometimes only cross-sectional data are available for analysis; other times your research question may only require a cross-sectional study to answer it.

Cross-sectional studies cannot establish a cause-and-effect relationship or analyse behaviour over a period of time. To investigate cause and effect, you need to do a longitudinal study or an experimental study .

Longitudinal studies and cross-sectional studies are two different types of research design . In a cross-sectional study you collect data from a population at a specific point in time; in a longitudinal study you repeatedly collect data from the same sample over an extended period of time.

Longitudinal studies are better to establish the correct sequence of events, identify changes over time, and provide insight into cause-and-effect relationships, but they also tend to be more expensive and time-consuming than other types of studies.

The 1970 British Cohort Study , which has collected data on the lives of 17,000 Brits since their births in 1970, is one well-known example of a longitudinal study .

Longitudinal studies can last anywhere from weeks to decades, although they tend to be at least a year long.

A correlation reflects the strength and/or direction of the association between two or more variables.

• A positive correlation means that both variables change in the same direction.
• A negative correlation means that the variables change in opposite directions.
• A zero correlation means there’s no relationship between the variables.

A correlational research design investigates relationships between two variables (or more) without the researcher controlling or manipulating any of them. It’s a non-experimental type of quantitative research .

A correlation coefficient is a single number that describes the strength and direction of the relationship between your variables.

Different types of correlation coefficients might be appropriate for your data based on their levels of measurement and distributions . The Pearson product-moment correlation coefficient (Pearson’s r ) is commonly used to assess a linear relationship between two quantitative variables.

The third variable and directionality problems are two main reasons why correlation isn’t causation .

The third variable problem means that a confounding variable affects both variables to make them seem causally related when they are not.

The directionality problem is when two variables correlate and might actually have a causal relationship, but it’s impossible to conclude which variable causes changes in the other.

As a rule of thumb, questions related to thoughts, beliefs, and feelings work well in focus groups . Take your time formulating strong questions, paying special attention to phrasing. Be careful to avoid leading questions , which can bias your responses.

Overall, your focus group questions should be:

• Open-ended and flexible
• Impossible to answer with ‘yes’ or ‘no’ (questions that start with ‘why’ or ‘how’ are often best)
• Unambiguous, getting straight to the point while still stimulating discussion
• Unbiased and neutral

Social desirability bias is the tendency for interview participants to give responses that will be viewed favourably by the interviewer or other participants. It occurs in all types of interviews and surveys , but is most common in semi-structured interviews , unstructured interviews , and focus groups .

Social desirability bias can be mitigated by ensuring participants feel at ease and comfortable sharing their views. Make sure to pay attention to your own body language and any physical or verbal cues, such as nodding or widening your eyes.

This type of bias in research can also occur in observations if the participants know they’re being observed. They might alter their behaviour accordingly.

A focus group is a research method that brings together a small group of people to answer questions in a moderated setting. The group is chosen due to predefined demographic traits, and the questions are designed to shed light on a topic of interest. It is one of four types of interviews .

The four most common types of interviews are:

• Structured interviews : The questions are predetermined in both topic and order.
• Semi-structured interviews : A few questions are predetermined, but other questions aren’t planned.
• Unstructured interviews : None of the questions are predetermined.
• Focus group interviews : The questions are presented to a group instead of one individual.

An unstructured interview is the most flexible type of interview, but it is not always the best fit for your research topic.

Unstructured interviews are best used when:

• You are an experienced interviewer and have a very strong background in your research topic, since it is challenging to ask spontaneous, colloquial questions
• Your research question is exploratory in nature. While you may have developed hypotheses, you are open to discovering new or shifting viewpoints through the interview process.
• You are seeking descriptive data, and are ready to ask questions that will deepen and contextualise your initial thoughts and hypotheses
• Your research depends on forming connections with your participants and making them feel comfortable revealing deeper emotions, lived experiences, or thoughts

A semi-structured interview is a blend of structured and unstructured types of interviews. Semi-structured interviews are best used when:

• You have prior interview experience. Spontaneous questions are deceptively challenging, and it’s easy to accidentally ask a leading question or make a participant uncomfortable.
• Your research question is exploratory in nature. Participant answers can guide future research questions and help you develop a more robust knowledge base for future research.

The interviewer effect is a type of bias that emerges when a characteristic of an interviewer (race, age, gender identity, etc.) influences the responses given by the interviewee.

There is a risk of an interviewer effect in all types of interviews , but it can be mitigated by writing really high-quality interview questions.

A structured interview is a data collection method that relies on asking questions in a set order to collect data on a topic. They are often quantitative in nature. Structured interviews are best used when:

• You already have a very clear understanding of your topic. Perhaps significant research has already been conducted, or you have done some prior research yourself, but you already possess a baseline for designing strong structured questions.
• You are constrained in terms of time or resources and need to analyse your data quickly and efficiently
• Your research question depends on strong parity between participants, with environmental conditions held constant

More flexible interview options include semi-structured interviews , unstructured interviews , and focus groups .

When conducting research, collecting original data has significant advantages:

• You can tailor data collection to your specific research aims (e.g., understanding the needs of your consumers or user testing your website).
• You can control and standardise the process for high reliability and validity (e.g., choosing appropriate measurements and sampling methods ).

However, there are also some drawbacks: data collection can be time-consuming, labour-intensive, and expensive. In some cases, it’s more efficient to use secondary data that has already been collected by someone else, but the data might be less reliable.

Data collection is the systematic process by which observations or measurements are gathered in research. It is used in many different contexts by academics, governments, businesses, and other organisations.

A mediator variable explains the process through which two variables are related, while a moderator variable affects the strength and direction of that relationship.

A confounder is a third variable that affects variables of interest and makes them seem related when they are not. In contrast, a mediator is the mechanism of a relationship between two variables: it explains the process by which they are related.

If something is a mediating variable :

• It’s caused by the independent variable
• It influences the dependent variable
• When it’s taken into account, the statistical correlation between the independent and dependent variables is higher than when it isn’t considered

Including mediators and moderators in your research helps you go beyond studying a simple relationship between two variables for a fuller picture of the real world. They are important to consider when studying complex correlational or causal relationships.

Mediators are part of the causal pathway of an effect, and they tell you how or why an effect takes place. Moderators usually help you judge the external validity of your study by identifying the limitations of when the relationship between variables holds.

You can think of independent and dependent variables in terms of cause and effect: an independent variable is the variable you think is the cause , while a dependent variable is the effect .

In an experiment, you manipulate the independent variable and measure the outcome in the dependent variable. For example, in an experiment about the effect of nutrients on crop growth:

• The  independent variable  is the amount of nutrients added to the crop field.
• The  dependent variable is the biomass of the crops at harvest time.

Defining your variables, and deciding how you will manipulate and measure them, is an important part of experimental design .

Discrete and continuous variables are two types of quantitative variables :

• Discrete variables represent counts (e.g., the number of objects in a collection).
• Continuous variables represent measurable amounts (e.g., water volume or weight).

Quantitative variables are any variables where the data represent amounts (e.g. height, weight, or age).

Categorical variables are any variables where the data represent groups. This includes rankings (e.g. finishing places in a race), classifications (e.g. brands of cereal), and binary outcomes (e.g. coin flips).

You need to know what type of variables you are working with to choose the right statistical test for your data and interpret your results .

Determining cause and effect is one of the most important parts of scientific research. It’s essential to know which is the cause – the independent variable – and which is the effect – the dependent variable.

You want to find out how blood sugar levels are affected by drinking diet cola and regular cola, so you conduct an experiment .

• The type of cola – diet or regular – is the independent variable .
• The level of blood sugar that you measure is the dependent variable – it changes depending on the type of cola.

No. The value of a dependent variable depends on an independent variable, so a variable cannot be both independent and dependent at the same time. It must be either the cause or the effect, not both.

Yes, but including more than one of either type requires multiple research questions .

For example, if you are interested in the effect of a diet on health, you can use multiple measures of health: blood sugar, blood pressure, weight, pulse, and many more. Each of these is its own dependent variable with its own research question.

You could also choose to look at the effect of exercise levels as well as diet, or even the additional effect of the two combined. Each of these is a separate independent variable .

To ensure the internal validity of an experiment , you should only change one independent variable at a time.

To ensure the internal validity of your research, you must consider the impact of confounding variables. If you fail to account for them, you might over- or underestimate the causal relationship between your independent and dependent variables , or even find a causal relationship where none exists.

A confounding variable is closely related to both the independent and dependent variables in a study. An independent variable represents the supposed cause , while the dependent variable is the supposed effect . A confounding variable is a third variable that influences both the independent and dependent variables.

Failing to account for confounding variables can cause you to wrongly estimate the relationship between your independent and dependent variables.

There are several methods you can use to decrease the impact of confounding variables on your research: restriction, matching, statistical control, and randomisation.

In restriction , you restrict your sample by only including certain subjects that have the same values of potential confounding variables.

In matching , you match each of the subjects in your treatment group with a counterpart in the comparison group. The matched subjects have the same values on any potential confounding variables, and only differ in the independent variable .

In statistical control , you include potential confounders as variables in your regression .

In randomisation , you randomly assign the treatment (or independent variable) in your study to a sufficiently large number of subjects, which allows you to control for all potential confounding variables.

In scientific research, concepts are the abstract ideas or phenomena that are being studied (e.g., educational achievement). Variables are properties or characteristics of the concept (e.g., performance at school), while indicators are ways of measuring or quantifying variables (e.g., yearly grade reports).

The process of turning abstract concepts into measurable variables and indicators is called operationalisation .

In statistics, ordinal and nominal variables are both considered categorical variables .

Even though ordinal data can sometimes be numerical, not all mathematical operations can be performed on them.

A control variable is any variable that’s held constant in a research study. It’s not a variable of interest in the study, but it’s controlled because it could influence the outcomes.

Control variables help you establish a correlational or causal relationship between variables by enhancing internal validity .

If you don’t control relevant extraneous variables , they may influence the outcomes of your study, and you may not be able to demonstrate that your results are really an effect of your independent variable .

‘Controlling for a variable’ means measuring extraneous variables and accounting for them statistically to remove their effects on other variables.

Researchers often model control variable data along with independent and dependent variable data in regression analyses and ANCOVAs . That way, you can isolate the control variable’s effects from the relationship between the variables of interest.

An extraneous variable is any variable that you’re not investigating that can potentially affect the dependent variable of your research study.

A confounding variable is a type of extraneous variable that not only affects the dependent variable, but is also related to the independent variable.

There are 4 main types of extraneous variables :

• Demand characteristics : Environmental cues that encourage participants to conform to researchers’ expectations
• Experimenter effects : Unintentional actions by researchers that influence study outcomes
• Situational variables : Eenvironmental variables that alter participants’ behaviours
• Participant variables : Any characteristic or aspect of a participant’s background that could affect study results

The difference between explanatory and response variables is simple:

• An explanatory variable is the expected cause, and it explains the results.
• A response variable is the expected effect, and it responds to other variables.

The term ‘ explanatory variable ‘ is sometimes preferred over ‘ independent variable ‘ because, in real-world contexts, independent variables are often influenced by other variables. This means they aren’t totally independent.

Multiple independent variables may also be correlated with each other, so ‘explanatory variables’ is a more appropriate term.

On graphs, the explanatory variable is conventionally placed on the x -axis, while the response variable is placed on the y -axis.

• If you have quantitative variables , use a scatterplot or a line graph.
• If your response variable is categorical, use a scatterplot or a line graph.
• If your explanatory variable is categorical, use a bar graph.

A correlation is usually tested for two variables at a time, but you can test correlations between three or more variables.

An independent variable is the variable you manipulate, control, or vary in an experimental study to explore its effects. It’s called ‘independent’ because it’s not influenced by any other variables in the study.

Independent variables are also called:

• Explanatory variables (they explain an event or outcome)
• Predictor variables (they can be used to predict the value of a dependent variable)
• Right-hand-side variables (they appear on the right-hand side of a regression equation)

A dependent variable is what changes as a result of the independent variable manipulation in experiments . It’s what you’re interested in measuring, and it ‘depends’ on your independent variable.

In statistics, dependent variables are also called:

• Response variables (they respond to a change in another variable)
• Outcome variables (they represent the outcome you want to measure)
• Left-hand-side variables (they appear on the left-hand side of a regression equation)

Deductive reasoning is commonly used in scientific research, and it’s especially associated with quantitative research .

In research, you might have come across something called the hypothetico-deductive method . It’s the scientific method of testing hypotheses to check whether your predictions are substantiated by real-world data.

Deductive reasoning is a logical approach where you progress from general ideas to specific conclusions. It’s often contrasted with inductive reasoning , where you start with specific observations and form general conclusions.

Deductive reasoning is also called deductive logic.

Inductive reasoning is a method of drawing conclusions by going from the specific to the general. It’s usually contrasted with deductive reasoning, where you proceed from general information to specific conclusions.

Inductive reasoning is also called inductive logic or bottom-up reasoning.

In inductive research , you start by making observations or gathering data. Then, you take a broad scan of your data and search for patterns. Finally, you make general conclusions that you might incorporate into theories.

Inductive reasoning is a bottom-up approach, while deductive reasoning is top-down.

Inductive reasoning takes you from the specific to the general, while in deductive reasoning, you make inferences by going from general premises to specific conclusions.

There are many different types of inductive reasoning that people use formally or informally.

Here are a few common types:

• Inductive generalisation : You use observations about a sample to come to a conclusion about the population it came from.
• Statistical generalisation: You use specific numbers about samples to make statements about populations.
• Causal reasoning: You make cause-and-effect links between different things.
• Sign reasoning: You make a conclusion about a correlational relationship between different things.
• Analogical reasoning: You make a conclusion about something based on its similarities to something else.

It’s often best to ask a variety of people to review your measurements. You can ask experts, such as other researchers, or laypeople, such as potential participants, to judge the face validity of tests.

While experts have a deep understanding of research methods , the people you’re studying can provide you with valuable insights you may have missed otherwise.

Face validity is important because it’s a simple first step to measuring the overall validity of a test or technique. It’s a relatively intuitive, quick, and easy way to start checking whether a new measure seems useful at first glance.

Good face validity means that anyone who reviews your measure says that it seems to be measuring what it’s supposed to. With poor face validity, someone reviewing your measure may be left confused about what you’re measuring and why you’re using this method.

Face validity is about whether a test appears to measure what it’s supposed to measure. This type of validity is concerned with whether a measure seems relevant and appropriate for what it’s assessing only on the surface.

Statistical analyses are often applied to test validity with data from your measures. You test convergent validity and discriminant validity with correlations to see if results from your test are positively or negatively related to those of other established tests.

You can also use regression analyses to assess whether your measure is actually predictive of outcomes that you expect it to predict theoretically. A regression analysis that supports your expectations strengthens your claim of construct validity .

When designing or evaluating a measure, construct validity helps you ensure you’re actually measuring the construct you’re interested in. If you don’t have construct validity, you may inadvertently measure unrelated or distinct constructs and lose precision in your research.

Construct validity is often considered the overarching type of measurement validity ,  because it covers all of the other types. You need to have face validity , content validity, and criterion validity to achieve construct validity.

Construct validity is about how well a test measures the concept it was designed to evaluate. It’s one of four types of measurement validity , which includes construct validity, face validity , and criterion validity.

There are two subtypes of construct validity.

• Convergent validity : The extent to which your measure corresponds to measures of related constructs
• Discriminant validity: The extent to which your measure is unrelated or negatively related to measures of distinct constructs

Attrition bias can skew your sample so that your final sample differs significantly from your original sample. Your sample is biased because some groups from your population are underrepresented.

With a biased final sample, you may not be able to generalise your findings to the original population that you sampled from, so your external validity is compromised.

There are seven threats to external validity : selection bias , history, experimenter effect, Hawthorne effect , testing effect, aptitude-treatment, and situation effect.

The two types of external validity are population validity (whether you can generalise to other groups of people) and ecological validity (whether you can generalise to other situations and settings).

The external validity of a study is the extent to which you can generalise your findings to different groups of people, situations, and measures.

Attrition bias is a threat to internal validity . In experiments, differential rates of attrition between treatment and control groups can skew results.

This bias can affect the relationship between your independent and dependent variables . It can make variables appear to be correlated when they are not, or vice versa.

Internal validity is the extent to which you can be confident that a cause-and-effect relationship established in a study cannot be explained by other factors.

There are eight threats to internal validity : history, maturation, instrumentation, testing, selection bias , regression to the mean, social interaction, and attrition .

A sampling error is the difference between a population parameter and a sample statistic .

A statistic refers to measures about the sample , while a parameter refers to measures about the population .

Populations are used when a research question requires data from every member of the population. This is usually only feasible when the population is small and easily accessible.

Systematic sampling is a probability sampling method where researchers select members of the population at a regular interval – for example, by selecting every 15th person on a list of the population. If the population is in a random order, this can imitate the benefits of simple random sampling .

There are three key steps in systematic sampling :

• Define and list your population , ensuring that it is not ordered in a cyclical or periodic order.
• Decide on your sample size and calculate your interval, k , by dividing your population by your target sample size.
• Choose every k th member of the population as your sample.

Yes, you can create a stratified sample using multiple characteristics, but you must ensure that every participant in your study belongs to one and only one subgroup. In this case, you multiply the numbers of subgroups for each characteristic to get the total number of groups.

For example, if you were stratifying by location with three subgroups (urban, rural, or suburban) and marital status with five subgroups (single, divorced, widowed, married, or partnered), you would have 3 × 5 = 15 subgroups.

You should use stratified sampling when your sample can be divided into mutually exclusive and exhaustive subgroups that you believe will take on different mean values for the variable that you’re studying.

Using stratified sampling will allow you to obtain more precise (with lower variance ) statistical estimates of whatever you are trying to measure.

For example, say you want to investigate how income differs based on educational attainment, but you know that this relationship can vary based on race. Using stratified sampling, you can ensure you obtain a large enough sample from each racial group, allowing you to draw more precise conclusions.

In stratified sampling , researchers divide subjects into subgroups called strata based on characteristics that they share (e.g., race, gender, educational attainment).

Once divided, each subgroup is randomly sampled using another probability sampling method .

Multistage sampling can simplify data collection when you have large, geographically spread samples, and you can obtain a probability sample without a complete sampling frame.

But multistage sampling may not lead to a representative sample, and larger samples are needed for multistage samples to achieve the statistical properties of simple random samples .

In multistage sampling , you can use probability or non-probability sampling methods.

For a probability sample, you have to probability sampling at every stage. You can mix it up by using simple random sampling , systematic sampling , or stratified sampling to select units at different stages, depending on what is applicable and relevant to your study.

Cluster sampling is a probability sampling method in which you divide a population into clusters, such as districts or schools, and then randomly select some of these clusters as your sample.

The clusters should ideally each be mini-representations of the population as a whole.

There are three types of cluster sampling : single-stage, double-stage and multi-stage clustering. In all three types, you first divide the population into clusters, then randomly select clusters for use in your sample.

• In single-stage sampling , you collect data from every unit within the selected clusters.
• In double-stage sampling , you select a random sample of units from within the clusters.
• In multi-stage sampling , you repeat the procedure of randomly sampling elements from within the clusters until you have reached a manageable sample.

Cluster sampling is more time- and cost-efficient than other probability sampling methods , particularly when it comes to large samples spread across a wide geographical area.

However, it provides less statistical certainty than other methods, such as simple random sampling , because it is difficult to ensure that your clusters properly represent the population as a whole.

If properly implemented, simple random sampling is usually the best sampling method for ensuring both internal and external validity . However, it can sometimes be impractical and expensive to implement, depending on the size of the population to be studied,

If you have a list of every member of the population and the ability to reach whichever members are selected, you can use simple random sampling.

The American Community Survey  is an example of simple random sampling . In order to collect detailed data on the population of the US, the Census Bureau officials randomly select 3.5 million households per year and use a variety of methods to convince them to fill out the survey.

Simple random sampling is a type of probability sampling in which the researcher randomly selects a subset of participants from a population . Each member of the population has an equal chance of being selected. Data are then collected from as large a percentage as possible of this random subset.

Sampling bias occurs when some members of a population are systematically more likely to be selected in a sample than others.

In multistage sampling , or multistage cluster sampling, you draw a sample from a population using smaller and smaller groups at each stage.

This method is often used to collect data from a large, geographically spread group of people in national surveys, for example. You take advantage of hierarchical groupings (e.g., from county to city to neighbourhood) to create a sample that’s less expensive and time-consuming to collect data from.

In non-probability sampling , the sample is selected based on non-random criteria, and not every member of the population has a chance of being included.

Common non-probability sampling methods include convenience sampling , voluntary response sampling, purposive sampling , snowball sampling , and quota sampling .

Probability sampling means that every member of the target population has a known chance of being included in the sample.

Probability sampling methods include simple random sampling , systematic sampling , stratified sampling , and cluster sampling .

Samples are used to make inferences about populations . Samples are easier to collect data from because they are practical, cost-effective, convenient, and manageable.

While a between-subjects design has fewer threats to internal validity , it also requires more participants for high statistical power than a within-subjects design .

Advantages:

• Prevents carryover effects of learning and fatigue.
• Shorter study duration.

Disadvantages:

• Needs larger samples for high power.
• Uses more resources to recruit participants, administer sessions, cover costs, etc.
• Individual differences may be an alternative explanation for results.

In a factorial design, multiple independent variables are tested.

If you test two variables, each level of one independent variable is combined with each level of the other independent variable to create different conditions.

Yes. Between-subjects and within-subjects designs can be combined in a single study when you have two or more independent variables (a factorial design). In a mixed factorial design, one variable is altered between subjects and another is altered within subjects.

Within-subjects designs have many potential threats to internal validity , but they are also very statistically powerful .

• Only requires small samples
• Statistically powerful
• Removes the effects of individual differences on the outcomes
• Internal validity threats reduce the likelihood of establishing a direct relationship between variables
• Time-related effects, such as growth, can influence the outcomes
• Carryover effects mean that the specific order of different treatments affect the outcomes

Quasi-experimental design is most useful in situations where it would be unethical or impractical to run a true experiment .

Quasi-experiments have lower internal validity than true experiments, but they often have higher external validity  as they can use real-world interventions instead of artificial laboratory settings.

In experimental research, random assignment is a way of placing participants from your sample into different groups using randomisation. With this method, every member of the sample has a known or equal chance of being placed in a control group or an experimental group.

A quasi-experiment is a type of research design that attempts to establish a cause-and-effect relationship. The main difference between this and a true experiment is that the groups are not randomly assigned.

In a between-subjects design , every participant experiences only one condition, and researchers assess group differences between participants in various conditions.

In a within-subjects design , each participant experiences all conditions, and researchers test the same participants repeatedly for differences between conditions.

The word ‘between’ means that you’re comparing different conditions between groups, while the word ‘within’ means you’re comparing different conditions within the same group.

A confounding variable , also called a confounder or confounding factor, is a third variable in a study examining a potential cause-and-effect relationship.

A confounding variable is related to both the supposed cause and the supposed effect of the study. It can be difficult to separate the true effect of the independent variable from the effect of the confounding variable.

In your research design , it’s important to identify potential confounding variables and plan how you will reduce their impact.

Triangulation can help:

• Reduce bias that comes from using a single method, theory, or investigator
• Enhance validity by approaching the same topic with different tools
• Establish credibility by giving you a complete picture of the research problem

But triangulation can also pose problems:

• It’s time-consuming and labour-intensive, often involving an interdisciplinary team.
• Your results may be inconsistent or even contradictory.

There are four main types of triangulation :

• Data triangulation : Using data from different times, spaces, and people
• Investigator triangulation : Involving multiple researchers in collecting or analysing data
• Theory triangulation : Using varying theoretical perspectives in your research
• Methodological triangulation : Using different methodologies to approach the same topic

Experimental designs are a set of procedures that you plan in order to examine the relationship between variables that interest you.

To design a successful experiment, first identify:

• A testable hypothesis
• One or more independent variables that you will manipulate
• One or more dependent variables that you will measure

When designing the experiment, first decide:

• How your variable(s) will be manipulated
• How you will control for any potential confounding or lurking variables
• How many subjects you will include
• How you will assign treatments to your subjects

Exploratory research explores the main aspects of a new or barely researched question.

Explanatory research explains the causes and effects of an already widely researched question.

The key difference between observational studies and experiments is that, done correctly, an observational study will never influence the responses or behaviours of participants. Experimental designs will have a treatment condition applied to at least a portion of participants.

An observational study could be a good fit for your research if your research question is based on things you observe. If you have ethical, logistical, or practical concerns that make an experimental design challenging, consider an observational study. Remember that in an observational study, it is critical that there be no interference or manipulation of the research subjects. Since it’s not an experiment, there are no control or treatment groups either.

These are four of the most common mixed methods designs :

• Convergent parallel: Quantitative and qualitative data are collected at the same time and analysed separately. After both analyses are complete, compare your results to draw overall conclusions. 
• Embedded: Quantitative and qualitative data are collected at the same time, but within a larger quantitative or qualitative design. One type of data is secondary to the other.
• Explanatory sequential: Quantitative data is collected and analysed first, followed by qualitative data. You can use this design if you think your qualitative data will explain and contextualise your quantitative findings.
• Exploratory sequential: Qualitative data is collected and analysed first, followed by quantitative data. You can use this design if you think the quantitative data will confirm or validate your qualitative findings.

Triangulation in research means using multiple datasets, methods, theories and/or investigators to address a research question. It’s a research strategy that can help you enhance the validity and credibility of your findings.

Triangulation is mainly used in qualitative research , but it’s also commonly applied in quantitative research . Mixed methods research always uses triangulation.

Operationalisation means turning abstract conceptual ideas into measurable observations.

For example, the concept of social anxiety isn’t directly observable, but it can be operationally defined in terms of self-rating scores, behavioural avoidance of crowded places, or physical anxiety symptoms in social situations.

Before collecting data , it’s important to consider how you will operationalise the variables that you want to measure.

Hypothesis testing is a formal procedure for investigating our ideas about the world using statistics. It is used by scientists to test specific predictions, called hypotheses , by calculating how likely it is that a pattern or relationship between variables could have arisen by chance.

There are five common approaches to qualitative research :

• Grounded theory involves collecting data in order to develop new theories.
• Ethnography involves immersing yourself in a group or organisation to understand its culture.
• Narrative research involves interpreting stories to understand how people make sense of their experiences and perceptions.
• Phenomenological research involves investigating phenomena through people’s lived experiences.
• Action research links theory and practice in several cycles to drive innovative changes.

There are various approaches to qualitative data analysis , but they all share five steps in common:

• Prepare and organise your data.
• Review and explore your data.
• Develop a data coding system.
• Assign codes to the data.
• Identify recurring themes.

The specifics of each step depend on the focus of the analysis. Some common approaches include textual analysis , thematic analysis , and discourse analysis .

In mixed methods research , you use both qualitative and quantitative data collection and analysis methods to answer your research question .

Methodology refers to the overarching strategy and rationale of your research project . It involves studying the methods used in your field and the theories or principles behind them, in order to develop an approach that matches your objectives.

Methods are the specific tools and procedures you use to collect and analyse data (e.g. experiments, surveys , and statistical tests ).

In shorter scientific papers, where the aim is to report the findings of a specific study, you might simply describe what you did in a methods section .

In a longer or more complex research project, such as a thesis or dissertation , you will probably include a methodology section , where you explain your approach to answering the research questions and cite relevant sources to support your choice of methods.

The research methods you use depend on the type of data you need to answer your research question .

• If you want to measure something or test a hypothesis , use quantitative methods . If you want to explore ideas, thoughts, and meanings, use qualitative methods .
• If you want to analyse a large amount of readily available data, use secondary data. If you want data specific to your purposes with control over how they are generated, collect primary data.
• If you want to establish cause-and-effect relationships between variables , use experimental methods. If you want to understand the characteristics of a research subject, use descriptive methods.

Ask our team

Want to contact us directly? No problem. We are always here for you.

• Chat with us
• Email [email protected]
• Call +44 (0)20 3917 4242
• WhatsApp +31 20 261 6040

Our support team is here to help you daily via chat, WhatsApp, email, or phone between 9:00 a.m. to 11:00 p.m. CET.

Our APA experts default to APA 7 for editing and formatting. For the Citation Editing Service you are able to choose between APA 6 and 7.

Yes, if your document is longer than 20,000 words, you will get a sample of approximately 2,000 words. This sample edit gives you a first impression of the editor’s editing style and a chance to ask questions and give feedback.

How does the sample edit work?

You will receive the sample edit within 24 hours after placing your order. You then have 24 hours to let us know if you’re happy with the sample or if there’s something you would like the editor to do differently.

Read more about how the sample edit works

Yes, you can upload your document in sections.

We try our best to ensure that the same editor checks all the different sections of your document. When you upload a new file, our system recognizes you as a returning customer, and we immediately contact the editor who helped you before.

However, we cannot guarantee that the same editor will be available. Your chances are higher if

• You send us your text as soon as possible and
• You can be flexible about the deadline.

Please note that the shorter your deadline is, the lower the chance that your previous editor is not available.

If your previous editor isn’t available, then we will inform you immediately and look for another qualified editor. Fear not! Every Scribbr editor follows the  Scribbr Improvement Model  and will deliver high-quality work.

Yes, our editors also work during the weekends and holidays.

Because we have many editors available, we can check your document 24 hours per day and 7 days per week, all year round.

If you choose a 72 hour deadline and upload your document on a Thursday evening, you’ll have your thesis back by Sunday evening!

Yes! Our editors are all native speakers, and they have lots of experience editing texts written by ESL students. They will make sure your grammar is perfect and point out any sentences that are difficult to understand. They’ll also notice your most common mistakes, and give you personal feedback to improve your writing in English.

Every Scribbr order comes with our award-winning Proofreading & Editing service , which combines two important stages of the revision process.

For a more comprehensive edit, you can add a Structure Check or Clarity Check to your order. With these building blocks, you can customize the kind of feedback you receive.

You might be familiar with a different set of editing terms. To help you understand what you can expect at Scribbr, we created this table:

View an example

When you place an order, you can specify your field of study and we’ll match you with an editor who has familiarity with this area.

However, our editors are language specialists, not academic experts in your field. Your editor’s job is not to comment on the content of your dissertation, but to improve your language and help you express your ideas as clearly and fluently as possible.

This means that your editor will understand your text well enough to give feedback on its clarity, logic and structure, but not on the accuracy or originality of its content.

Good academic writing should be understandable to a non-expert reader, and we believe that academic editing is a discipline in itself. The research, ideas and arguments are all yours – we’re here to make sure they shine!

After your document has been edited, you will receive an email with a link to download the document.

The editor has made changes to your document using ‘Track Changes’ in Word. This means that you only have to accept or ignore the changes that are made in the text one by one.

It is also possible to accept all changes at once. However, we strongly advise you not to do so for the following reasons:

• You can learn a lot by looking at the mistakes you made.
• The editors don’t only change the text – they also place comments when sentences or sometimes even entire paragraphs are unclear. You should read through these comments and take into account your editor’s tips and suggestions.
• With a final read-through, you can make sure you’re 100% happy with your text before you submit!

You choose the turnaround time when ordering. We can return your dissertation within 24 hours , 3 days or 1 week . These timescales include weekends and holidays. As soon as you’ve paid, the deadline is set, and we guarantee to meet it! We’ll notify you by text and email when your editor has completed the job.

Very large orders might not be possible to complete in 24 hours. On average, our editors can complete around 13,000 words in a day while maintaining our high quality standards. If your order is longer than this and urgent, contact us to discuss possibilities.

Always leave yourself enough time to check through the document and accept the changes before your submission deadline.

Scribbr is specialised in editing study related documents. We check:

• Graduation projects
• Dissertations
• Admissions essays
• College essays
• Application essays
• Personal statements
• Process reports
• Reflections
• Internship reports
• Academic papers
• Research proposals
• Prospectuses

Calculate the costs

The fastest turnaround time is 24 hours.

You can upload your document at any time and choose between four deadlines:

At Scribbr, we promise to make every customer 100% happy with the service we offer. Our philosophy: Your complaint is always justified – no denial, no doubts.

Our customer support team is here to find the solution that helps you the most, whether that’s a free new edit or a refund for the service.

Yes, in the order process you can indicate your preference for American, British, or Australian English .

If you don’t choose one, your editor will follow the style of English you currently use. If your editor has any questions about this, we will contact you.

• Bipolar Disorder
• Therapy Center
• When To See a Therapist
• Types of Therapy
• Best Online Therapy
• Best Couples Therapy
• Managing Stress
• Sleep and Dreaming
• Understanding Emotions
• Self-Improvement
• Healthy Relationships
• Student Resources
• Personality Types
• Sweepstakes
• Guided Meditations
• Verywell Mind Insights
• 2024 Verywell Mind 25
• Mental Health in the Classroom
• Editorial Process
• Meet Our Review Board
• Crisis Support

Correlation Studies in Psychology Research

Determining the relationship between two or more variables.

Verywell / Brianna Gilmartin

• Characteristics

Potential Pitfalls

Frequently asked questions.

A correlational study is a type of research design that looks at the relationships between two or more variables. Correlational studies are non-experimental, which means that the experimenter does not manipulate or control any of the variables.

A correlation refers to a relationship between two variables. Correlations can be strong or weak and positive or negative. Sometimes, there is no correlation.

There are three possible outcomes of a correlation study: a positive correlation, a negative correlation, or no correlation. Researchers can present the results using a numerical value called the correlation coefficient, a measure of the correlation strength. It can range from –1.00 (negative) to +1.00 (positive). A correlation coefficient of 0 indicates no correlation.

• Positive correlations : Both variables increase or decrease at the same time. A correlation coefficient close to +1.00 indicates a strong positive correlation.
• Negative correlations : As the amount of one variable increases, the other decreases (and vice versa). A correlation coefficient close to -1.00 indicates a strong negative correlation.
• No correlation : There is no relationship between the two variables. A correlation coefficient of 0 indicates no correlation.

Characteristics of a Correlational Study

Correlational studies are often used in psychology, as well as other fields like medicine. Correlational research is a preliminary way to gather information about a topic. The method is also useful if researchers are unable to perform an experiment.

Researchers use correlations to see if a relationship between two or more variables exists, but the variables themselves are not under the control of the researchers.

While correlational research can demonstrate a relationship between variables, it cannot prove that changing one variable will change another. In other words, correlational studies cannot prove cause-and-effect relationships.

When you encounter research that refers to a "link" or an "association" between two things, they are most likely talking about a correlational study.

Types of Correlational Research

There are three types of correlational research: naturalistic observation, the survey method, and archival research. Each type has its own purpose, as well as its pros and cons.

Naturalistic Observation

The naturalistic observation method involves observing and recording variables of interest in a natural setting without interference or manipulation.  

Can inspire ideas for further research

Option if lab experiment not available

Variables are viewed in natural setting

Can be time-consuming and expensive

Extraneous variables can't be controlled

No scientific control of variables

Subjects might behave differently if aware of being observed

This method is well-suited to studies where researchers want to see how variables behave in their natural setting or state.   Inspiration can then be drawn from the observations to inform future avenues of research.

In some cases, it might be the only method available to researchers; for example, if lab experimentation would be precluded by access, resources, or ethics. It might be preferable to not being able to conduct research at all, but the method can be costly and usually takes a lot of time.  

Naturalistic observation presents several challenges for researchers. For one, it does not allow them to control or influence the variables in any way nor can they change any possible external variables.

However, this does not mean that researchers will get reliable data from watching the variables, or that the information they gather will be free from bias.

For example, study subjects might act differently if they know that they are being watched. The researchers might not be aware that the behavior that they are observing is not necessarily the subject's natural state (i.e., how they would act if they did not know they were being watched).

Researchers also need to be aware of their biases, which can affect the observation and interpretation of a subject's behavior.  

Surveys and questionnaires are some of the most common methods used for psychological research. The survey method involves having a  random sample  of participants complete a survey, test, or questionnaire related to the variables of interest.   Random sampling is vital to the generalizability of a survey's results.

Cheap, easy, and fast

Can collect large amounts of data in a short amount of time

Results can be affected by poor survey questions

Results can be affected by unrepresentative sample

Outcomes can be affected by participants

If researchers need to gather a large amount of data in a short period of time, a survey is likely to be the fastest, easiest, and cheapest option.  

It's also a flexible method because it lets researchers create data-gathering tools that will help ensure they get the information they need (survey responses) from all the sources they want to use (a random sample of participants taking the survey).

Survey data might be cost-efficient and easy to get, but it has its downsides. For one, the data is not always reliable—particularly if the survey questions are poorly written or the overall design or delivery is weak.   Data is also affected by specific faults, such as unrepresented or underrepresented samples .

The use of surveys relies on participants to provide useful data. Researchers need to be aware of the specific factors related to the people taking the survey that will affect its outcome.

For example, some people might struggle to understand the questions. A person might answer a particular way to try to please the researchers or to try to control how the researchers perceive them (such as trying to make themselves "look better").

Sometimes, respondents might not even realize that their answers are incorrect or misleading because of mistaken memories .

Archival Research

Many areas of psychological research benefit from analyzing studies that were conducted long ago by other researchers, as well as reviewing historical records and case studies.

For example, in an experiment known as  "The Irritable Heart ," researchers used digitalized records containing information on American Civil War veterans to learn more about post-traumatic stress disorder (PTSD).

Large amount of data

Can be less expensive

Researchers cannot change participant behavior

Can be unreliable

Information might be missing

No control over data collection methods

Using records, databases, and libraries that are publicly accessible or accessible through their institution can help researchers who might not have a lot of money to support their research efforts.

Free and low-cost resources are available to researchers at all levels through academic institutions, museums, and data repositories around the world.

Another potential benefit is that these sources often provide an enormous amount of data that was collected over a very long period of time, which can give researchers a way to view trends, relationships, and outcomes related to their research.

While the inability to change variables can be a disadvantage of some methods, it can be a benefit of archival research. That said, using historical records or information that was collected a long time ago also presents challenges. For one, important information might be missing or incomplete and some aspects of older studies might not be useful to researchers in a modern context.

A primary issue with archival research is reliability. When reviewing old research, little information might be available about who conducted the research, how a study was designed, who participated in the research, as well as how data was collected and interpreted.

Researchers can also be presented with ethical quandaries—for example, should modern researchers use data from studies that were conducted unethically or with questionable ethics?

You've probably heard the phrase, "correlation does not equal causation." This means that while correlational research can suggest that there is a relationship between two variables, it cannot prove that one variable will change another.

For example, researchers might perform a correlational study that suggests there is a relationship between academic success and a person's self-esteem. However, the study cannot show that academic success changes a person's self-esteem.

To determine why the relationship exists, researchers would need to consider and experiment with other variables, such as the subject's social relationships, cognitive abilities, personality, and socioeconomic status.

The difference between a correlational study and an experimental study involves the manipulation of variables. Researchers do not manipulate variables in a correlational study, but they do control and systematically vary the independent variables in an experimental study. Correlational studies allow researchers to detect the presence and strength of a relationship between variables, while experimental studies allow researchers to look for cause and effect relationships.

If the study involves the systematic manipulation of the levels of a variable, it is an experimental study. If researchers are measuring what is already present without actually changing the variables, then is a correlational study.

The variables in a correlational study are what the researcher measures. Once measured, researchers can then use statistical analysis to determine the existence, strength, and direction of the relationship. However, while correlational studies can say that variable X and variable Y have a relationship, it does not mean that X causes Y.

The goal of correlational research is often to look for relationships, describe these relationships, and then make predictions. Such research can also often serve as a jumping off point for future experimental research. 

Heath W. Psychology Research Methods . Cambridge University Press; 2018:134-156.

Schneider FW. Applied Social Psychology . 2nd ed. SAGE; 2012:50-53.

Curtis EA, Comiskey C, Dempsey O. Importance and use of correlational research .  Nurse Researcher . 2016;23(6):20-25. doi:10.7748/nr.2016.e1382

Carpenter S. Visualizing Psychology . 3rd ed. John Wiley & Sons; 2012:14-30.

Pizarro J, Silver RC, Prause J. Physical and mental health costs of traumatic war experiences among civil war veterans .  Arch Gen Psychiatry . 2006;63(2):193. doi:10.1001/archpsyc.63.2.193

Post SG. The echo of Nuremberg: Nazi data and ethics .  J Med Ethics . 1991;17(1):42-44. doi:10.1136/jme.17.1.42

Lau F. Chapter 12 Methods for Correlational Studies . In: Lau F, Kuziemsky C, eds. Handbook of eHealth Evaluation: An Evidence-based Approach . University of Victoria.

Akoglu H. User's guide to correlation coefficients .  Turk J Emerg Med . 2018;18(3):91-93. doi:10.1016/j.tjem.2018.08.001

Price PC. Research Methods in Psychology . California State University.

By Kendra Cherry, MSEd Kendra Cherry, MS, is a psychosocial rehabilitation specialist, psychology educator, and author of the "Everything Psychology Book."

Want to create or adapt books like this? Learn more about how Pressbooks supports open publishing practices.

2.2 Psychologists Use Descriptive, Correlational, and Experimental Research Designs to Understand Behavior

Learning objectives.

• Differentiate the goals of descriptive, correlational, and experimental research designs and explain the advantages and disadvantages of each.
• Explain the goals of descriptive research and the statistical techniques used to interpret it.
• Summarize the uses of correlational research and describe why correlational research cannot be used to infer causality.
• Review the procedures of experimental research and explain how it can be used to draw causal inferences.

Psychologists agree that if their ideas and theories about human behavior are to be taken seriously, they must be backed up by data. However, the research of different psychologists is designed with different goals in mind, and the different goals require different approaches. These varying approaches, summarized in Table 2.2 “Characteristics of the Three Research Designs” , are known as research designs . A research design is the specific method a researcher uses to collect, analyze, and interpret data . Psychologists use three major types of research designs in their research, and each provides an essential avenue for scientific investigation. Descriptive research is research designed to provide a snapshot of the current state of affairs . Correlational research is research designed to discover relationships among variables and to allow the prediction of future events from present knowledge . Experimental research is research in which initial equivalence among research participants in more than one group is created, followed by a manipulation of a given experience for these groups and a measurement of the influence of the manipulation . Each of the three research designs varies according to its strengths and limitations, and it is important to understand how each differs.

Table 2.2 Characteristics of the Three Research Designs

Stangor, C. (2011). Research methods for the behavioral sciences (4th ed.). Mountain View, CA: Cengage.

Descriptive Research: Assessing the Current State of Affairs

Descriptive research is designed to create a snapshot of the current thoughts, feelings, or behavior of individuals. This section reviews three types of descriptive research: case studies , surveys , and naturalistic observation .

Sometimes the data in a descriptive research project are based on only a small set of individuals, often only one person or a single small group. These research designs are known as case studies — descriptive records of one or more individual’s experiences and behavior . Sometimes case studies involve ordinary individuals, as when developmental psychologist Jean Piaget used his observation of his own children to develop his stage theory of cognitive development. More frequently, case studies are conducted on individuals who have unusual or abnormal experiences or characteristics or who find themselves in particularly difficult or stressful situations. The assumption is that by carefully studying individuals who are socially marginal, who are experiencing unusual situations, or who are going through a difficult phase in their lives, we can learn something about human nature.

Sigmund Freud was a master of using the psychological difficulties of individuals to draw conclusions about basic psychological processes. Freud wrote case studies of some of his most interesting patients and used these careful examinations to develop his important theories of personality. One classic example is Freud’s description of “Little Hans,” a child whose fear of horses the psychoanalyst interpreted in terms of repressed sexual impulses and the Oedipus complex (Freud (1909/1964).

Political polls reported in newspapers and on the Internet are descriptive research designs that provide snapshots of the likely voting behavior of a population.

Another well-known case study is Phineas Gage, a man whose thoughts and emotions were extensively studied by cognitive psychologists after a railroad spike was blasted through his skull in an accident. Although there is question about the interpretation of this case study (Kotowicz, 2007), it did provide early evidence that the brain’s frontal lobe is involved in emotion and morality (Damasio et al., 2005). An interesting example of a case study in clinical psychology is described by Rokeach (1964), who investigated in detail the beliefs and interactions among three patients with schizophrenia, all of whom were convinced they were Jesus Christ.

In other cases the data from descriptive research projects come in the form of a survey — a measure administered through either an interview or a written questionnaire to get a picture of the beliefs or behaviors of a sample of people of interest . The people chosen to participate in the research (known as the sample ) are selected to be representative of all the people that the researcher wishes to know about (the population ). In election polls, for instance, a sample is taken from the population of all “likely voters” in the upcoming elections.

The results of surveys may sometimes be rather mundane, such as “Nine out of ten doctors prefer Tymenocin,” or “The median income in Montgomery County is $36,712.” Yet other times (particularly in discussions of social behavior), the results can be shocking: “More than 40,000 people are killed by gunfire in the United States every year,” or “More than 60% of women between the ages of 50 and 60 suffer from depression.” Descriptive research is frequently used by psychologists to get an estimate of the prevalence (or incidence ) of psychological disorders.

A final type of descriptive research—known as naturalistic observation —is research based on the observation of everyday events . For instance, a developmental psychologist who watches children on a playground and describes what they say to each other while they play is conducting descriptive research, as is a biopsychologist who observes animals in their natural habitats. One example of observational research involves a systematic procedure known as the strange situation , used to get a picture of how adults and young children interact. The data that are collected in the strange situation are systematically coded in a coding sheet such as that shown in Table 2.3 “Sample Coding Form Used to Assess Child’s and Mother’s Behavior in the Strange Situation” .

Table 2.3 Sample Coding Form Used to Assess Child’s and Mother’s Behavior in the Strange Situation

The results of descriptive research projects are analyzed using descriptive statistics — numbers that summarize the distribution of scores on a measured variable . Most variables have distributions similar to that shown in Figure 2.5 “Height Distribution” , where most of the scores are located near the center of the distribution, and the distribution is symmetrical and bell-shaped. A data distribution that is shaped like a bell is known as a normal distribution .

Table 2.4 Height and Family Income for 25 Students

Figure 2.5 Height Distribution

The distribution of the heights of the students in a class will form a normal distribution. In this sample the mean ( M ) = 67.12 and the standard deviation ( s ) = 2.74.

A distribution can be described in terms of its central tendency —that is, the point in the distribution around which the data are centered—and its dispersion , or spread. The arithmetic average, or arithmetic mean , is the most commonly used measure of central tendency . It is computed by calculating the sum of all the scores of the variable and dividing this sum by the number of participants in the distribution (denoted by the letter N ). In the data presented in Figure 2.5 “Height Distribution” , the mean height of the students is 67.12 inches. The sample mean is usually indicated by the letter M .

In some cases, however, the data distribution is not symmetrical. This occurs when there are one or more extreme scores (known as outliers ) at one end of the distribution. Consider, for instance, the variable of family income (see Figure 2.6 “Family Income Distribution” ), which includes an outlier (a value of $3,800,000). In this case the mean is not a good measure of central tendency. Although it appears from Figure 2.6 “Family Income Distribution” that the central tendency of the family income variable should be around $70,000, the mean family income is actually $223,960. The single very extreme income has a disproportionate impact on the mean, resulting in a value that does not well represent the central tendency.

The median is used as an alternative measure of central tendency when distributions are not symmetrical. The median is the score in the center of the distribution, meaning that 50% of the scores are greater than the median and 50% of the scores are less than the median . In our case, the median household income ($73,000) is a much better indication of central tendency than is the mean household income ($223,960).

Figure 2.6 Family Income Distribution

The distribution of family incomes is likely to be nonsymmetrical because some incomes can be very large in comparison to most incomes. In this case the median or the mode is a better indicator of central tendency than is the mean.

A final measure of central tendency, known as the mode , represents the value that occurs most frequently in the distribution . You can see from Figure 2.6 “Family Income Distribution” that the mode for the family income variable is $93,000 (it occurs four times).

In addition to summarizing the central tendency of a distribution, descriptive statistics convey information about how the scores of the variable are spread around the central tendency. Dispersion refers to the extent to which the scores are all tightly clustered around the central tendency, like this:

Or they may be more spread out away from it, like this:

One simple measure of dispersion is to find the largest (the maximum ) and the smallest (the minimum ) observed values of the variable and to compute the range of the variable as the maximum observed score minus the minimum observed score. You can check that the range of the height variable in Figure 2.5 “Height Distribution” is 72 – 62 = 10. The standard deviation , symbolized as s , is the most commonly used measure of dispersion . Distributions with a larger standard deviation have more spread. The standard deviation of the height variable is s = 2.74, and the standard deviation of the family income variable is s = $745,337.

An advantage of descriptive research is that it attempts to capture the complexity of everyday behavior. Case studies provide detailed information about a single person or a small group of people, surveys capture the thoughts or reported behaviors of a large population of people, and naturalistic observation objectively records the behavior of people or animals as it occurs naturally. Thus descriptive research is used to provide a relatively complete understanding of what is currently happening.

Despite these advantages, descriptive research has a distinct disadvantage in that, although it allows us to get an idea of what is currently happening, it is usually limited to static pictures. Although descriptions of particular experiences may be interesting, they are not always transferable to other individuals in other situations, nor do they tell us exactly why specific behaviors or events occurred. For instance, descriptions of individuals who have suffered a stressful event, such as a war or an earthquake, can be used to understand the individuals’ reactions to the event but cannot tell us anything about the long-term effects of the stress. And because there is no comparison group that did not experience the stressful situation, we cannot know what these individuals would be like if they hadn’t had the stressful experience.

Correlational Research: Seeking Relationships Among Variables

In contrast to descriptive research, which is designed primarily to provide static pictures, correlational research involves the measurement of two or more relevant variables and an assessment of the relationship between or among those variables. For instance, the variables of height and weight are systematically related (correlated) because taller people generally weigh more than shorter people. In the same way, study time and memory errors are also related, because the more time a person is given to study a list of words, the fewer errors he or she will make. When there are two variables in the research design, one of them is called the predictor variable and the other the outcome variable . The research design can be visualized like this, where the curved arrow represents the expected correlation between the two variables:

Figure 2.2.2

One way of organizing the data from a correlational study with two variables is to graph the values of each of the measured variables using a scatter plot . As you can see in Figure 2.10 “Examples of Scatter Plots” , a scatter plot is a visual image of the relationship between two variables . A point is plotted for each individual at the intersection of his or her scores for the two variables. When the association between the variables on the scatter plot can be easily approximated with a straight line, as in parts (a) and (b) of Figure 2.10 “Examples of Scatter Plots” , the variables are said to have a linear relationship .

When the straight line indicates that individuals who have above-average values for one variable also tend to have above-average values for the other variable, as in part (a), the relationship is said to be positive linear . Examples of positive linear relationships include those between height and weight, between education and income, and between age and mathematical abilities in children. In each case people who score higher on one of the variables also tend to score higher on the other variable. Negative linear relationships , in contrast, as shown in part (b), occur when above-average values for one variable tend to be associated with below-average values for the other variable. Examples of negative linear relationships include those between the age of a child and the number of diapers the child uses, and between practice on and errors made on a learning task. In these cases people who score higher on one of the variables tend to score lower on the other variable.

Relationships between variables that cannot be described with a straight line are known as nonlinear relationships . Part (c) of Figure 2.10 “Examples of Scatter Plots” shows a common pattern in which the distribution of the points is essentially random. In this case there is no relationship at all between the two variables, and they are said to be independent . Parts (d) and (e) of Figure 2.10 “Examples of Scatter Plots” show patterns of association in which, although there is an association, the points are not well described by a single straight line. For instance, part (d) shows the type of relationship that frequently occurs between anxiety and performance. Increases in anxiety from low to moderate levels are associated with performance increases, whereas increases in anxiety from moderate to high levels are associated with decreases in performance. Relationships that change in direction and thus are not described by a single straight line are called curvilinear relationships .

Figure 2.10 Examples of Scatter Plots

Some examples of relationships between two variables as shown in scatter plots. Note that the Pearson correlation coefficient ( r ) between variables that have curvilinear relationships will likely be close to zero.

Adapted from Stangor, C. (2011). Research methods for the behavioral sciences (4th ed.). Mountain View, CA: Cengage.

The most common statistical measure of the strength of linear relationships among variables is the Pearson correlation coefficient , which is symbolized by the letter r . The value of the correlation coefficient ranges from r = –1.00 to r = +1.00. The direction of the linear relationship is indicated by the sign of the correlation coefficient. Positive values of r (such as r = .54 or r = .67) indicate that the relationship is positive linear (i.e., the pattern of the dots on the scatter plot runs from the lower left to the upper right), whereas negative values of r (such as r = –.30 or r = –.72) indicate negative linear relationships (i.e., the dots run from the upper left to the lower right). The strength of the linear relationship is indexed by the distance of the correlation coefficient from zero (its absolute value). For instance, r = –.54 is a stronger relationship than r = .30, and r = .72 is a stronger relationship than r = –.57. Because the Pearson correlation coefficient only measures linear relationships, variables that have curvilinear relationships are not well described by r , and the observed correlation will be close to zero.

It is also possible to study relationships among more than two measures at the same time. A research design in which more than one predictor variable is used to predict a single outcome variable is analyzed through multiple regression (Aiken & West, 1991). Multiple regression is a statistical technique, based on correlation coefficients among variables, that allows predicting a single outcome variable from more than one predictor variable . For instance, Figure 2.11 “Prediction of Job Performance From Three Predictor Variables” shows a multiple regression analysis in which three predictor variables are used to predict a single outcome. The use of multiple regression analysis shows an important advantage of correlational research designs—they can be used to make predictions about a person’s likely score on an outcome variable (e.g., job performance) based on knowledge of other variables.

Figure 2.11 Prediction of Job Performance From Three Predictor Variables

Multiple regression allows scientists to predict the scores on a single outcome variable using more than one predictor variable.

An important limitation of correlational research designs is that they cannot be used to draw conclusions about the causal relationships among the measured variables. Consider, for instance, a researcher who has hypothesized that viewing violent behavior will cause increased aggressive play in children. He has collected, from a sample of fourth-grade children, a measure of how many violent television shows each child views during the week, as well as a measure of how aggressively each child plays on the school playground. From his collected data, the researcher discovers a positive correlation between the two measured variables.

Although this positive correlation appears to support the researcher’s hypothesis, it cannot be taken to indicate that viewing violent television causes aggressive behavior. Although the researcher is tempted to assume that viewing violent television causes aggressive play,

there are other possibilities. One alternate possibility is that the causal direction is exactly opposite from what has been hypothesized. Perhaps children who have behaved aggressively at school develop residual excitement that leads them to want to watch violent television shows at home:

Although this possibility may seem less likely, there is no way to rule out the possibility of such reverse causation on the basis of this observed correlation. It is also possible that both causal directions are operating and that the two variables cause each other:

Still another possible explanation for the observed correlation is that it has been produced by the presence of a common-causal variable (also known as a third variable ). A common-causal variable is a variable that is not part of the research hypothesis but that causes both the predictor and the outcome variable and thus produces the observed correlation between them . In our example a potential common-causal variable is the discipline style of the children’s parents. Parents who use a harsh and punitive discipline style may produce children who both like to watch violent television and who behave aggressively in comparison to children whose parents use less harsh discipline:

In this case, television viewing and aggressive play would be positively correlated (as indicated by the curved arrow between them), even though neither one caused the other but they were both caused by the discipline style of the parents (the straight arrows). When the predictor and outcome variables are both caused by a common-causal variable, the observed relationship between them is said to be spurious . A spurious relationship is a relationship between two variables in which a common-causal variable produces and “explains away” the relationship . If effects of the common-causal variable were taken away, or controlled for, the relationship between the predictor and outcome variables would disappear. In the example the relationship between aggression and television viewing might be spurious because by controlling for the effect of the parents’ disciplining style, the relationship between television viewing and aggressive behavior might go away.

Common-causal variables in correlational research designs can be thought of as “mystery” variables because, as they have not been measured, their presence and identity are usually unknown to the researcher. Since it is not possible to measure every variable that could cause both the predictor and outcome variables, the existence of an unknown common-causal variable is always a possibility. For this reason, we are left with the basic limitation of correlational research: Correlation does not demonstrate causation. It is important that when you read about correlational research projects, you keep in mind the possibility of spurious relationships, and be sure to interpret the findings appropriately. Although correlational research is sometimes reported as demonstrating causality without any mention being made of the possibility of reverse causation or common-causal variables, informed consumers of research, like you, are aware of these interpretational problems.

In sum, correlational research designs have both strengths and limitations. One strength is that they can be used when experimental research is not possible because the predictor variables cannot be manipulated. Correlational designs also have the advantage of allowing the researcher to study behavior as it occurs in everyday life. And we can also use correlational designs to make predictions—for instance, to predict from the scores on their battery of tests the success of job trainees during a training session. But we cannot use such correlational information to determine whether the training caused better job performance. For that, researchers rely on experiments.

Experimental Research: Understanding the Causes of Behavior

The goal of experimental research design is to provide more definitive conclusions about the causal relationships among the variables in the research hypothesis than is available from correlational designs. In an experimental research design, the variables of interest are called the independent variable (or variables ) and the dependent variable . The independent variable in an experiment is the causing variable that is created (manipulated) by the experimenter . The dependent variable in an experiment is a measured variable that is expected to be influenced by the experimental manipulation . The research hypothesis suggests that the manipulated independent variable or variables will cause changes in the measured dependent variables. We can diagram the research hypothesis by using an arrow that points in one direction. This demonstrates the expected direction of causality:

Figure 2.2.3

Research Focus: Video Games and Aggression

Consider an experiment conducted by Anderson and Dill (2000). The study was designed to test the hypothesis that viewing violent video games would increase aggressive behavior. In this research, male and female undergraduates from Iowa State University were given a chance to play with either a violent video game (Wolfenstein 3D) or a nonviolent video game (Myst). During the experimental session, the participants played their assigned video games for 15 minutes. Then, after the play, each participant played a competitive game with an opponent in which the participant could deliver blasts of white noise through the earphones of the opponent. The operational definition of the dependent variable (aggressive behavior) was the level and duration of noise delivered to the opponent. The design of the experiment is shown in Figure 2.17 “An Experimental Research Design” .

Figure 2.17 An Experimental Research Design

Two advantages of the experimental research design are (1) the assurance that the independent variable (also known as the experimental manipulation) occurs prior to the measured dependent variable, and (2) the creation of initial equivalence between the conditions of the experiment (in this case by using random assignment to conditions).

Experimental designs have two very nice features. For one, they guarantee that the independent variable occurs prior to the measurement of the dependent variable. This eliminates the possibility of reverse causation. Second, the influence of common-causal variables is controlled, and thus eliminated, by creating initial equivalence among the participants in each of the experimental conditions before the manipulation occurs.

The most common method of creating equivalence among the experimental conditions is through random assignment to conditions , a procedure in which the condition that each participant is assigned to is determined through a random process, such as drawing numbers out of an envelope or using a random number table . Anderson and Dill first randomly assigned about 100 participants to each of their two groups (Group A and Group B). Because they used random assignment to conditions, they could be confident that, before the experimental manipulation occurred, the students in Group A were, on average, equivalent to the students in Group B on every possible variable, including variables that are likely to be related to aggression, such as parental discipline style, peer relationships, hormone levels, diet—and in fact everything else.

Then, after they had created initial equivalence, Anderson and Dill created the experimental manipulation—they had the participants in Group A play the violent game and the participants in Group B play the nonviolent game. Then they compared the dependent variable (the white noise blasts) between the two groups, finding that the students who had viewed the violent video game gave significantly longer noise blasts than did the students who had played the nonviolent game.

Anderson and Dill had from the outset created initial equivalence between the groups. This initial equivalence allowed them to observe differences in the white noise levels between the two groups after the experimental manipulation, leading to the conclusion that it was the independent variable (and not some other variable) that caused these differences. The idea is that the only thing that was different between the students in the two groups was the video game they had played.

Despite the advantage of determining causation, experiments do have limitations. One is that they are often conducted in laboratory situations rather than in the everyday lives of people. Therefore, we do not know whether results that we find in a laboratory setting will necessarily hold up in everyday life. Second, and more important, is that some of the most interesting and key social variables cannot be experimentally manipulated. If we want to study the influence of the size of a mob on the destructiveness of its behavior, or to compare the personality characteristics of people who join suicide cults with those of people who do not join such cults, these relationships must be assessed using correlational designs, because it is simply not possible to experimentally manipulate these variables.

Key Takeaways

• Descriptive, correlational, and experimental research designs are used to collect and analyze data.
• Descriptive designs include case studies, surveys, and naturalistic observation. The goal of these designs is to get a picture of the current thoughts, feelings, or behaviors in a given group of people. Descriptive research is summarized using descriptive statistics.
• Correlational research designs measure two or more relevant variables and assess a relationship between or among them. The variables may be presented on a scatter plot to visually show the relationships. The Pearson Correlation Coefficient ( r ) is a measure of the strength of linear relationship between two variables.
• Common-causal variables may cause both the predictor and outcome variable in a correlational design, producing a spurious relationship. The possibility of common-causal variables makes it impossible to draw causal conclusions from correlational research designs.
• Experimental research involves the manipulation of an independent variable and the measurement of a dependent variable. Random assignment to conditions is normally used to create initial equivalence between the groups, allowing researchers to draw causal conclusions.

Exercises and Critical Thinking

• There is a negative correlation between the row that a student sits in in a large class (when the rows are numbered from front to back) and his or her final grade in the class. Do you think this represents a causal relationship or a spurious relationship, and why?
• Think of two variables (other than those mentioned in this book) that are likely to be correlated, but in which the correlation is probably spurious. What is the likely common-causal variable that is producing the relationship?
• Imagine a researcher wants to test the hypothesis that participating in psychotherapy will cause a decrease in reported anxiety. Describe the type of research design the investigator might use to draw this conclusion. What would be the independent and dependent variables in the research?

Aiken, L., & West, S. (1991). Multiple regression: Testing and interpreting interactions . Newbury Park, CA: Sage.

Ainsworth, M. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment: A psychological study of the strange situation . Hillsdale, NJ: Lawrence Erlbaum Associates.

Anderson, C. A., & Dill, K. E. (2000). Video games and aggressive thoughts, feelings, and behavior in the laboratory and in life. Journal of Personality and Social Psychology, 78 (4), 772–790.

Damasio, H., Grabowski, T., Frank, R., Galaburda, A. M., Damasio, A. R., Cacioppo, J. T., & Berntson, G. G. (2005). The return of Phineas Gage: Clues about the brain from the skull of a famous patient. In Social neuroscience: Key readings. (pp. 21–28). New York, NY: Psychology Press.

Freud, S. (1964). Analysis of phobia in a five-year-old boy. In E. A. Southwell & M. Merbaum (Eds.), Personality: Readings in theory and research (pp. 3–32). Belmont, CA: Wadsworth. (Original work published 1909)

Kotowicz, Z. (2007). The strange case of Phineas Gage. History of the Human Sciences, 20 (1), 115–131.

Rokeach, M. (1964). The three Christs of Ypsilanti: A psychological study . New York, NY: Knopf.

Introduction to Psychology Copyright © 2015 by University of Minnesota is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License , except where otherwise noted.

Have a language expert improve your writing

Run a free plagiarism check in 10 minutes, generate accurate citations for free.

• Knowledge Base

Methodology

• What Is a Case Study? | Definition, Examples & Methods

What Is a Case Study? | Definition, Examples & Methods

Published on May 8, 2019 by Shona McCombes . Revised on November 20, 2023.

A case study is a detailed study of a specific subject, such as a person, group, place, event, organization, or phenomenon. Case studies are commonly used in social, educational, clinical, and business research.

A case study research design usually involves qualitative methods , but quantitative methods are sometimes also used. Case studies are good for describing , comparing, evaluating and understanding different aspects of a research problem .

Table of contents

When to do a case study, step 1: select a case, step 2: build a theoretical framework, step 3: collect your data, step 4: describe and analyze the case, other interesting articles.

A case study is an appropriate research design when you want to gain concrete, contextual, in-depth knowledge about a specific real-world subject. It allows you to explore the key characteristics, meanings, and implications of the case.

Case studies are often a good choice in a thesis or dissertation . They keep your project focused and manageable when you don’t have the time or resources to do large-scale research.

You might use just one complex case study where you explore a single subject in depth, or conduct multiple case studies to compare and illuminate different aspects of your research problem.

Here's why students love Scribbr's proofreading services

Discover proofreading & editing

Once you have developed your problem statement and research questions , you should be ready to choose the specific case that you want to focus on. A good case study should have the potential to:

• Provide new or unexpected insights into the subject
• Challenge or complicate existing assumptions and theories
• Propose practical courses of action to resolve a problem
• Open up new directions for future research

TipIf your research is more practical in nature and aims to simultaneously investigate an issue as you solve it, consider conducting action research instead.

Unlike quantitative or experimental research , a strong case study does not require a random or representative sample. In fact, case studies often deliberately focus on unusual, neglected, or outlying cases which may shed new light on the research problem.

Example of an outlying case studyIn the 1960s the town of Roseto, Pennsylvania was discovered to have extremely low rates of heart disease compared to the US average. It became an important case study for understanding previously neglected causes of heart disease.

However, you can also choose a more common or representative case to exemplify a particular category, experience or phenomenon.

Example of a representative case studyIn the 1920s, two sociologists used Muncie, Indiana as a case study of a typical American city that supposedly exemplified the changing culture of the US at the time.

While case studies focus more on concrete details than general theories, they should usually have some connection with theory in the field. This way the case study is not just an isolated description, but is integrated into existing knowledge about the topic. It might aim to:

• Exemplify a theory by showing how it explains the case under investigation
• Expand on a theory by uncovering new concepts and ideas that need to be incorporated
• Challenge a theory by exploring an outlier case that doesn’t fit with established assumptions

To ensure that your analysis of the case has a solid academic grounding, you should conduct a literature review of sources related to the topic and develop a theoretical framework . This means identifying key concepts and theories to guide your analysis and interpretation.

There are many different research methods you can use to collect data on your subject. Case studies tend to focus on qualitative data using methods such as interviews , observations , and analysis of primary and secondary sources (e.g., newspaper articles, photographs, official records). Sometimes a case study will also collect quantitative data.

Example of a mixed methods case studyFor a case study of a wind farm development in a rural area, you could collect quantitative data on employment rates and business revenue, collect qualitative data on local people’s perceptions and experiences, and analyze local and national media coverage of the development.

The aim is to gain as thorough an understanding as possible of the case and its context.

Receive feedback on language, structure, and formatting

Professional editors proofread and edit your paper by focusing on:

• Academic style
• Vague sentences
• Style consistency

See an example

In writing up the case study, you need to bring together all the relevant aspects to give as complete a picture as possible of the subject.

How you report your findings depends on the type of research you are doing. Some case studies are structured like a standard scientific paper or thesis , with separate sections or chapters for the methods , results and discussion .

Others are written in a more narrative style, aiming to explore the case from various angles and analyze its meanings and implications (for example, by using textual analysis or discourse analysis ).

In all cases, though, make sure to give contextual details about the case, connect it back to the literature and theory, and discuss how it fits into wider patterns or debates.

If you want to know more about statistics , methodology , or research bias , make sure to check out some of our other articles with explanations and examples.

• Normal distribution
• Degrees of freedom
• Null hypothesis
• Discourse analysis
• Control groups
• Mixed methods research
• Non-probability sampling
• Quantitative research
• Ecological validity

Research bias

• Rosenthal effect
• Implicit bias
• Cognitive bias
• Selection bias
• Negativity bias
• Status quo bias

Cite this Scribbr article

If you want to cite this source, you can copy and paste the citation or click the “Cite this Scribbr article” button to automatically add the citation to our free Citation Generator.

McCombes, S. (2023, November 20). What Is a Case Study? | Definition, Examples & Methods. Scribbr. Retrieved September 3, 2024, from https://www.scribbr.com/methodology/case-study/

Is this article helpful?

Shona McCombes

Other students also liked, primary vs. secondary sources | difference & examples, what is a theoretical framework | guide to organizing, what is action research | definition & examples, what is your plagiarism score.

Case Study vs Experiment: Know the Difference

A case study and experiment are the two prominent approaches often used at the forefront of scholarly inquiry. While case studies study the complexities of real-life situations, aiming for depth and contextual understanding, experiments seek to uncover causal relationships through controlled manipulation and observation. Both these research methods are indispensable tools in understanding phenomena, yet they diverge significantly in their approaches, aims, and applications.

In this article, we’ll unpack the key differences between case studies and experiments, exploring their strengths, limitations, and the unique insights they offer when working with quantitative data. In the meantime, feel free to use our specialized case study writing service if you seek to streamline your efforts when handling this academic task.

What Is a Case Study?

A case study is a research method that involves an in-depth examination of a particular individual, group, event, or phenomenon within its real-life context. It aims to provide a detailed and comprehensive analysis of the subject under investigation, often using multiple data sources such as interviews, observations, documents, and archival records.

The case study method is used in psychology, sociology, anthropology, education, and business to explore complex issues, understand unique situations, and generate rich, contextualized insights. They allow scholars to explore the intricacies of real-world phenomena, uncovering patterns, relationships, and underlying factors in social sciences that may not be readily apparent through other research methods.

Overall, case studies offer a holistic and nuanced understanding of the subject of interest, facilitating deeper exploration and interpretation of complex social and human phenomena. If you’re struggling with this assignment, simply say, ‘ write my case study for me ,’ and our experts will help you promptly.

What Is an Experiment?

Compared to the case study method, an experiment investigates cause-and-effect relationships by systematically manipulating one or more variables and observing the effects on other variables. In an experiment, students aim to establish causal relationships between an independent variable (the factors being manipulated) and a dependent variable (the outcomes being measured).

Experiments are characterized by their controlled and systematic approach, often involving the random assignment of participants to different experimental conditions to minimize bias and ensure the validity of the findings. They are commonly used in such fields of social sciences as psychology, biology, physics, and medicine to test hypotheses, identify causal mechanisms, and provide empirical evidence for theories.

An experiment method allows scholars to establish causal relationships with high confidence, providing valuable insights into the underlying mechanisms of behavior, natural phenomena, and social processes. Other research methods include:

• Survey method: Collects research data from individuals through questionnaires or interviews to gather information about attitudes, opinions, behaviors, and characteristics of a population.
• Observation method: Systematically observes and records behavior, events, or phenomena as they naturally occur in real-life settings to study social interactions, environmental factors, or naturalistic behavior.
• Qualitative and quantitative research method: Qualitative research explores meanings, perceptions, and experiences using interviews or content analysis, while quantitative research analyzes numerical data to test hypotheses or identify patterns and relationships.
• Archival research method: Analyzes existing documents, records, or data sources such as historical documents or organizational archives to investigate trends, patterns, or historical events.
• Action research method: Involves collaboration between scholars and practitioners to identify and address practical problems or challenges within specific organizational or community contexts, aiming to generate actionable knowledge and facilitate positive change.

Difference Between Case Study and Experiment

Case study and experiment definitions.

A case study method involves a deep investigation into a specific individual, group, event, or phenomenon within its real-life context, aiming to provide rich and detailed insights into complex issues. Learners gather research data from multiple sources, such as interviews, observations, documents, and archival records, to comprehensively understand the subject under study.

Case studies are particularly useful for exploring unique or rare phenomena, offering a holistic view that captures the intricacies and nuances of the situation. However, findings from case studies may be challenging to generalize to broader populations due to the specificity of the case and the lack of experimental control. To learn more about how to write a case study , please refer to our guide.

An experiment is a research method that systematically manipulates one or more variables to observe their effects on other variables, aiming to establish cause-and-effect relationships under controlled conditions. Researchers design experiments with high control over variables, often using standardized procedures and quantitative measures for research data collection.

Experiments are well-suited for testing hypotheses and identifying causal relationships in controlled environments, allowing educatees to conclude the effects of specific interventions or manipulations. However, experiments may lack the depth and contextual richness of case studies, and findings are typically limited to the specific conditions of the experiment.

Case Study and Experiment Characteristic Features

• In case studies , variables are observed rather than manipulated. Researchers do not typically control variables; instead, they examine how naturally occurring variables interact within the case context.
• Experiments involve manipulating one or more variables to observe their effects on other variables. Students actively control and manipulate variables to test hypotheses and establish cause-and-effect relationships.
• Case studies may not always begin with a specific hypothesis. Instead, researchers often seek to generate hypotheses based on the data collected during the study.
• Experiments are typically conducted to test specific hypotheses. Researchers formulate a hypothesis based on existing theory or observations, and the experiment is designed to confirm or refute this hypothesis.

Manipulating Variables

• Variables are not manipulated in case studies . Instead, researchers observe and analyze how naturally occurring variables influence the phenomenon of interest.
• In experiments , researchers actively manipulate one independent or dependent variable or more to observe their effects on other variables. This manipulation allows researchers to establish causal relationships between variables.
• Case studies often involve collecting qualitative data from multiple sources, such as interviews, observations, documents, and archival records. Researchers analyze this research data to provide a detailed and contextualized understanding of the case.
• Experiments typically involve the collection of quantitative data using standardized procedures and measures. Researchers use statistical analysis to interpret the research data and draw conclusions about the effects of the manipulated variables.

Areas of Implementation

• Case studies are widely used in social sciences, such as psychology, sociology, anthropology, education, and business, to explore complex issues, understand unique situations, and generate rich, contextualized insights.
• Experiments are common in fields such as psychology, biology, physics, and medicine, where researchers seek to test hypotheses, identify causal mechanisms, and provide empirical evidence for theories through controlled manipulation and observation.

Frequently asked questions

She was flawless! first time using a website like this, I've ordered article review and i totally adored it! grammar punctuation, content - everything was on point

This writer is my go to, because whenever I need someone who I can trust my task to - I hire Joy. She wrote almost every paper for me for the last 2 years

Term paper done up to a highest standard, no revisions, perfect communication. 10s across the board!!!!!!!

I send him instructions and that's it. my paper was done 10 hours later, no stupid questions, he nailed it.

Sometimes I wonder if Michael is secretly a professor because he literally knows everything. HE DID SO WELL THAT MY PROF SHOWED MY PAPER AS AN EXAMPLE. unbelievable, many thanks

New posts to your inbox!

Stay in touch

• Skip to secondary menu
• Skip to main content
• Skip to primary sidebar

Statistics By Jim

Making statistics intuitive

Correlational Study Overview & Examples

By Jim Frost 2 Comments

What is a Correlational Study?

A correlational study is an experimental design that evaluates only the correlation between variables. The researchers record measurements but do not control or manipulate the variables. Correlational research is a form of observational study .

A correlation indicates that as the value of one variable increases, the other tends to change in a specific direction:

• Positive correlation : Two variables increase or decrease together (as height increases, weight tends to increase).
• Negative correlation : As one variable increases, the other tends to decrease (as school absences increase, grades tend to fall).
• No correlation : No relationship exists between the two variables. As one increases, the other does not change in a specific direction (as absences increase, height doesn’t tend to increase or decrease).

For example, researchers conducting correlational research explored the relationship between social media usage and levels of anxiety in young adults. Participants reported their demographic information and daily time on various social media platforms and completed a standardized anxiety assessment tool.

The correlational study looked for relationships between social media usage and anxiety. Is increased social media usage associated with higher anxiety? Is it worse for particular demographics?

Learn more about Interpreting Correlation .

Using Correlational Research

Correlational research design is crucial in various disciplines, notably psychology and medicine. This type of design is generally cheaper, easier, and quicker to conduct than an experiment because the researchers don’t control any variables or conditions. Consequently, these studies often serve as an initial assessment, especially when random assignment and controlling variables for a true experiment are not feasible or unethical.

However, an unfortunate aspect of a correlational study is its limitation in establishing causation. While these studies can reveal connections between variables, they cannot prove that altering one variable will cause changes in another. Hence, correlational research can determine whether relationships exist but cannot confirm causality.

Remember, correlation doesn’t necessarily imply causation !

Correlational Study vs Experiment

The difference between the two designs is simple.

In a correlational study, the researchers don’t systematically control any variables. They’re simply observing events and do not want to influence outcomes.

In an experiment, researchers manipulate variables and explicitly hope to affect the outcomes. For example, they might control the treatment condition by giving a medication or placebo to each subject. They also randomly assign subjects to the control and treatment groups, which helps establish causality.

Learn more about Randomized Controlled Trials (RCTs) , which statisticians consider to be true experiments.

Types of Correlation Studies and Examples

Researchers divide these studies into three broad types.

Secondary Data Sources

One approach to correlational research is to utilize pre-existing data, which may include official records, public polls, or data from earlier studies. This method can be cost-effective and time-efficient because other researchers have already gathered the data. These existing data sources can provide large sample sizes and longitudinal data , thereby showing relationship trends.

However, it also comes with potential drawbacks. The data may be incomplete or irrelevant to the new research question. Additionally, as a researcher, you won’t have control over the original data collection methods, potentially impacting the data’s reliability and validity .

Using existing data makes this approach a retrospective study .

Surveys in Correlation Research

Surveys are a great way to collect data for correlational studies while using a consistent instrument across all respondents. You can use various formats, such as in-person, online, and by phone. And you can ask the questions necessary to obtain the particular variables you need for your project. In short, it’s easy to customize surveys to match your study’s requirements.

However, you’ll need to carefully word all the questions to be clear and not introduce bias in the results. This process can take multiple iterations and pilot studies to produce the finished survey.

For example, you can use a survey to find correlations between various demographic variables and political opinions.

Naturalistic Observation

Naturalistic observation is a method of collecting field data for a correlational study. Researchers observe and measure variables in a natural environment. The process can include counting events, categorizing behavior, and describing outcomes without interfering with the activities.

For example, researchers might observe and record children’s behavior after watching television. Does a relationship exist between the type of television program and behaviors?

Naturalistic observations occur in a prospective study .

Analyzing Data from a Correlational Study

Statistical analysis of correlational research frequently involves correlation and regression analysis .

A correlation coefficient describes the strength and direction of the relationship between two variables with a single number.

Regression analysis can evaluate how multiple variables relate to a single outcome. For example, in the social media correlational study example, how do the demographic variables and daily social media usage collectively correlate with anxiety?

Curtis EA, Comiskey C, Dempsey O.  Importance and use of correlational research .  Nurse Researcher . 2016;23(6):20-25. doi:10.7748/nr.2016.e1382

Share this:

Reader Interactions

January 14, 2024 at 4:34 pm

Hi Jim. Have you written a blog note dedicated to clinical trials? If not, besides the note on hypothesis testing, are there other blogs ypo have written that touch on clinical trials?

January 14, 2024 at 5:49 pm

Hi Stan, I haven’t written a blog post specifically about clinical trials, but I have the following related posts:

Randomized Controlled Trials Clinical Trial about a COVID vaccine Clinical Trials about flu vaccines

Comments and Questions Cancel reply

Observational vs. Experimental Study: A Comprehensive Guide

Explore the fundamental disparities between experimental and observational studies in this comprehensive guide by Santos Research Center, Corp. Uncover concepts such as control group, random sample, cohort studies, response variable, and explanatory variable that shape the foundation of these methodologies. Discover the significance of randomized controlled trials and case control studies, examining causal relationships and the role of dependent variables and independent variables in research designs.

This enlightening exploration also delves into the meticulous scientific study process, involving survey members, systematic reviews, and statistical analyses. Investigate the careful balance of control group and treatment group dynamics, highlighting how researchers meticulously assign variables and analyze statistical patterns to discern meaningful insights. From dissecting issues like lung cancer to understanding sleep patterns, this guide emphasizes the precision of controlled experiments and controlled trials, where variables are isolated and scrutinized, paving the way for a deeper comprehension of the world through empirical research.

Introduction to Observational and Experimental Studies

These two studies are the cornerstones of scientific inquiry, each offering a distinct approach to unraveling the mysteries of the natural world.

Observational studies allow us to observe, document, and gather data without direct intervention. They provide a means to explore real-world scenarios and trends, making them valuable when manipulating variables is not feasible or ethical. From surveys to meticulous observations, these studies shed light on existing conditions and relationships.

Experimental studies , in contrast, put researchers in the driver's seat. They involve the deliberate manipulation of variables to understand their impact on specific outcomes. By controlling the conditions, experimental studies establish causal relationships, answering questions of causality with precision. This approach is pivotal for hypothesis testing and informed decision-making.

At Santos Research Center, Corp., we recognize the importance of both observational and experimental studies. We employ these methodologies in our diverse research projects to ensure the highest quality of scientific investigation and to answer a wide range of research questions.

Observational Studies: A Closer Look

In our exploration of research methodologies, let's zoom in on observational research studies—an essential facet of scientific inquiry that we at Santos Research Center, Corp., expertly employ in our diverse research projects.

What is an Observational Study?

Observational research studies involve the passive observation of subjects without any intervention or manipulation by researchers. These studies are designed to scrutinize the relationships between variables and test subjects, uncover patterns, and draw conclusions grounded in real-world data.

Researchers refrain from interfering with the natural course of events in controlled experiment. Instead, they meticulously gather data by keenly observing and documenting information about the test subjects and their surroundings. This approach permits the examination of variables that cannot be ethically or feasibly manipulated, making it particularly valuable in certain research scenarios.

Types of Observational Studies

Now, let's delve into the various forms that observational studies can take, each with its distinct characteristics and applications.

Cohort Studies:  A cohort study is a type of observational study that entails tracking one group of individuals over an extended period. Its primary goal is to identify potential causes or risk factors for specific outcomes or treatment group. Cohort studies provide valuable insights into the development of conditions or diseases and the factors that influence them.

Case-Control Studies:  Case-control studies, on the other hand, involve the comparison of individuals with a particular condition or outcome to those without it (the control group). These studies aim to discern potential causal factors or associations that may have contributed to the development of the condition under investigation.

Cross-Sectional Studies:  Cross-sectional studies take a snapshot of a diverse group of individuals at a single point in time. By collecting data from this snapshot, researchers gain insights into the prevalence of a specific condition or the relationships between variables at that precise moment. Cross-sectional studies are often used to assess the health status of the different groups within a population or explore the interplay between various factors.

Advantages and Limitations of Observational Studies

Observational studies, as we've explored, are a vital pillar of scientific research, offering unique insights into real-world phenomena. In this section, we will dissect the advantages and limitations that characterize these studies, shedding light on the intricacies that researchers grapple with when employing this methodology.

Advantages: One of the paramount advantages of observational studies lies in their utilization of real-world data. Unlike controlled experiments that operate in artificial settings, observational studies embrace the complexities of the natural world. This approach enables researchers to capture genuine behaviors, patterns, and occurrences as they unfold. As a result, the data collected reflects the intricacies of real-life scenarios, making it highly relevant and applicable to diverse settings and populations.

Moreover, in a randomized controlled trial, researchers looked to randomly assign participants to a group. Observational studies excel in their capacity to examine long-term trends. By observing one group of subjects over extended periods, research scientists gain the ability to track developments, trends, and shifts in behavior or outcomes. This longitudinal perspective is invaluable when studying phenomena that evolve gradually, such as chronic diseases, societal changes, or environmental shifts. It allows for the detection of subtle nuances that may be missed in shorter-term investigations.

Limitations: However, like any research methodology, observational studies are not without their limitations. One significant challenge of statistical study lies in the potential for biases. Since researchers do not intervene in the subjects' experiences, various biases can creep into the data collection process. These biases may arise from participant self-reporting, observer bias, or selection bias in random sample, among others. Careful design and rigorous data analysis are crucial for mitigating these biases.

Another limitation is the presence of confounding variables. In observational studies, it can be challenging to isolate the effect of a specific variable from the myriad of other factors at play. These confounding variables can obscure the true relationship between the variables of interest, making it difficult to establish causation definitively. Research scientists must employ statistical techniques to control for or adjust these confounding variables.

Additionally, observational studies face constraints in their ability to establish causation. While they can identify associations and correlations between variables, they cannot prove causality or causal relationship. Establishing causation typically requires controlled experiments where researchers can manipulate independent variables systematically. In observational studies, researchers can only infer potential causation based on the observed associations.

Experimental Studies: Delving Deeper

In the intricate landscape of scientific research, we now turn our gaze toward experimental studies—a dynamic and powerful method that Santos Research Center, Corp. skillfully employs in our pursuit of knowledge.

What is an Experimental Study?

While some studies observe and gather data passively, experimental studies take a more proactive approach. Here, researchers actively introduce an intervention or treatment to an experiment group study its effects on one or more variables. This methodology empowers researchers to manipulate independent variables deliberately and examine their direct impact on dependent variables.

Experimental research are distinguished by their exceptional ability to establish cause-and-effect relationships. This invaluable characteristic allows researchers to unlock the mysteries of how one variable influences another, offering profound insights into the scientific questions at hand. Within the controlled environment of an experimental study, researchers can systematically test hypotheses, shedding light on complex phenomena.

Key Features of Experimental Studies

Central to statistical analysis, the rigor and reliability of experimental studies are several key features that ensure the validity of their findings.

Randomized Controlled Trials:  Randomization is a critical element in experimental studies, as it ensures that subjects are assigned to groups in a random assignment. This randomly assigned allocation minimizes the risk of unintentional biases and confounding variables, strengthening the credibility of the study's outcomes.

Control Groups:  Control groups play a pivotal role in experimental studies by serving as a baseline for comparison. They enable researchers to assess the true impact of the intervention being studied. By comparing the outcomes of the intervention group to those of survey members of the control group, researchers can discern whether the intervention caused the observed changes.

Blinding:  Both single-blind and double-blind techniques are employed in experimental studies to prevent biases from influencing the study or controlled trial's outcomes. Single-blind studies keep either the subjects or the researchers unaware of certain aspects of the study, while double-blind studies extend this blindness to both parties, enhancing the objectivity of the study.

These key features work in concert to uphold the integrity and trustworthiness of the results generated through experimental studies.

Advantages and Limitations of Experimental Studies

As with any research methodology, this one comes with its unique set of advantages and limitations.

Advantages:  These studies offer the distinct advantage of establishing causal relationships between two or more variables together. The controlled environment allows researchers to exert authority over variables, ensuring that changes in the dependent variable can be attributed to the independent variable. This meticulous control results in high-quality, reliable data that can significantly contribute to scientific knowledge.

Limitations:  However, experimental ones are not without their challenges. They may raise ethical concerns, particularly when the interventions involve potential risks to subjects. Additionally, their controlled nature can limit their real-world applicability, as the conditions in experiments may not accurately mirror those in the natural world. Moreover, executing an experimental study in randomized controlled, often demands substantial resources, with other variables including time, funding, and personnel.

Observational vs Experimental: A Side-by-Side Comparison

Having previously examined observational and experimental studies individually, we now embark on a side-by-side comparison to illuminate the key distinctions and commonalities between these foundational research approaches.

Key Differences and Notable Similarities

Methodologies

• Observational Studies : Characterized by passive observation, where researchers collect data without direct intervention, allowing the natural course of events to unfold.
• Experimental Studies : Involve active intervention, where researchers deliberately manipulate variables to discern their impact on specific outcomes, ensuring control over the experimental conditions.
• Observational Studies : Designed to identify patterns, correlations, and associations within existing data, shedding light on relationships within real-world settings.
• Experimental Studies : Geared toward establishing causality by determining the cause-and-effect relationships between variables, often in controlled laboratory environments.
• Observational Studies : Yield real-world data, reflecting the complexities and nuances of natural phenomena.
• Experimental Studies : Generate controlled data, allowing for precise analysis and the establishment of clear causal connections.

Observational studies excel at exploring associations and uncovering patterns within the intricacies of real-world settings, while experimental studies shine as the gold standard for discerning cause-and-effect relationships through meticulous control and manipulation in controlled environments. Understanding these differences and similarities empowers researchers to choose the most appropriate method for their specific research objectives.

When to Use Which: Practical Applications

The decision to employ either observational or experimental studies hinges on the research objectives at hand and the available resources. Observational studies prove invaluable when variable manipulation is impractical or ethically challenging, making them ideal for delving into long-term trends and uncovering intricate associations between certain variables (response variable or explanatory variable). On the other hand, experimental studies emerge as indispensable tools when the aim is to definitively establish causation and methodically control variables.

At Santos Research Center, Corp., our approach to both scientific study and methodology is characterized by meticulous consideration of the specific research goals. We recognize that the quality of outcomes hinges on selecting the most appropriate method of research study. Our unwavering commitment to employing both observational and experimental research studies further underscores our dedication to advancing scientific knowledge across diverse domains.

Conclusion: The Synergy of Experimental and Observational Studies in Research

In conclusion, both observational and experimental studies are integral to scientific research, offering complementary approaches with unique strengths and limitations. At Santos Research Center, Corp., we leverage these methodologies to contribute meaningfully to the scientific community.

Explore our projects and initiatives at Santos Research Center, Corp. by visiting our website or contacting us at (813) 249-9100, where our unwavering commitment to rigorous research practices and advancing scientific knowledge awaits.

Recent Posts

At Santos Research Center, a medical research facility dedicated to advancing TBI treatments, we emphasize the importance of tailored rehabilitation...

Learn about COVID-19 rebound after Paxlovid, its symptoms, causes, and management strategies. Join our study at Santos Research Center. Apply now!

Learn everything about Respiratory Syncytial Virus (RSV), from symptoms and diagnosis to treatment and prevention. Stay informed and protect your health with...

Discover key insights on Alzheimer's disease, including symptoms, stages, and care tips. Learn how to manage the condition and find out how you can...

Discover expert insights on migraines, from symptoms and causes to management strategies, and learn about our specialized support at Santos Research Center.

Explore our in-depth guide on UTIs, covering everything from symptoms and causes to effective treatments, and learn how to manage and prevent urinary tract infections.

Your definitive guide to COVID symptoms. Dive deep into the signs of COVID-19, understand the new variants, and get answers to your most pressing questions.

Santos Research Center, Corp. is a research facility conducting paid clinical trials, in partnership with major pharmaceutical companies & CROs. We work with patients from across the Tampa Bay area.

Contact Details

Navigation menu.

Research Methods

• Research Process
• Research Design & Method

Qualitative vs. Quantiative

Correlational vs. experimental, empirical vs. non-empirical.

• Survey Research
• Survey & Interview Data Analysis
• Resources for Research
• Ethical Considerations in Research

Qualitative Research gathers data about lived experiences, emotions or behaviors, and the meanings individuals attach to them. It assists in enabling researchers to gain a better understanding of complex concepts, social interactions or cultural phenomena. This type of research is useful in the exploration of how or why things have occurred, interpreting events and describing actions.

Quantitative Research gathers numerical data which can be ranked, measured or categorized through statistical analysis. It assists with uncovering patterns or relationships, and for making generalizations. This type of research is useful for finding out how many, how much, how often, or to what extent.

Correlational Research cannot determine causal relationships. Instead they examine relationships between variables.

Experimental Research can establish causal relationship and variables can be manipulated.

Empirical Studies are based on evidence. The data is collected through experimentation or observation.

Non-empirical Studies do not require researchers to collect first-hand data.

• << Previous: Research Design & Method
• Next: Survey Research >>
• Last Updated: Apr 5, 2023 4:19 PM
• URL: https://semo.libguides.com/ResearchMethods

Case Study vs. Single-Case Experimental Designs

What's the difference.

Case study and single-case experimental designs are both research methods used in psychology and other social sciences to investigate individual cases or subjects. However, they differ in their approach and purpose. Case studies involve in-depth examination of a single case, such as an individual, group, or organization, to gain a comprehensive understanding of the phenomenon being studied. On the other hand, single-case experimental designs focus on studying the effects of an intervention or treatment on a single subject over time. These designs use repeated measures and control conditions to establish cause-and-effect relationships. While case studies provide rich qualitative data, single-case experimental designs offer more rigorous experimental control and allow for the evaluation of treatment effectiveness.

Further Detail

Introduction.

When conducting research in various fields, it is essential to choose the appropriate study design to answer research questions effectively. Two commonly used designs are case study and single-case experimental designs. While both approaches aim to provide valuable insights into specific phenomena, they differ in several key attributes. This article will compare and contrast the attributes of case study and single-case experimental designs, highlighting their strengths and limitations.

Definition and Purpose

A case study is an in-depth investigation of a particular individual, group, or event. It involves collecting and analyzing qualitative or quantitative data to gain a comprehensive understanding of the subject under study. Case studies are often used to explore complex phenomena, generate hypotheses, or provide detailed descriptions of unique cases.

On the other hand, single-case experimental designs are a type of research design that focuses on studying a single individual or a small group over time. These designs involve manipulating an independent variable and measuring its effects on a dependent variable. Single-case experimental designs are particularly useful for examining cause-and-effect relationships and evaluating the effectiveness of interventions or treatments.

Data Collection and Analysis

In terms of data collection, case studies rely on various sources such as interviews, observations, documents, and artifacts. Researchers often employ multiple methods to gather rich and diverse data, allowing for a comprehensive analysis of the case. The data collected in case studies are typically qualitative in nature, although quantitative data may also be included.

In contrast, single-case experimental designs primarily rely on quantitative data collection methods. Researchers use standardized measures and instruments to collect data on the dependent variable before, during, and after the manipulation of the independent variable. This allows for a systematic analysis of the effects of the intervention or treatment on the individual or group being studied.

Generalizability

One of the key differences between case studies and single-case experimental designs is their generalizability. Case studies are often conducted on unique or rare cases, making it challenging to generalize the findings to a larger population. The focus of case studies is on providing detailed insights into specific cases rather than making broad generalizations.

On the other hand, single-case experimental designs aim to establish causal relationships and can provide evidence for generalizability. By systematically manipulating the independent variable and measuring its effects on the dependent variable, researchers can draw conclusions about the effectiveness of interventions or treatments that may be applicable to similar cases or populations.

Internal Validity

Internal validity refers to the extent to which a study accurately measures the cause-and-effect relationship between variables. In case studies, establishing internal validity can be challenging due to the lack of control over extraneous variables. The presence of multiple data sources and the potential for subjective interpretation may also introduce bias.

In contrast, single-case experimental designs prioritize internal validity by employing rigorous control over extraneous variables. Researchers carefully design the intervention or treatment, implement it consistently, and measure the dependent variable under controlled conditions. This allows for a more confident determination of the causal relationship between the independent and dependent variables.

Time and Resources

Case studies often require significant time and resources due to their in-depth nature. Researchers need to spend considerable time collecting and analyzing data from various sources, conducting interviews, and immersing themselves in the case. Additionally, case studies may involve multiple researchers or a research team, further increasing the required resources.

On the other hand, single-case experimental designs can be more time and resource-efficient. Since they focus on a single individual or a small group, data collection and analysis can be more streamlined. Researchers can also implement interventions or treatments in a controlled manner, reducing the time and resources needed for data collection.

Ethical Considerations

Both case studies and single-case experimental designs require researchers to consider ethical implications. In case studies, researchers must ensure the privacy and confidentiality of the individuals or groups being studied. Informed consent and ethical guidelines for data collection and analysis should be followed to protect the rights and well-being of the participants.

Similarly, in single-case experimental designs, researchers must consider ethical considerations when implementing interventions or treatments. The well-being and safety of the individual or group being studied should be prioritized, and informed consent should be obtained. Additionally, researchers should carefully monitor and evaluate the potential risks and benefits associated with the intervention or treatment.

Case studies and single-case experimental designs are valuable research approaches that offer unique insights into specific phenomena. While case studies provide in-depth descriptions and exploratory analyses of individual cases, single-case experimental designs focus on establishing causal relationships and evaluating interventions or treatments. Researchers should carefully consider the attributes and goals of their study when choosing between these two designs, ensuring that the selected approach aligns with their research questions and objectives.

Comparisons may contain inaccurate information about people, places, or facts. Please report any issues.

Distinguishing Between Case Studies & Experiments

Case Study vs Experiment

Case studies and experiments are two distinct research methods used across various disciplines, providing researchers with the ability to study and analyze a subject through different approaches. This variety in research methods allows the researcher to gather both qualitative and quantitative data, cross-check the data, and assign greater validity to the conclusions and overall findings of the research. A case study is a research method in which the researcher explores the subject in depth, while an experiment is a research method where two specific groups or variables are used to test a hypothesis. This article will examine the differences between case study and experiment further.

What is a Case Study?

A case study is a research method where an individual, event, or significant place is studied in depth. In the case of an individual, the researcher studies the person’s life history, which can include important days or special experiences. The case study method is used in various social sciences such as sociology, anthropology, and psychology. Through a case study, the researcher can identify and understand the subjective experiences of an individual regarding a specific topic. For example, a researcher studying the impact of second rape on the lives of rape victims can conduct several case studies to understand the subjective experiences of individuals and social mechanisms that contribute to this phenomenon. The case study is a qualitative research method that can be subjective.

What is an Experiment?

An experiment, unlike a case study, can be classified as a quantitative research method, as it provides statistically significant data and an objective, empirical approach. Experiments are primarily used in natural sciences, as they allow the scientist to control variables. In social sciences, controlling variables can be challenging and may lead to faulty conclusions. In an experiment, there are mainly two variables: the independent variable and the dependent variable. The researcher tries to test their hypothesis by manipulating these variables. There are different types of experiments, such as laboratory experiments (conducted in laboratories where conditions can be strictly controlled) and natural experiments (which take place in real-life settings). As seen, case study methods and experiments are very different from one another. However, most researchers prefer to use triangulation when conducting research to minimize biases.

Key Takeaways

• Case studies are in-depth explorations of a subject, providing qualitative data, while experiments test hypotheses by manipulating variables, providing quantitative data.
• Experiments are primarily used in natural sciences, whereas case studies are primarily used in social sciences.
• Experiments involve testing the correlation between two variables (independent and dependent), while case studies focus on exploring a subject in depth without testing correlations between variables.

LEAVE A REPLY Cancel reply

Save my name, email, and website in this browser for the next time I comment.

Related Articles

Difference between power & authority, distinguishing could of & could have, distinguishing pixie & bob haircuts, distinguishing between debate & discussion, distinguishing between dialogue & conversation, distinguishing between a present & a gift, distinguishing between will & can, distinguishing between up & upon.

• Skip to main content
• Skip to primary sidebar
• Skip to footer
• QuestionPro

• Solutions Industries Gaming Automotive Sports and events Education Government Travel & Hospitality Financial Services Healthcare Cannabis Technology Use Case AskWhy Communities Audience Contactless surveys Mobile LivePolls Member Experience GDPR Positive People Science 360 Feedback Surveys
• Resources Blog eBooks Survey Templates Case Studies Training Help center

Home Market Research

Descriptive Correlational: Descriptive vs Correlational Research

Descriptive research and Correlational research are two important types of research studies that help researchers make ambitious and measured decisions in their respective fields. Both descriptive research and correlational research are used in descriptive correlational research. 

Descriptive research is defined as a research method that involves observing behavior to describe attributes objectively and systematically. A descriptive research project seeks to comprehend phenomena or groups in depth.

Correlational research , on the other hand, is a method that describes and predicts how variables are naturally related in the real world without the researcher attempting to alter them or assign causation between them.

The main objective of descriptive research is to create a snapshot of the current state of affairs, whereas correlational research helps in comparing two or more entities or variables.

What is descriptive correlational research?

Descriptive correlational research is a type of research design that tries to explain the relationship between two or more variables without making any claims about cause and effect. It includes collecting and analyzing data on at least two variables to see if there is a link between them. 

In descriptive correlational research, researchers collect data to explain the variables of interest and figure out how they relate. The main goal is to give a full account of the variables and how they are related without changing them or assuming that one thing causes another.

In descriptive correlational research, researchers do not change any variables or try to find cause-and-effect connections. Instead, they just watch and measure the variables of interest and then look at the patterns and relationships that emerge from the data.

Experimental research involves the independent variable to see how it affects the dependent variable, while descriptive correlational research just describes the relationship between variables. 

In descriptive correlational research, correlational research designs measure the magnitude and direction of the relationship between two or more variables, revealing their associations. At the outset creating initial equivalence between the groups or variables being compared is essential in descriptive correlational research

The independent variable occurs prior to the measurement of the measured dependent variable in descriptive correlational research. Its goal is to explain the traits or actions of a certain population or group and look at the connections between independent and dependent variables.

How are descriptive research and correlational research carried out?

Descriptive research is carried out using three methods, namely:  

• Case studies – Case studies involve in-depth research and study of individuals or groups. Case studies lead to a hypothesis and widen a further scope of studying a phenomenon. However, case studies should not be used to determine cause and effect as they don’t have the capacity to make accurate predictions.
• Surveys – A survey is a set of questions that is administered to a population, also known as respondents. Surveys are a popular market research tool that helps collect meaningful insights from the respondents. To gather good quality data, a survey should have good survey questions, which should be a balanced mix of open-ended and close-ended questions .
• Naturalistic Observation – Naturalistic observations are carried out in the natural environment without disturbing the person/ object in observation. It is much like taking notes about people in a supermarket without letting them know. This leads to a greater validity of collected data because people are unaware they are being observed here. This tends to bring out their natural characteristics.

Correlational research also uses naturalistic observation to collect data. However, in addition, it uses archival data to gather information. Archival data is collected from previously conducted research of a similar nature. Archival data is collected through primary research.

In contrast to naturalistic observation, information collected through archived is straightforward. For example, counting the number of people named Jacinda in the United States using their social security number.  

Descriptive Research vs Correlational Research

Features of Descriptive Correlational Research

The key features of descriptive correlational research include the following:

01. Description

The main goal, just like with descriptive research, is to describe the variables of interest thoroughly. Researchers aim to explain a certain group or event’s traits, behaviors, or attitudes. 

02. Relationships

Like correlational research, descriptive correlational research looks at how two or more factors are related. It looks at how variables are connected to each other, such as how they change over time or how they are linked.

03. Quantitative analysis

Most methods for analyzing quantitative analysis data are used in descriptive correlational research. Researchers use statistical methods to study and measure the size and direction of relationships between variables.

04. No manipulation

As with correlational research, the researcher does not change or control the variables. The data is taken in its natural environment without any changes or interference.

05. Cross-sectional or longitudinal

Cross-sectional or longitudinal designs can be used for descriptive correlational research. It collects data at one point in time, while longitudinal research collects data over a long period of time to look at changes and relationships over time. 

Examples of descriptive correlational research

For example, descriptive correlational research could look at the link between a person’s age and how much money they make. The researcher would take a sample of people’s ages and incomes and then look at the data to see if there is a link between the two factors.

• Example 1 : A research project is done to find out if there is a link between how long college students sleep and how well they do in school. They keep track of how many hours kids sleep each night and what their GPAs are. By studying the data, the researcher can describe how the students sleep and find out if there is a link between how long they sleep and how well they do in school.
• Example 2 : A researcher wants to know how people’s exercise habits affect their physical health if they are between the ages of 40 and 60. They take notes on things like how often and how hard you work out, your body mass index (BMI), blood pressure, and cholesterol numbers. By analyzing the data, the researcher can describe the participants’ exercise habits and physical health and look for any links between these factors.
• Example 3 : Let’s say a researcher wants to find out if college students who work out feel less stressed. Using a poll, the researcher finds out how many hours students spend exercising each week and how stressed they feel. By looking at the data, the researcher may find that there is a moderate negative correlation between exercise and stress levels. This means that as exercise grows, stress levels tend to go down. 

Descriptive correlational research is a good way to learn about the characteristics of a population or group and the relationships between its different parts. It lets researchers describe variables in detail and look into their relationships without suggesting that one variable caused another. 

Descriptive correlational research gives useful insights and can be used as a starting point for more research or to come up with hypotheses. It’s important to be aware of the problems with this type of study, such as the fact that it can’t show cause and effect and relies on cross-sectional data. 

Still, descriptive correlational research helps us understand things and makes making decisions in many areas easier.

QuestionPro is a very useful tool for descriptive correlational research. Its many features and easy-to-use interface help researchers collect and study data quickly, giving them a better understanding of the characteristics and relationships between variables in a certain population or group. 

The different kinds of questions, analytical research tools, and reporting features on the software improve the research process and help researchers come up with useful results. QuestionPro makes it easier to do descriptive correlational research, which makes it a useful tool for learning important things and making decisions in many fields.

LEARN MORE         FREE TRIAL

MORE LIKE THIS

Net Trust Score: Tool for Measuring Trust in Organization

Sep 2, 2024

Why You Should Attend XDAY 2024

Aug 30, 2024

Alchemer vs Qualtrics: Find out which one you should choose

Target Population: What It Is + Strategies for Targeting

Aug 29, 2024

Other categories

• Academic Research
• Artificial Intelligence
• Assessments
• Brand Awareness
• Case Studies
• Communities
• Consumer Insights
• Customer effort score
• Customer Engagement
• Customer Experience
• Customer Loyalty
• Customer Research
• Customer Satisfaction
• Employee Benefits
• Employee Engagement
• Employee Retention
• Friday Five
• General Data Protection Regulation
• Insights Hub
• Life@QuestionPro
• Market Research
• Mobile diaries
• Mobile Surveys
• New Features
• Online Communities
• Question Types
• Questionnaire
• QuestionPro Products
• Release Notes
• Research Tools and Apps
• Revenue at Risk
• Survey Templates
• Training Tips
• Tuesday CX Thoughts (TCXT)
• Uncategorized
• What’s Coming Up
• Workforce Intelligence

Warning: The NCBI web site requires JavaScript to function. more...

An official website of the United States government

The .gov means it's official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings
• Browse Titles

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Lau F, Kuziemsky C, editors. Handbook of eHealth Evaluation: An Evidence-based Approach [Internet]. Victoria (BC): University of Victoria; 2017 Feb 27.

Handbook of eHealth Evaluation: An Evidence-based Approach [Internet].

Chapter 12 methods for correlational studies.

Francis Lau .

12.1. Introduction

Correlational studies aim to find out if there are differences in the characteristics of a population depending on whether or not its subjects have been exposed to an event of interest in the naturalistic setting. In eHealth, correlational studies are often used to determine whether the use of an eHealth system is associated with a particular set of user characteristics and/or quality of care patterns ( Friedman & Wyatt, 2006 ). An example is a computerized provider order entry ( cpoe ) study to differentiate the background, usage and performance between clinical users and non-users of the cpoe system after its implementation in a hospital.

Correlational studies are different from comparative studies in that the evaluator does not control the allocation of subjects into comparison groups or assignment of the intervention to specific groups. Instead, the evaluator defines a set of variables including an outcome of interest then tests for hypothesized relations among these variables. The outcome is known as the dependent variable and the variables being tested for association are the independent variables. Correlational studies are similar to comparative studies in that they take on an objectivist view where the variables can be defined, measured and analyzed for the presence of hypothesized relations. As such, correlational studies face the same challenges as comparative studies in terms of their internal and external validity. Of particular importance are the issues of design choices, selection bias, confounders, and reporting consistency.

In this chapter we describe the basic types of correlational studies seen in the eHealth literature and their methodological considerations. Also included are three case examples to show how these studies are done.

12.2. Types of Correlational Studies

Correlational studies, better known as observational studies in epidemiology, are used to examine event exposure, disease prevalence and risk factors in a population ( Elwood, 2007 ). In eHealth, the exposure typically refers to the use of an eHealth system by a population of subjects in a given setting. These subjects may be patients, providers or organizations identified through a set of variables that are thought to differ in their measured values depending on whether or not the subjects were “exposed” to the eHealth system.

There are three basic types of correlational studies that are used in eHealth evaluation: cohort, cross-sectional, and case-control studies ( Vandenbroucke et al., 2014 ). These are described below.

• Cohort studies – A sample of subjects is observed over time where those exposed and not exposed to the eHealth system are compared for differences in one or more predefined outcomes, such as adverse event rates. Cohort studies may be prospective in nature where subjects are followed for a time period into the future or retrospective for a period into the past. The comparisons are typically made at the beginning of the study as baseline measures, then repeated over time at predefined intervals for differences and trends. Some cohort studies involve only a single group of subjects. Their focus is to describe the characteristics of subjects based on a set of variables, such as the pattern of ehr use by providers and their quality of care in an organization over a given time period.
• Cross-sectional studies – These are considered a type of cohort study where only one comparison is made between exposed and unexposed subjects. They provide a snapshot of the outcome and the associated characteristics of the cohort at a specific point in time.
• Case-control studies – Subjects in a sample that are exposed to the eHealth system are matched with those not exposed but otherwise similar in composition, then compared for differences in some predefined outcomes. Case-control studies are retrospective in nature where subjects already exposed to the event are selected then matched with unexposed subjects, using historical cases to ensure they have similar characteristics.

A cross-sectional survey is a type of cross-sectional study where the data source is drawn from postal questionnaires and interviews. This topic will be covered in the chapter on methods for survey studies.

12.3. Methodological Considerations

While correlational studies are considered less rigorous than rct s, they are the preferred designs when it is neither feasible nor ethical to conduct experimental trials. Key methodological issues arise in terms of: (a) design options, (b) biases and confounders, (c) controlling for confounding effects, (d) adherence to good practices, and (e) reporting consistency. These issues are discussed below.

12.3.1. Design Options

There are growing populations with multiple chronic conditions and healthcare interventions. They have made it difficult to design rct s with sufficient sample size and long-term follow-up to account for all the variability this phenomenon entails. Also rct s are intended to test the efficacy of an intervention in a restricted sample of subjects under ideal settings. They have limited generalizability to the population at large in routine settings ( Fleurence, Naci, & Jansen, 2010 ). As such, correlational studies, especially those involving the use of routinely collected ehr data from the general population, have become viable alternatives to rct s. There are advantages and disadvantages to each of the three design options presented above. They are listed below.

• Cohort studies – These studies typically follow the cohorts over time, which allow one to examine causal relationships between exposure and one or more outcomes. They also allow one to measure change in exposure and outcomes over time. However, these studies can be costly and time-consuming to conduct if the outcomes are rare or occur in the future. With prospective cohorts they can be prone to dropout. With retrospective cohorts accurate historical records are required which may not be available or complete ( Levin, 2003a ).
• Case-control studies – These studies are suited to examine infrequent or rare outcomes since they are selected at the outset to ensure sufficient cases. Yet the selection of exposed and matching cases can be problematic, as not all relevant characteristics are known. Moreover, the cases may not be representative of the population of interest. The focus on exposed cases that occur infrequently may overestimate their risks ( Levin, 2003b ).
• Cross-sectional studies – These studies are easier and quicker to conduct than others as they involve a one-time effort over a short period using a sample from the population of interest. They can be used to generate hypotheses and examine multiple outcomes and characteristics at the same time with no loss to follow-up. On the other hand, these studies only give a snapshot of the situation at one time point, making it difficult for causal inference of the exposure and outcomes. The results might be different had another time period been chosen ( Levin, 2006 ).

12.3.2. Biases and Confounders

Shamliyan, Kane, and Dickinson (2010) conducted a systematic review on tools used to assess the quality of observational studies. Despite the large number of quality scales and checklists found in the literature, they concluded that the universal concerns are in the areas of selection bias, confounding, and misclassification. These concerns, also mentioned by Vandenbroucke and colleagues (2014) in their reporting guidelines for observational studies, are summarized below.

• Selection bias – When subjects are selected through their exposure to the event rather than by random or concealed allocation, there is a risk that the subjects are not comparable due to the presence of systematic differences in their baseline characteristics. For example, a correlational study that examines the association between ehr use and quality of care may have younger providers with more computer savvy in the exposed group because they use ehr more and with more facility than those in the unexposed group. It is also possible to have sicker patients in the exposed group since they require more frequent ehr use than unexposed patients who may be healthier and have less need for the ehr . This is sometimes referred to as response bias, where the characteristics of subjects agreed to be in the study are different from those who declined to take part.
• Confounding – Extraneous factors that influence the outcome but are also associated with the exposure are said to have a confounding effect. One such type is confounding by indication where sicker patients are both more likely to receive treatments and also more likely to have adverse outcomes. For example, a study of cds alerts and adverse drug events may find a positive but spurious association due to the inclusion of sicker patients with multiple conditions and medications, which increases their chance of adverse events regardless of cds alerts.
• Misclassification – When there are systematic differences in the completeness or accuracy of the data recorded on the subjects, there is a risk of misclassification in their exposures or outcomes. This is also known as information or detection bias. An example is where sicker patients may have more complete ehr data because they received more tests, treatments and outcome tracking than those who are healthier and require less attention. As such, the exposure and outcomes of sicker patients may be overestimated.

It is important to note that bias and confounding are not synonymous. Bias is caused by finding the wrong association from flawed information or subject selection. Confounding is factually correct with respect to the relationship found, but is incorrect in its interpretation due to an extraneous factor that is associated with both the exposure and outcome.

12.3.3. Controlling for Confounding Effects

There are three common methods to control for confounding effects. These are by matching, stratification, and modelling. They are described below ( Higgins & Green, 2011 ).

• Matching – The selection of subjects with similar characteristics so that they are comparable; the matching can be done at the individual subject level where each exposed subject is matched with one or more unexposed subjects as controls. It can also be done at the group level with equal numbers of exposed and unexposed subjects. Another way to match subjects is by propensity score, that is, a measure derived from a set of characteristics in the subjects. An example is the retrospective cohort study by Zhou, Leith, Li, and Tom (2015) to examine the association between caregiver phr use and healthcare utilization by pediatric patients. In that study, a propensity score-matching algorithm was used to match phr -registered children to non-registered children. The matching model used registration as the outcome variable and all child and caregiver characteristics as the independent variables.
• Stratification – Subjects are categorized into subgroups based on a set of characteristics such as age and sex then analyzed for the effect within each subgroup. An example is the retrospective cohort study by Staes et al. (2008) , examining the impact of computerized alerts on the quality of outpatient lab monitoring for transplant patients. In that study, the before/after comparison of the timeliness of reporting and clinician responses was stratified by the type of test (creatinine, cyclosporine A, and tacrolimus) and report source (hospital laboratory or other labs).
• Modelling – The use of statistical models to compute adjusted effects while accounting for relevant characteristics such as age and sex differences among subjects. An example is the retrospective cohort study by Beck and colleagues (2012) to compare documentation consistency and care plan improvement before and after the implementation of an electronic asthma-specific history and physical template. In that study, before/after group characteristics were compared for differences using t -tests for continuous variables and χ 2 statistics for categorical variables. Logistic regression was used to adjust for group differences in age, gender, insurance, albuterol use at admission, and previous hospitalization.

12.3.4. Adherence to Good Practices in Prospective Observational Studies

The ispor Good Research Practices Task Force published a set of recommendations in designing, conducting and reporting prospective observational studies for comparative effectiveness research ( Berger et al., 2012 ) that are relevant to eHealth evaluation. Their key recommendations are listed below.

• Key policy questions should be defined to allow inferences to be drawn.
• Hypothesis testing protocol design to include the hypothesis/questions, treatment groups and outcomes, measured and unmeasured confounders, primary analyses, and required sample size.
• Rationale for prospective observational study design over others (e.g., rct ) is based on question, feasibility, intervention characteristics and ability to answer the question versus cost and timeliness.
• Study design choice is able to address potential biases and confounders through the use of inception cohorts, multiple comparator groups, matching designs and unaffected outcomes.
• Explanation of study design and analytic choices is transparent.
• Study execution is carried out in ways that ensure relevance and reasonable follow-up is not different from the usual practice.
• Study registration takes place on publicly available sites prior to its initiation.

12.3.5. The Need for Reporting Consistency

Vandenbroucke et al. (2014) published an expanded version of the Strengthening the Reporting of Observational Studies in Epidemiology ( strobe ) statement to improve the reporting of observational studies that can be applied in eHealth evaluation. It is made up of 22 items, of which 18 are common to cohort, case-control and cross-sectional studies, with four being specific to each of the three designs. The 22 reporting items are listed below (for details refer to the cited reference).

• Title and abstract – one item that covers the type of design used, and a summary of what was done and found.
• Introduction – two items on study background/rationale, objectives and/or hypotheses.
• Methods – nine items on design, setting, participants, variables, data sources/measurement, bias, study size, quantitative variables and statistical methods used.
• Results – five items on participants, descriptive, outcome data, main results and other analyses.
• Discussion – four items on key results, limitations, interpretation and generalizability.
• Other information – one item on funding source.

The four items specific to study design relate to the reporting of participants, statistical methods, descriptive results and outcome data. They are briefly described below for the three types of designs.

• Cohort studies – Participant eligibility criteria and sources, methods of selection, follow-up and handling dropouts, description of follow-up time and duration, and number of outcome events or summary measures over time. For matched studies include matching criteria and number of exposed and unexposed subjects.
• Cross-sectional studies – Participant eligibility criteria, sources and methods of selection, analytical methods accounting for sampling strategy as needed, and number of outcome events or summary measures.
• Case-control studies – Participant eligibility criteria, sources and methods of case/control selection with rationale for choices, methods of matching cases/controls, and number of exposures by category or summary measures of exposures. For matched studies include matching criteria and number of controls per case.

12.4. Case Examples

12.4.1. cohort study of automated immunosuppressive care.

Park and colleagues (2010) conducted a retrospective cohort study to examine the association between the use of a cds (clinical decision support) system in post-liver transplant immunosuppressive care and the rates of rejection episode and drug toxicity. The study is summarized below.

• Setting – A liver transplant program in the United States that had implemented an automated cds system to manage immunosuppressive therapy for its post-liver transplant recipients after discharge. The system consolidated all clinical information to expedite immunosuppressive review, ordering, and follow-up with recipients. Prior to automation, a paper charting system was used that involved manually tracking lab tests, transcribing results into a paper spreadsheet, finding physicians to review results and orders, and contacting recipients to notify them of changes.
• Subjects – The study population included recipients of liver transplants between 2004 and 2008 who received outpatient immunosuppressive therapy that included tacrolimus medications.
• Design – A retrospective cohort study with a before/after design to compare recipients managed by the paper charting system against those managed by the cds system for up to one year after discharge.
• Measures – The outcome variables were the percentages of recipients with at least one rejection and/or tacrolimus toxicity episode during the one-year follow-up period. The independent variables included recipient, intraoperative, donor and postoperative characteristics, and use of paper charting or cds . Examples of recipient variables were age, gender, body mass index, presence of diabetes and hypertension, and pre-transplant lab results. Examples of intraoperative data were blood type match, type of transplant and volume of blood transfused. Examples of donor data included percentage of fat in the liver. Examples of post-transplantation data included the type of immunosuppressive induction therapy and the management method.
• Analysis – Mean, standard deviation and t -tests were computed for continuous variables after checking for normal distribution. Percentages and Fisher’s exact test were computed for categorical variables. Autoregressive integrated moving average analysis was done to determine change in outcomes over time. Logistic regression with variables thought to be clinically relevant was used to identify significant univariable and multivariable factors associated with the outcomes. P values of less than 0.05 were considered significant.
• Findings – Overall, the cds system was associated with significantly fewer episodes of rejection and tacrolimus toxicity. The integrated moving average analysis showed a significant decrease in outcome rates after the cds system was implemented compared with paper charting. Multivariable analysis showed the cds system had lower odds of a rejection episode than paper charting ( or 0.20; p < 0.01) and lower odds of tacrolimus toxicity ( or 0.5; p < 0.01). Other significant non-system related factors included the use of specific drugs, the percentage of fat in the donor liver and the volume of packed red cells transfused.

12.4.2. Cross-sectional Analysis of EHR Documentation and Care Quality

Linder, Schnipper, and Middleton (2012) conducted a cross-sectional study to examine the association between the type of ehr documentation used by physicians and the quality of care provided. The study is summarized below.

• Setting – An integrated primary care practice-based research network affiliated with an academic centre in the United States. The network uses an in-house ehr system with decision support for preventive services, chronic care management, and medication monitoring and alerts. The ehr data include problem and medication lists, coded allergies and lab tests.
• Subjects – Physicians and patients from 10 primary care practices that were part of an rct to examine the use of a decision support tool to manage patients with coronary artery disease and diabetes ( cad/DM ). Eligible patients were those with cad/DM in their ehr problem list prior to the rct start date.
• Design – A nine-month retrospective cross-sectional analysis of ehr data collected from the rct . Three physician documentation styles were defined based on 188,554 visit notes in the ehr : (a) dictation, (b) structured documentation, and (c) free text note. Physicians were divided into three groups based on their predominant style defined as more than 25% of their notes composed by a given method.
• Measures – The outcome variables were 15 ehr -based cad/DM quality measures assessed 30 days after primary care visits. They covered quality of documentation, medication use, lab testing, physiologic measures, and vaccinations. Measures collected prior to the day of visit were eligible and considered fulfilled with the presence of coded ehr data on vital signs, medications, allergies, problem lists, lab tests, and vaccinations. Independent variables on physicians and patients were included as covariates. For physicians, they included age, gender, training level, proportion of cad/DM patients in their panel, total patient visits, and self-reported experience with the ehr . For patients, they included socio-demographic factors, the number of clinic visits and hospitalizations, the number of problems and medications in the ehr , and whether their physician was in the intervention group.
• Analysis – Baseline characteristics of physicians and patients were compared using descriptive statistics. Continuous variables were compared using anova . For categorical variables, Fisher’s exact test was used for physician variables and χ 2 test for patient variables. Multivariate logistic regression models were used for each quality measure to adjust for patient and physician clustering and potential confounders. Bonferroni procedure was used to account for multiple comparisons for the 15 quality measures.
• Findings – During the study period, 234 physicians documented 18,569 visits from 7,000 cad/DM patients. Of these physicians, 146 (62%) typed free-text notes, 68 (25%) used structured documentation, and 20 (9%) dictated notes. After adjusting for cluster effect, physicians who dictated their notes had the worst quality of care in all 15 measures. In particular, physicians who dictated notes were significantly worse in three of 15 measures (antiplatelet medication, tobacco use, diabetic eye exam); physicians who used structured documentation were better in three measures (blood pressure, body mass, diabetic foot exam); and those who used free-text were better in one measure (influenza vaccination). In summary, physicians who dictated notes had worse quality of care than those with structured documentation.

12.4.3. Case-control Comparison of Internet Portal Use

Nielsen, Halamka, and Kinkel (2012) conducted a case-control study to evaluate whether there was an association between active Internet patient portal use by Multiple Sclerosis ( ms ) patients and medical resource utilization. Patient predictors and barriers to portal use were also identified. The study is summarized below.

• Setting – An academic ms centre in the United States with an in-house Internet patient portal site that was accessed by ms patients to schedule clinic appointments, request prescription refills and referrals, view test results, upload personal health information, and communicate with providers via secure e-mails.
• Subjects – 240 adult ms patients actively followed during 2008 and 2009 were randomly selected from the ehr ; 120 of these patients had submitted at least one message during that period and were defined as portal users. Another 120 patients who did not enrol in the portal or send any message were selected as non-users for comparison.
• Design – A retrospective case-control study facilitated through a chart review comparing portal users against non-users from the same period. Patient demographic and clinical information was extracted from the ehr , while portal usage, including feature access type and frequency and e-mail message content, were provided by it staff.
• Measures – Patient variables included age, gender, race, insurance type, employment status, number of medical problems, disease duration, psychiatric history, number of medications, and physical disability scores. Provider variables included prescription type and frequency. Portal usage variables included feature access type and frequency for test results, appointments, prescription requests and logins, and categorized messaging contents.
• Analysis – Comparison of patient demographic, clinical and medical resource utilization data from users and non-users were made using descriptive statistics, Wilcoxon rank sum test, Fisher’s exact test and χ 2 test. Multivariate logistic regression was used to identify patient predictors and barriers to portal use. Provider prescribing habits against patient’s psychiatric history and portal use were examined by two-way analysis of variance. All statistical tests used p value of 0.05 with no adjustment made for multiple comparisons. A logistic multivariate regression model was created to predict portal use based on patient demographics, clinical condition, socio-economic status, and physical disability metrics.
• Findings – Portal users were mostly young professionals with little physical disability. The most frequently used feature was secure patient-provider messaging, often for medication requests or refills, and self-reported side effects. Predictors and barriers of portal use were the number of medications prescribed ( or 1.69, p < 0.0001), Caucasian ethnicity ( or 5.04, p = 0.007), arm and hand disability ( or 0.23, p = 0.01), and impaired vision ( or 0.31, p = 0.01). For medical resource utilization, portal users had more frequent clinic visits, medication use and prescriptions from centre staff providers. Patients with a history of psychiatric disease were prescribed more ms medications than those without any history ( p < 0.0001). In summary, ms patients used the Internet more than the general population, but physical disability limited their access and need to be addressed.

12.4.4. Limitations

A general limitation of a correlational study is that it can determine association between exposure and outcomes but cannot predict causation. The more specific limitations of the three case examples cited by the authors are listed below.

• Automated immunosuppressive care – Baseline differences existed between groups with unknown effects; possible other unmeasured confounders; possible Hawthorne effects from focus on immunosuppressive care.
• ehr documentation and care quality – Small sample size; only three documentation styles were considered (e.g., scribe and voice recognition software were excluded) and unsure if they were stable during study period; quality measures specific to cad/DM conditions only; complex methods of adjusting for clustering and confounding that did not account for unmeasured confounders; the level of physician training (e.g., attending versus residents) not adjusted.
• Internet portal use – Small sample size not representative of the study population; referral centre site could over-represent complex patients requiring advanced care; all patients had health insurance.

12.5. Summary

In this chapter we described cohort, case-control and cross-sectional studies as three types of correlational studies used in eHealth evaluation. The methodological issues addressed include bias and confounding, controlling for confounders, adherence to good practices and consistency in reporting. Three case examples were included to show how eHealth correlational studies are done.

1 ISPOR – International Society for Pharmacoeconomics and Outcomes Research

• Beck A. F., Sauers H. S., Kahn R. S., Yau C., Weiser J., Simmons J.M. Improved documentation and care planning with an asthma-specific history and physical. Hospital Pediatrics. 2012; 2 (4):194–201. [ PubMed : 24313025 ]
• Berger M. L., Dreyer N., Anderson F., Towse A., Sedrakyan A., Normand S.L. Prospective observational studies to address comparative effectiveness: The ispor good research practices task force report. Value in Health. 2012; 15 (2):217–230. Retrieved from http://www ​.sciencedirect ​.com/science/article ​/pii/S1098301512000071 . [ PubMed : 22433752 ]
• Elwood M. Critical appraisal of epidemiological studies and clinical studies. 3rd ed. Oxford: Oxford University Press; 2007.
• Fleurence R. L., Naci H., Jansen J.P. The critical role of observational evidence in comparative effectiveness research. Health Affairs. 2010; 29 (10):1826–1833. [ PubMed : 20921482 ]
• Friedman C. P., Wyatt J.C. Evaluation methods in biomedical informatics. 2nd ed. New York: Springer Science + Business Media, Inc; 2006.
• Higgins J. P. T., Green S., editors. Cochrane handbook for systematic reviews of interventions. London: The Cochrane Collaboration; 2011. (Version 5.1.0, updated March 2011) Retrieved from http://handbook ​.cochrane.org/
• Levin K. A. Study design iv : Cohort studies. Evidence-based Dentistry. 2003a; 7 :51–52. [ PubMed : 16858385 ]
• Levin K. A. Study design v : Case-control studies. Evidence-based Dentistry. 2003b; 7 :83–84. [ PubMed : 17003803 ]
• Levin K. A. Study design iii : Cross-sectional studies. Evidence-based Dentistry. 2006; 7 :24–25. [ PubMed : 16557257 ]
• Linder J. A., Schnipper J. L., Middleton B. Method of electronic health record documentation and quality of primary care. Journal of the American Medical Informatics Association. 2012; 19 (6):1019–1024. [ PMC free article : PMC3534457 ] [ PubMed : 22610494 ]
• Nielsen A. S., Halamka J. D., Kinkel R.P. Internet portal use in an academic multiple sclerosis center. Journal of the American Medical Informatics Association. 2012; 19 (1):128–133. [ PMC free article : PMC3240754 ] [ PubMed : 21571744 ]
• Park E. S., Peccoud M. R., Wicks K. A., Halldorson J. B., Carithers R. L. Jr., Reyes J. D., Perkins J.D. Use of an automated clinical management system improves outpatient immunosuppressive care following liver transplantation. Journal of the American Medical Informatics Association. 2010; 17 (4):396–402. [ PMC free article : PMC2995663 ] [ PubMed : 20595306 ]
• Shamliyan T., Kane R. L., Dickinson S. A systematic review of tools used to assess the quality of observational studies that examine incidence or prevalence and risk factors for diseases. Journal of Clinical Epidemiology. 2010; 63 (10):1061–1070. [ PubMed : 20728045 ]
• Staes C. J., Evans R. S., Rocha B. H. S. C., Sorensen J. B., Huff S. M., Arata J., Narus S.P. Computerized alerts improve outpatient laboratory monitoring of transplant patients. Journal of the American Medical Informatics Association. 2008; 15 (3):324–332. [ PMC free article : PMC2410008 ] [ PubMed : 18308982 ]
• Vandenbroucke J. P., von Elm E., Altman D. G., Gotzsche P. C., Mulrow C. D., Pocock S. J., Egger M. for the strobe Initiative. Strengthening the reporting of observational studies in epidemiology ( strobe ): explanation and elaboration. International Journal of Surgery. 2014; 12 (12):1500–1524. Retrieved from http://www.sciencedirect.com/science/article/pii/ s174391911400212x . [ PubMed : 25046751 ]
• Zhou Y. Y., Leith W. M., Li H., Tom J.O. Personal health record use for children and health care utilization: propensity score-matched cohort analysis. Journal of the American Medical Informatics Association. 2015; 22 (4):748–754. [ PubMed : 25656517 ]

This publication is licensed under a Creative Commons License, Attribution-Noncommercial 4.0 International License (CC BY-NC 4.0): see https://creativecommons.org/licenses/by-nc/4.0/

• Cite this Page Lau F. Chapter 12 Methods for Correlational Studies. In: Lau F, Kuziemsky C, editors. Handbook of eHealth Evaluation: An Evidence-based Approach [Internet]. Victoria (BC): University of Victoria; 2017 Feb 27.
• PDF version of this title (4.5M)

In this Page

• Introduction
• Types of Correlational Studies
• Methodological Considerations
• Case Examples

Related information

• PMC PubMed Central citations
• PubMed Links to PubMed

Recent Activity

• Chapter 12 Methods for Correlational Studies - Handbook of eHealth Evaluation: A... Chapter 12 Methods for Correlational Studies - Handbook of eHealth Evaluation: An Evidence-based Approach

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

Connect with NLM

National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894

Web Policies FOIA HHS Vulnerability Disclosure

Help Accessibility Careers

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 02 September 2024

Green spaces provide substantial but unequal urban cooling globally

• Yuxiang Li 1 ,
• Jens-Christian Svenning   ORCID: orcid.org/0000-0002-3415-0862 2 ,
• Weiqi Zhou   ORCID: orcid.org/0000-0001-7323-4906 3 , 4 , 5 ,
• Kai Zhu   ORCID: orcid.org/0000-0003-1587-3317 6 ,
• Jesse F. Abrams   ORCID: orcid.org/0000-0003-0411-8519 7 ,
• Timothy M. Lenton   ORCID: orcid.org/0000-0002-6725-7498 7 ,
• William J. Ripple 8 ,
• Zhaowu Yu   ORCID: orcid.org/0000-0003-4576-4541 9 ,
• Shuqing N. Teng 1 ,
• Robert R. Dunn 10 &
• Chi Xu   ORCID: orcid.org/0000-0002-1841-9032 1  

Nature Communications volume  15 , Article number:  7108 ( 2024 ) Cite this article

28 Altmetric

Metrics details

• Climate-change mitigation
• Urban ecology

Climate warming disproportionately impacts countries in the Global South by increasing extreme heat exposure. However, geographic disparities in adaptation capacity are unclear. Here, we assess global inequality in green spaces, which urban residents critically rely on to mitigate outdoor heat stress. We use remote sensing data to quantify daytime cooling by urban greenery in the warm seasons across the ~500 largest cities globally. We show a striking contrast, with Global South cities having ~70% of the cooling capacity of cities in the Global North (2.5 ± 1.0 °C vs. 3.6 ± 1.7 °C). A similar gap occurs for the cooling adaptation benefits received by an average resident in these cities (2.2 ± 0.9 °C vs. 3.4 ± 1.7 °C). This cooling adaptation inequality is due to discrepancies in green space quantity and quality between cities in the Global North and South, shaped by socioeconomic and natural factors. Our analyses further suggest a vast potential for enhancing cooling adaptation while reducing global inequality.

Similar content being viewed by others

Global climate-driven trade-offs between the water retention and cooling benefits of urban greening

Water, energy and climate benefits of urban greening throughout Europe under different climatic scenarios

Magnitude of urban heat islands largely explained by climate and population

Introduction.

Heat extremes are projected to be substantially intensified by global warming 1 , 2 , imposing a major threat to human mortality and morbidity in the coming decades 3 , 4 , 5 , 6 . This threat is particularly concerning as a majority of people now live in cities 7 , including those cities suffering some of the hottest climate extremes. Cities face two forms of warming: warming due to climate change and warming due to the urban heat island effect 8 , 9 , 10 . These two forms of warming have the potential to be additive, or even multiplicative. Climate change in itself is projected to result in rising maximum temperatures above 50 °C for a considerable fraction of the world if 2 °C global warming is exceeded 2 ; the urban heat island effect will cause up to >10 °C additional (surface) warming 11 . Exposures to temperatures above 35 °C with high humidity or above 40 °C with low humidity can lead to lethal heat stress for humans 12 . Even before such lethal temperatures are reached, worker productivity 13 and general health and well-being 14 can suffer. Heat extremes are especially risky for people living in the Global South 15 , 16 due to warmer climates at low latitudes. Climate models project that the lethal temperature thresholds will be exceeded with increasing frequencies and durations, and such extreme conditions will be concentrated in low-latitude regions 17 , 18 , 19 . These low-latitude regions overlap with the major parts of the Global South where population densities are already high and where population growth rates are also high. Consequently, the number of people exposed to extreme heat will likely increase even further, all things being equal 16 , 20 . That population growth will be accompanied by expanded urbanization and intensified urban heat island effects 21 , 22 , potentially exacerbating future Global North-Global South heat stress exposure inequalities.

Fortunately, we know that heat stress can be buffered, in part, by urban vegetation 23 . Urban green spaces, and especially urban forests, have proven an effective means through which to ameliorate heat stress through shading 24 , 25 and transpirational cooling 26 , 27 . The buffering effect of urban green spaces is influenced by their area (relative to the area of the city) and their spatial configuration 28 . In this context, green spaces become a kind of infrastructure that can and should be actively managed. At broad spatial scales, the effect of this urban green infrastructure is also mediated by differences among regions, whether in their background climate 29 , composition of green spaces 30 , or other factors 31 , 32 , 33 , 34 . The geographic patterns of the buffering effects of green spaces, whether due to geographic patterns in their areal extent or region-specific effects, have so far been poorly characterized.

On their own, the effects of climate change and urban heat islands on human health are likely to become severe. However, these effects will become even worse if they fall disproportionately in cities or countries with less economic ability to invest in green space 35 or in other forms of cooling 36 , 37 . A number of studies have now documented the so-called ‘luxury effect,’ wherein lower-income parts of cities tend to have less green space and, as a result, reduced biodiversity 38 , 39 . Where the luxury effect exists, green space and its benefits become, in essence, a luxury good 40 . If the luxury effect holds among cities, and lower-income cities also have smaller green spaces, the Global South may have the least potential to mitigate the combined effects of climate warming and urban heat islands, leading to exacerbated and rising inequalities in heat exposure 41 .

Here, we assess the global inequalities in the cooling capability of existing urban green infrastructure across urban areas worldwide. To this end, we use remotely sensed data to quantify three key variables, i.e., (1) cooling efficiency, (2) cooling capacity, and (3) cooling benefit of existing urban green infrastructure for ~500 major cities across the world. Urban green infrastructure and temperature are generally negatively and relatively linearly correlated at landscape scales, i.e., higher quantities of urban green infrastructure yield lower temperatures 42 , 43 . Cooling efficiency is widely used as a measure of the extent to which a given proportional increase in the area of urban green infrastructure leads to a decrease in temperature, i.e., the slope of the urban green infrastructure-temperature relationship 42 , 44 , 45 (see Methods for details). This simple metric allows quantifying the quality of urban green infrastructure in terms of ameliorating the urban heat island effect. Meanwhile, the extent to which existing urban green infrastructure cools down an entire city’s surface temperatures (compared to the non-vegetated built-up areas) is referred to as cooling capacity. Hence, cooling capacity is a function of the total quantity of urban green infrastructure and its cooling efficiency (see Methods).

As a third step, we account for the spatial distributions of urban green infrastructure and populations to quantify the benefit of cooling mitigation received by an average urban inhabitant in each city given their location. This cooling benefit is a more direct measure of the cooling realized by people, after accounting for the within-city geography of urban green infrastructure and population density. We focus on cooling capacity and cooling benefit as the measures of the cooling capability of individual cities for assessing their global inequalities. We are particularly interested in linking cooling adaptation inequality with income inequality 40 , 46 . While this can be achieved using existing income metrics for country classifications 47 , here we use the traditional Global North/South classification due to its historical ties to geography which is influential in climate research.

Results and discussion

Our analyses indicate that existing green infrastructure of an average city has a capability of cooling down surface temperatures by ~3 °C during warm seasons. However, a concerning disparity is evident; on average Global South cities have only two-thirds the cooling capacity and cooling benefit compared to Global North cities. This inequality is attributable to the differences in both quantity and quality of existing urban green infrastructure among cities. Importantly, we find that there exists considerable potential for many cities to enhance the cooling capability of their green infrastructure; achieving this potential could dramatically reduce global inequalities in adaptation to outdoor heat stress.

Quantifying cooling inequality

Our analyses showed that both the quantity and quality of the existing urban green infrastructure vary greatly among the world’s ~500 most populated cities (see Methods for details, and Fig.  1 for examples). The quantity of urban green infrastructure measured based on remotely sensed indicators of spectral greenness (Normalized Difference Vegetation Index, NDVI, see Methods) had a coefficient of variation (CV) of 35%. Similarly, the quality of urban green infrastructure in terms of cooling efficiency (daytime land surface temperatures during peak summer) had a CV of 37% (Supplementary Figs.  1 , 2 ). The global mean value of cooling capacity is 2.9 °C; existing urban green infrastructure ameliorates warm-season heat stress by 2.9 °C of surface temperature in an average city. In truth, however, the variation in cooling capacity was great (global CV in cooling capacity as large as ~50%), such that few cities were average. This variation is strongly geographically structured. Cities closer to the equator - tropical and subtropical cities - tend to have relatively weak cooling capacities (Fig.  2a, b ). As Global South countries are predominantly located at low latitudes, this pattern leads to a situation in which Global South cities, which tend to be hotter and relatively lower-income, have, on average, approximately two-thirds the cooling capacity of the Global North cities (2.5 ± 1.0 vs. 3.6 ± 1.7°C, Wilcoxon test, p  = 2.7e-12; Fig.  2c ). The cities that most need to rely on green infrastructure are, at present, those that are least able to do so.

a , e , i , m , q Los Angeles, US. b , f , j , n , r Paris, France. c , g , k , o , s Shanghai, China. d , h , l , p , t Cairo, Egypt. Local cooling efficiency is calculated for different local climate zone types to account for within-city heterogeneity. In densely populated parts of cities, local cooling capacity tends to be lower due to reduced green space area, whereas local cooling benefit (local cooling capacity multiplied by a weight term of local population density relative to city mean) tends to be higher as more urban residents can receive cooling amelioration.

a Global distribution of cooling capacity for the 468 major urbanized areas. b Latitudinal pattern of cooling capacity. c Cooling capacity difference between the Global North and South cities. The cooling capacity offered by urban green infrastructure evinces a latitudinal pattern wherein lower-latitude cities have weaker cooling capacity ( b , cubic-spline fitting of cooling capacity with 95% confidence interval is shown), representing a significant inequality between Global North and South countries: city-level cooling capacity for Global North cities are about 1.5-fold higher than in Global South cities ( c ). Data are presented as box plots, where median values (center black lines), 25th percentiles (box lower bounds), 75th percentiles (box upper bounds), whiskers extending to 1.5-fold of the interquartile range (IQR), and outliers are shown. The tails of the cooling capacity distributions are truncated at zero as all cities have positive values of cooling capacity. Notice that no cities in the Global South have a cooling capacity greater than 5.5 °C ( c ). This is because no cities in the Global South have proportional green space areas as great as those seen in the Global North (see also Fig.  4b ). A similar pattern is found for cooling benefit (Supplementary Fig.  3 ). The two-sided non-parametric Wilcoxon test was used for statistical comparisons.

When we account for the locations of urban green infrastructure relative to humans within cities, the cooling benefit of urban green infrastructure realized by an average urban resident generally becomes slightly lower than suggested by cooling capacity (see Methods; Supplementary Fig.  3 ). Urban residents tend to be densest in the parts of cities with less green infrastructure. As a result, the average urban resident experiences less cooling amelioration than expected. However, this heterogeneity has only a minor effect on global-scale inequality. As a result, the geographic trends in cooling capacity and cooling benefit are similar: mean cooling benefit for an average urban resident also presents a 1.5-fold gap between Global South and North cities (2.2 ± 0.9 vs. 3.4 ± 1.7 °C, Wilcoxon test, p  = 3.2e-13; Supplementary Fig.  3c ). Urban green infrastructure is a public good that has the potential to help even the most marginalized populations stay cool; unfortunately, this public benefit is least available in the Global South. When walking outdoors, the average person in an average Global South city receives only two-thirds the cooling amelioration from urban green infrastructure experienced by a person in an average Global North city. The high cooling amelioration capacity and benefit of the Global North cities is heavily influenced by North America (specifically, Canada and the US), which have both the highest cooling efficiency and the largest area of green infrastructure, followed by Europe (Supplementary Fig.  4 ).

One way to illustrate the global inequality of cooling capacity or benefit is to separately look at the cities that are most and least effective in ameliorating outdoor heat stress. Our results showed that ~85% of the 50 most effective cities (with highest cooling capacity or cooling benefit) are located in the Global North, while ~80% of the 50 least effective are Global South cities (Fig.  3 , Supplementary Fig.  5 ). This is true without taking into account the differences in the background temperatures and climate warming of these cities, which will exacerbate the effects on human health; cities in the Global South are likely to be closer to the limits of human thermal comfort and even, increasingly, the limits of the temperatures and humidities (wet-bulb temperatures) at which humans can safely work or even walk, such that the ineffectiveness of green spaces in those cities in cooling will lead to greater negative effects on human health 48 , work 14 , and gross domestic product (GDP) 49 . In addition, Global South cities commonly have higher population densities (Fig.  3 , Supplementary Fig.  5 ) and are projected to have faster population growth 50 . This situation will plausibly intensify the urban heat island effect because of the need of those populations for housing (and hence tensions between the need for buildings and the need for green spaces). It will also increase the number of people exposed to extreme urban heat island effects. Therefore, it is critical to increase cooling benefit via expanding urban green spaces, so that more people can receive the cooling mitigation from a given new neighboring green space if they live closer to each other. Doing so will require policies that incentivize urban green spaces as well as architectural innovations that make innovations such as plant-covered buildings easier and cheaper to implement.

The axes on the right are an order of magnitude greater than those on the left, such that the cooling capacity of Charlotte in the United States is about 37-fold greater than that of Mogadishu (Somalia) and 29-fold greater than that of Sana’a (Yemen). The cities presenting lowest cooling capacities are most associated with Global South cities at higher population densities.

Of course, cities differ even within the Global North or within the Global South. For example, some Global South cities have high green space areas (or relatively high cooling efficiency in combination with moderate green space areas) and hence high cooling capacity. These cities, such as Pune (India), will be important to study in more detail, to shed light on the mechanistic details of their cooling abilities as well as the sociopolitical and other factors that facilitated their high green area coverage and cooling capabilities (Supplementary Figs.  6 , 7 ).

We conducted our primary analyses using a spatial grain of 100-m grid cells and Landsat NDVI data for quantifying spectral greenness. Our results, however, were robust at the coarser spatial grain of 1 km. We find a slightly larger global cooling inequality (~2-fold gap between Global South and North cities) at the 1-km grain using MODIS data (see Methods and Supplementary Fig.  17 ). MODIS data have been frequently used for quantifying urban heat island effects and cooling mitigation 44 , 45 , 51 . Our results reinforce its robustness for comparing urban thermal environments between cities across broad scales.

Influencing factors

The global inequality of cooling amelioration could have a number of proximate causes. To understand their relative influence, we first separately examined the effects of quality (cooling efficiency) and quantity (NDVI as a proxy indicator of urban green space area) of urban green infrastructure. The simplest null model is one in which cooling capacity (at the city scale) and cooling benefit (at the human scale) are driven primarily by the proportional area in a city dedicated to green spaces. Indeed, we found that both cooling capacity and cooling benefit were strongly correlated with urban green space area (Fig.  4 , Supplementary Fig.  8 ). This finding is useful with regards to practical interventions. In general, cities that invest in saving or restoring more green spaces will receive more cooling benefits from those green spaces. By contrast, differences among cities in cooling efficiency played a more minor role in determining the cooling capacity and benefit of cities (Fig.  4 , Supplementary Fig.  8 ).

a Relationship between cooling efficiency and cooling capacity. b Relationship between green space area (measured by mean Landsat NDVI in the hottest month of 2018) and cooling capacity. Note that the highest level of urban green space area in the Global South cities is much lower than that in the Global North (dashed line in b ). Gray bands indicate 95% confidence intervals. Two-sided t-tests were conducted. c A piecewise structural equation model based on assumed direct and indirect (through influencing cooling efficiency and urban green space area) effects of essential natural and socioeconomic factors on cooling capacity. Mean annual temperature and precipitation, and topographic variation (elevation range) are selected to represent basic background natural conditions; GDP per capita is selected to represent basic socioeconomic conditions. The spatial extent of built-up areas is included to correct for city size. A bi-directional relationship (correlation) is fitted between mean annual temperature and precipitation. Red and blue solid arrows indicate significantly negative and positive coefficients with p  ≤ 0.05, respectively. Gray dashed arrows indicate p  > 0.05. The arrow width illustrates the effect size. Similar relationships are found for cooling benefits realized by an average urban resident (see Supplementary Fig.  8 ).

A further question is what shapes the quality and quantity of urban green infrastructure (which in turn are driving cooling capacity)? Many inter-correlated factors are possibly operating at multiple scales, making it difficult to disentangle their effects, especially since experiment-based causal inference is usually not feasible for large-scale urban systems. From a macroscopic perspective, we test the simple hypothesis that the background natural and socioeconomic conditions of cities jointly affect their cooling capacity and benefit in both direct and indirect ways. To this end, we constructed a minimal structural equation model including only the most essential variables reflecting background climate (mean annual temperature and precipitation), topographic variation (elevation range), as well as gross domestic product (GDP) per capita and city area (see Methods; Fig.  4c ).

We found that the quantity of green spaces in a city (again, in proportion to its size) was positively correlated with GDP per capita and city area; wealthier cities have more green spaces. It is well known that wealth and green spaces are positively correlated within cities (the luxury effect) 40 , 46 ; our analysis shows that a similar luxury effect occurs among them at a global scale. In addition, larger cities often have proportionally more green spaces, an effect that may be due to the tendency for large cities (particularly in the US and Canada) to have lower population densities. Cities that were hotter and had more topographic variation tended to have fewer green spaces and those that were more humid tended to have more green spaces. Given that temperature and humidity are highly correlated with the geography of the Global South and Global North, it is difficult to know whether these effects are due to the direct effects of temperature and precipitation, for example, on the growth rate of vegetation and hence the transition of abandoned lots into green spaces, or are associated with historical, cultural and political differences that via various mechanisms correlate to climate. Our structural equation model explained only a small fraction of variation among cities in their cooling efficiency, which is to say the quality of their green space. Cooling efficiency was modestly influenced by background temperature and precipitation—the warmer a city, the greater the cooling efficiency in that city; conversely, the more humid a city the less the cooling efficiency of that city.

Our analyses suggested that the lower cooling adaptation capabilities of Global South cities can be explained by their lower quantity of green infrastructure and, to a much lesser extent, their weaker cooling efficiency (quality; Supplementary Fig.  2 ). These patterns appear to be in part structured by GDP, but are also associated with climatic conditions 39 , and other factors. A key question, unresolved by our work, is whether the climatic correlates of the size of green spaces in cities are due to the effects of climate per se or if they, instead, reflect correlates between contemporary climate and the social, cultural, and political histories of cities in the Global South 52 . Since urban planning has much inertia, especially in big cities, those choices might be correlated with climate because of the climatic correlates of political histories. It is also possible that these dynamics relate, in part, to the ways in which climate influences vegetation structure. However, this seems less likely given that under non-urban conditions vegetation cover (and hence cooling capacity) is normally positively correlated with mean annual temperature across the globe, opposite to our observed negative relationships for urban systems (Supplementary Fig.  9g ). Still, it is possible that increased temperatures in cities due to the urban heat island effects may lead to temperature-vegetation cover-cooling capacity relationships that differ from those in natural environments 53 , 54 . Indeed, a recent study found that climate warming will put urban forests at risk, and the risk is disproportionately higher in the Global South 55 .

Our model serves as a starting point for unraveling the mechanisms underlying global cooling inequality. We cannot rule out the possibility that other unconsidered factors correlated with the studied variables play important roles. We invite systematic studies incorporating detailed sociocultural and ecological variables to address this question across scales.

Potential of enhancing cooling and reducing inequality

Can we reduce the inequality in cooling capacity and benefits that we have discovered among the world’s largest cities? Nuanced assessments of the potential to improve cooling mitigation require comprehensive considerations of socioeconomic, cultural, and technological aspects of urban management and policy. It is likely that cities differ greatly in their capacity to implement cooling through green infrastructure, whether as a function of culture, governance, policy or some mix thereof. However, any practical attempts to achieve greater cooling will occur in the context of the realities of climate and existing land use. To understand these realities, we modeled the maximum additional cooling capacity that is possible in cities, given existing constraints. We assume that this capacity depends on the quality (cooling efficiency) and quantity of urban green infrastructure. Our approach provides a straightforward metric of the cooling that could be achieved if all parts of a city’s green infrastructure were to be enhanced systematically.

The positive outlook is that our analyses suggest a considerable potential of improving cooling capacity by optimizing urban green infrastructure. An obvious way is through increases in urban green infrastructure quantity. We employ an approach in which we consider each local climate zone 56 to have a maximum NDVI and cooling efficiency (see Methods). For a given local climate zone, the city with the largest NDVI values or cooling efficiency sets the regional upper bounds for urban green infrastructure quantities or quality that can be achieved. Notably, these maxima are below the maxima for forests or other non-urban spaces for the simple reason that, as currently imagined, cities must contain gray (non-green) spaces in the form of roads and buildings. In this context, we conduct a thought experiment. What if we could systematically increase NDVI of all grid cells in each city, per local climate zone type, to a level corresponding to the median NDVI of grid cells in that upper bound city while keeping cooling efficiency unchanged (see Methods). If we were able to achieve this goal, the cooling capacity of cities would increase by ~2.4 °C worldwide. The increase would be even greater, ~3.8°C, if the 90th percentile (within the reference maximum city) was reached (Fig.  5a ). The potential for cooling benefit to the average urban resident is similar to that of cooling capacity (Supplementary Fig.  10a ). There is also potential to reduce urban temperatures if we can enhance cooling efficiency. However, the benefits of increases in cooling efficiency are modest (~1.5 °C increases at the 90th percentile of regional upper bounds) when holding urban green infrastructure quantity constant. In theory, if we could maximize both quantity and cooling efficiency of urban green infrastructure (to 90th percentiles of their regional upper bounds respectively), we would yield increases in cooling capacity and benefit up to ~10 °C, much higher than enhancing green space area or cooling efficiency alone (Fig.  5a , Supplementary Fig.  10a ). Notably, such co-maximization of green space area and cooling efficiency would substantially reduce global inequality to Gini <0.1 (Fig.  5b , Supplementary Fig.  10b ). Our analyses thus provide an important suggestion that enhancing both green space quantity and quality can yield a synergistic effect leading to much larger gains than any single aspect alone.

a The potential of enhancing cooling capacity via either enhancing urban green infrastructure quality (i.e., cooling efficiency) while holding quantity (i.e., green space area) fixed (yellow), or enhancing quantity while holding quality fixed (blue) is much lower than that of enhancing both quantity and quality (green). The x-axis indicates the targets of enhancing urban green infrastructure quantity and/or quality relative to the 50–90th percentiles of NDVI or cooling efficiency, see Methods). The dashed horizontal lines indicate the median cooling capacity of current cities. Data are presented as median values with the colored bands corresponding to 25–75th percentiles. b The potential of reducing cooling capacity inequality is also higher when enhancing both urban green infrastructure quantity and quality. The Gini index weighted by population density is used to measure inequality. Similar results were found for cooling benefit (Supplementary Fig.  10 ).

Different estimates of cooling capacity potential may be reached based on varying estimates and assumptions regarding the maximum possible quantity and quality of urban green infrastructure. There is no single, simple way to make these estimates, especially considering the huge between-city differences in society, culture, and structure across the globe. Our example case (above) begins from the upper bound city’s median NDVI, taking into account different local climate zone types and background climate regions (regional upper bounds). This is based on the assumption that for cities within the same climate regions, their average green space quantity may serve as an attainable target. Still, urban planning is often made at the level of individual cities, often only implemented to a limited extent and made with limited consideration of cities in other regions and countries. A potentially more realistic reference may be taken from the existing green infrastructure (again, per local climate zone type) within each particular city itself (see Methods): if a city’s sparsely vegetated areas was systematically elevated to the levels of 50–90th percentiles of NDVI within their corresponding local climate zones within the city, cooling capacity would still increase, but only by 0.5–1.5 °C and with only slightly reduced inequalities among cities (Supplementary Fig.  11 ). This highlights that ambitious policies, inspired by the greener cities worldwide, are necessary to realize the large cooling potential in urban green infrastructure.

In summary, our results demonstrate clear inequality in the extent to which urban green infrastructure cools cities and their denizens between the Global North and South. Much attention has been paid to the global inequality of indoor heat adaptation arising from the inequality of resources (e.g., less affordable air conditioning and more frequent power shortages in the Global South) 36 , 57 , 58 , 59 . Our results suggest that the inequality in outdoor adaptation is particularly concerning, especially as urban populations in the Global South are growing rapidly and are likely to face the most severe future temperature extremes 60 .

Previous studies have been focusing on characterizing urban heat island effects, urban vegetation patterns, resident exposure, and cooling effects in particular cities 26 , 28 , 34 , 61 , regions 22 , 25 , 62 , or continents 32 , 44 , 63 . Recent studies start looking at global patterns with respect to cooling efficiency or green space exposure 35 , 45 , 64 , 65 . Our approach is one drawn from the fields of large-scale ecology and macroecology. This approach is complementary to and, indeed, can, in the future, be combined with (1) mechanism driven biophysical models 66 , 67 to predict the influence of the composition and climate of green spaces on their cooling efficiency, (2) social theory aimed at understanding the factors that govern the amount of green space in cities as well as the disparity among cities 68 , (3) economic models of the effects of policy changes on the amount of greenspace and even (4) artist-driven projects that seek to understand the ways in which we might reimagine future cities 69 . Our simple explanatory model is, ultimately, one lens on a complex, global phenomenon.

Our results convey some positive outlook in that there is considerable potential to strengthen the cooling capability of cities and to reduce inequalities in cooling capacities at the same time. Realizing this nature-based solution, however, will be challenging. First, enhancing urban green infrastructure requires massive investments, which are more difficult to achieve in Global South cities. Second, it also requires smart planning strategies and advanced urban design and greening technologies 37 , 70 , 71 , 72 . Spatial planning of urban green spaces needs to consider not only the cooling amelioration effect, but also their multifunctional aspects that involve multiple ecosystem services, mental health benefits, accessibility, and security 73 . In theory, a city can maximize its cooling while also maximizing density through the combination of high-density living, ground-level green spaces, and vertical and rooftop gardens (or even forests). In practice, the current cities with the most green spaces tend to be lower-density cities 74 (Supplementary Fig.  12 ). Still, innovation and implementation of new technologies that allow green spaces and high-density living to be combined have the potential to reduce or disconnect the negative relationship between green space area and population density 71 , 75 . However, this development has yet to be realized. Another dimension of green spaces that deserves more attention is the geography of green spaces relative to where people are concentrated within cities. A critical question is how best should we distribute green spaces within cities to maximize cooling efficiency 76 and minimize within-city cooling inequality towards social equity 77 ? Last but not least, it is crucial to design and manage urban green spaces to be as resilient as possible to future climate stress 78 . For many cities, green infrastructure is likely to remain the primary means people will have to rely on to mitigate the escalating urban outdoor heat stress in the coming decades 79 .

We used the world population data from the World’s Cities in 2018 Data Booklet 80 to select 502 major cities with population over 1 million people (see Supplementary Data  1 for the complete list of the studied cities). Cities are divided into the Global North and Global South based on the Human Development Index (HDI) from the Human Development Report 2019 81 . For each selected city, we used the 2018 Global Artificial Impervious Area (GAIA) data at 30 m resolution 82 to determine its geographic extent. The derived urban boundary polygons thus encompass a majority of the built-up areas and urban residents. In using this approach, rather than urban administrative boundaries, we can focus on the relatively densely populated areas where cooling mitigation is most needed, and exclude areas dominated by (semi) natural landscapes that may bias the subsequent quantifications of the cooling effect. Our analyses on the cooling effect were conducted at the 100 m spatial resolution using Landsat data and WorldPop Global Project Population Data of 2018 83 . In order to test for the robustness of the results to coarser spatial scales, we also repeated the analyses at 1 km resolution using MODIS data, which have been extensively used for quantifying urban heat island effects and cooling mitigation 44 , 45 , 51 . We discarded the five cities with sizes <30 km 2 as they were too small for us to estimate their cooling efficiency based on linear regression (see section below for details). We combined closely located cities that form contiguous urban areas or urban agglomerations, if their urban boundary polygons from GAIA merged (e.g., Phoenix and Mesa in the United States were combined). Our approach yielded 468 polygons, each representing a major urbanized area that were the basis for all subsequent analyses. Because large water bodies can exert substantial and confounding cooling effects, we excluded permanent water bodies including lakes, reservoirs, rivers, and oceans using the Copernicus Global Land Service (CGLS) Land Cover data for 2018 at 10 m resolution 84 .

Quantifying the cooling effect

As a first step, we calculated cooling efficiency for each studied city within the GAIA-derived urban boundary. Cooling efficiency quantifies the extent to which a given area of green spaces in a city can reduce temperatures. It is a measure of the effectiveness (quality) of urban green spaces in terms of heat amelioration. Cooling efficiency is typically measured by calculating the slope of the relationship between remotely-sensed land surface temperature (LST) and vegetation cover through ordinary least square regression 42 , 44 , 45 . It is known that cooling efficiency varies between cities. Influencing factors might include background climate 29 , species composition 30 , 85 , landscape configuration 28 , topography 86 , proximity to large water bodies 33 , 87 , urban morphology 88 , and city management practices 31 . However, the mechanism underlying the global pattern of cooling efficiency remains unclear.

We used Landsat satellite data provided by the United States Geological Survey (USGS) to calculate the cooling efficiency of each studied city. We used the cloud-free Landsat 8 Level 2 LST and NDVI data. For each city we calculated the mean LST in each month of 2018 to identify the hottest month, and then derived the hottest month LST; we used the cloud-free Landsat 8 data to calculate the mean NDVI for the hottest month correspondingly.

We quantified cooling efficiency for different local climate zones 56 separately for each city, to account for within-city variability of thermal environments. To this end, we used the Copernicus Global Land Service data (CGLS) 84 and Global Human Settlement Layers (GHSL) Built-up height data 89 of 2018 at the 100 m resolution to identify five types of local climate zones: non-tree vegetation (shrubs, herbaceous vegetation, and cultivated vegetation according to the CGLS classification system), low-rise buildings (built up and bare according to the CGLS classification system, with building heights ≤10 m according to the GHSL data), medium-high-rise buildings (built up and bare areas with building heights >10 m), open tree cover (open forest with tree cover 15–70% according to the CGLS system), and closed tree cover (closed forest with tree cover >70%).

For each local climate zone type in each city, we constructed a regression model with NDVI as the predictor variable and LST as the response variable (using the ordinary least square method). We took into account the potential confounding factors including topographic elevation (derived from MERIT DEM dataset 90 ), building height (derived from the GHSL dataset 89 ), and distance to water bodies (derived from the GSHHG dataset 91 ), the model thus became: LST ~ NDVI + topography + building height + distance to water. Cooling efficiency was calculated as the absolute value of the regression coefficient of NDVI, after correcting for those confounding factors. To account for the multi-collinearity issue, we conducted variable selection based on the variance inflation factor (VIF) to achieve VIF < 5. Before the analysis, we discarded low-quality Landsat pixels, and filtered out the pixels with NDVI < 0 (normally less than 1% in a single city). Cooling efficiency is known to be influenced by within-city heterogeneity 92 , 93 , and, as a result, might sometimes better fit non-linear relationships at local scales 65 , 76 . However, our central aim is to assess global cooling inequality based on generalized relationships that fit the majority of global cities. Previous studies have shown that linear relationships can do this job 42 , 44 , 45 , therefore, here we used linear models to assess cooling efficiency.

As a second step, we calculated the cooling capacity of each city. Cooling capacity is a positive function of the magnitude of cooling efficiency and the proportional area of green spaces in a city and is calculated based on NDVI and the derived cooling efficiency (Eq.  1 , Supplementary Fig.  13 ):

where CC lcz and CE lcz are the cooling capacity and cooling efficiency for a given local climate zone type in a city, respectively; NDVI i is the mean NDVI for 100-m grid cell i ; NDVI min is the minimum NDVI across the city; and n is the total number of grid cells within the local climate zone. Local cooling capacity for each grid cell i (Fig.  1 , Supplementary Fig.  7 ) can be derived in this way as well (Supplementary Fig.  13 ). For a particular city, cooling capacity may be dependent on the spatial configuration of its land use/cover 28 , 94 , but here we condensed cooling capacity to city average (Eq.  2 ), thus did not take into account these local-scale factors.

where CC is the average cooling capacity of a city; n lcz is the number of grid cells of the local climate zone; m is the total number of grid cells within the whole city.

As a third step, we calculated the cooling benefit realized by an average urban resident (cooling benefit in short) in each city. Cooling benefit depends not only on the cooling capacity of a city, but also on where people live within a city relative to greener or grayer areas of the city. For example, cooling benefits in a city might be low even if the cooling capacity is high if the green parts and the dense-population parts of a city are inversely correlated. Here, we are calculating these averages while aware that in any particular city the exposure of a particular person will depend on the distribution of green spaces in a city, and the occupation, movement trajectories of a person, etc. On the scale of a city, we calculated cooling benefit following a previous study 35 , that is, simply adding a weight term of population size per 100-m grid cell into cooling capacity in Eq. ( 1 ):

Where CB lcz is the cooling benefit of a given local climate zone type in a specific city, pop i is the number of people within grid cell i , \(\overline{{pop}}\) is the mean population of the city.

Where CB is the average cooling benefit of a city. The population data were obtained from the 100-m resolution WorldPop Global Project Population Data of 2018 83 . Local cooling benefit for a given grid cell i can be calculated in a similar way, i.e., local cooling capacity multiplied by a weight term of local population density relative to mean population density. Local cooling benefits were mapped for example cities for the purpose of illustrating the effect of population spatial distribution (Fig.  1 , Supplementary Fig.  7 ), but their patterns were not examined here.

Based on the aforementioned three key variables quantified at 100 m grid cells, we conducted multivariate analyses to examine if and to what extent cooling efficiency and cooling benefit are shaped by essential natural and socioeconomic factors, including background climate (mean annual temperature from ECMWF ERA5 dataset 95 and precipitation from TerraClimate dataset 96 ), topography (elevation range 90 ), and GDP per capita 97 , with city size (geographic extent) corrected for. We did not include humidity because it is strongly correlated with temperature and precipitation, causing serious multi-collinearity problems. We used piecewise structural equation modeling to test the direct effects of these factors and indirect effects via influencing cooling efficiency and vegetation cover (Fig.  4c , Supplementary Fig.  8c ). To account for the potential influence of spatial autocorrelation, we used spatially autoregressive models (SAR) to test for the robustness of the observed effects of natural and socioeconomic factors on cooling capacity and benefit (Supplementary Fig.  14 ).

Testing for robustness

We conducted the following additional analyses to test for robustness. We obtained consistent results from these robustness analyses.

(1) We looked at the mean hottest-month LST and NDVI within 3 years (2017-2019) to check the consistency between the results based on relatively short (1 year) vs. long (3-year average) time periods (Supplementary Fig.  15 ).

(2) We carried out the approach at a coarser spatial scale of 1 km, using MODIS-derived NDVI and LST, as well as the population data 83 in the hottest month of 2018. In line with our finer-scale analysis of Landsat data, we selected the hottest month and excluded low-quality grids affected by cloud cover and water bodies 98 (water cover > 20% in 1 × 1 km 2 grid cells) of MODIS LST, and calculated the mean NDVI for the hottest month. We ultimately obtained 441 cities (or urban agglomerations) for analysis. At the 1 km resolution, some local climate zone types would yield insufficient samples for constructing cooling efficiency models. Therefore, instead of identifying local climate zone explicitly, we took an indirect approach to account for local climate confounding factors, that is, we constructed a multiple regression model for a whole city incorporating the hottest-month local temperature 95 , precipitation 96 , and humidity (based on NASA FLDAS dataset 99 ), albedo (derived from the MODIS MCD43A3 product 100 ), aerosol loading (derived from the MODIS MCD19A2 product 101 ), wind speed (based on TerraClimate dataset 96 ), topography elevation 90 , distance to water 91 , urban morphology (building height 102 ), and human activity intensity (VIIRS nighttime light data as a proxy indicator 103 ). We used the absolute value of the linear regression coefficient of NDVI as the cooling efficiency of the whole city (model: LST ~ NDVI + temperature + precipitation + humidity + distance to water + topography + building height + albedo + aerosol + wind speed + nighttime light), and calculated cooling capacity and cooling benefit based on the same method. Variable selection was conducted using the criterion of VIF < 5.

Our results indicated that MODIS-based cooling capacity and cooling benefit are significantly correlated with the Landsat-based counterparts (Supplementary Fig.  16 ); importantly, the gap between the Global South and North cities is around two-fold, close to the result from the Landsat-based result (Supplementary Fig.  17 ).

(3) For the calculation of cooling benefit, we considered different spatial scales of human accessibility to green spaces: assuming the population in each 100 × 100 m 2 grid cell could access to green spaces within neighborhoods of certain extents, we calculated cooling benefit by replacing NDVI i in Eq. ( 3 ) with mean NDVI within the 300 × 300 m 2 and 500 × 500 m 2 extents centered at the focal grid cell (Supplementary Fig.  18 ).

(4) Considering cities may vary in minimum NDVI, we assessed if this variation could affect resulting cooling capacity patterns. To this end, we calculated the cooling capacity for each studied city using NDVI = 0 as the reference (i.e., using NDVI = 0 instead of minimum NDVI in Supplementary Fig.  13b ), and correlated it with that using minimum NDVI as the reference (Supplementary Fig.  19 ).

Quantifying between-city inequality

Inequalities in access to the benefits of green spaces in cities exist within cities, as is increasingly well-documented 104 . Here, we focus instead on the inequalities among cities. We used the Gini coefficient to measure the inequality in cooling capacity and cooling benefit between all studied cities across the globe as well as between Global North or South cities. We calculated Gini using the population-density weighted method (Fig.  5b ), as well as the unweighted and population-size weighted methods (Supplementary Fig.  20 ).

Estimating the potential for more effective and equal cooling amelioration

We estimated the potential of enhancing cooling amelioration based on the assumptions that urban green space quality (cooling efficiency) and quantity (NDVI) can be increased to different levels, and that relative spatial distributions of green spaces and population can be idealized (so that their spatial matches can maximize cooling benefit). We assumed that macro-climate conditions act as the constraints of vegetation cover and cooling efficiency. We calculated the 50th, 60th, 70th, 80th, and 90th percentiles of NDVI within each type of local climate zone of each city. For a given local climate zone type, we obtained the city with the highest NDVI per percentile value as the regional upper bounds of urban green infrastructure quantity. The regional upper bounds of cooling efficiency are derived in a similar way. For each local climate zone in a city, we generated a potential NDVI distribution where all grid cells reach the regional upper bound values for the 50th, 60th, 70th, 80th, or 90th percentile of urban green space quantity or quality, respectively. NDVI values below these percentiles were increased, whereas those above these percentiles remained unchanged. The potential estimates are essentially dependent on the references, i.e., the optimal cooling efficiency and NDVI that a given city can reach. However, such references are obviously difficult to determine, because complex natural and socioeconomic conditions could play important roles in determining those cooling optima, and the dominant factors are unknown at a global scale. We employed the simplifying assumption that background climate could act as an essential constraint according to our results. We therefore used the Köppen climate classification system 105 to determine the reference separately in each climate region (tropical, arid, temperate, and continental climate regions were involved for all studied cities).

We calculated potential cooling capacity and cooling benefit based on these potential NDVI maps (Fixed cooling efficiency in Fig.  5 ). We then calculated the potentials if cooling efficiency of each city can be enhanced to 50–90th percentile across all urban local climate zones within the corresponding biogeographic region (Fixed green space area in Fig.  5 ). We also calculated the potentials if both NDVI and cooling efficiency were enhanced (Enhancing both in Fig.  5) to a certain corresponding level (i.e., i th percentile NDVI +  i th percentile cooling efficiency). We examined if there are additional effects of idealizing relative spatial distributions of urban green spaces and humans on cooling benefits. To this end, the pixel values of NDVI or population amount remained unchanged, but their one-to-one correspondences were based on their ranking: the largest population corresponds to the highest NDVI, and so forth. Under each scenario, we calculated cooling capacity and cooling benefit for each city, and the between-city inequality was measured by the Gini coefficient.

We used the Google Earth Engine to process the spatial data. The statistical analyses were conducted using R v4.3.3 106 , with car v3.1-2 107 , piecewiseSEM v2.1.2 108 , and ineq v0.2-13 109 packages. The global maps of cooling were created using the ArcGIS v10.3 software.

Reporting summary

Further information on research design is available in the  Nature Portfolio Reporting Summary linked to this article.

Data availability

City population statistics data is collected from the Population Division of the Department of Economic and Social Affairs of the United Nations ( https://www.un.org/development/desa/pd/content/worlds-cities-2018-data-booklet ). Global North-South division is based on Human Development Report 2019 which from United Nations Development Programme ( https://hdr.undp.org/content/human-development-report-2019 ). Global urban boundaries from GAIA data are available from Star Cloud Data Service Platform ( https://data-starcloud.pcl.ac.cn/resource/14 ) . Global water data is derived from 2018 Copernicus Global Land Service (CGLS 100-m) data ( https://developers.google.com/earth-engine/datasets/catalog/COPERNICUS_Landcover_100m_Proba-V-C3_Global ), European Space Agency (ESA) WorldCover 10 m 2020 product ( https://developers.google.com/earth-engine/datasets/catalog/ESA_WorldCover_v100 ), and GSHHG (A Global Self-consistent, Hierarchical, High-resolution Geography Database) at https://www.soest.hawaii.edu/pwessel/gshhg/ . Landsat 8 LST and NDVI data with 30 m resolution are available at  https://developers.google.com/earth-engine/datasets/catalog/LANDSAT_LC08_C02_T1_L2 . Land surface temperature (LST) data with 1 km from MODIS Aqua product (MYD11A1) is available at https://developers.google.com/earth-engine/datasets/catalog/MODIS_061_MYD11A1 . NDVI (1 km) dataset from MYD13A2 is available at https://developers.google.com/earth-engine/datasets/catalog/MODIS_061_MYD13A2 . Population data (100 m) is derived from WorldPop ( https://developers.google.com/earth-engine/datasets/catalog/WorldPop_GP_100m_pop ). Local climate zones are also based on 2018 CGLS data ( https://developers.google.com/earth-engine/datasets/catalog/COPERNICUS_Landcover_100m_Proba-V-C3_Global ), and built-up height data is available from Global Human Settlement Layers (GHSL, 100 m) ( https://developers.google.com/earth-engine/datasets/catalog/JRC_GHSL_P2023A_GHS_BUILT_H ). Temperature data is calculated from ERA5-Land Monthly Aggregated dataset ( https://developers.google.com/earth-engine/datasets/catalog/ECMWF_ERA5_LAND_MONTHLY_AGGR ). Precipitation and wind data are calculated from TerraClimate (Monthly Climate and Climatic Water Balance for Global Terrestrial Surfaces, University of Idaho) ( https://developers.google.com/earth-engine/datasets/catalog/IDAHO_EPSCOR_TERRACLIMATE ). Humidity data is calculated from Famine Early Warning Systems Network (FEWS NET) Land Data Assimilation System ( https://developers.google.com/earth-engine/datasets/catalog/NASA_FLDAS_NOAH01_C_GL_M_V001 ). Topography data from MERIT DEM (Multi-Error-Removed Improved-Terrain DEM) product is available at https://developers.google.com/earth-engine/datasets/catalog/MERIT_DEM_v1_0_3 . GDP from Gross Domestic Product and Human Development Index dataset is available at https://doi.org/10.5061/dryad.dk1j0 . VIIRS nighttime light data is available at https://developers.google.com/earth-engine/datasets/catalog/NOAA_VIIRS_DNB_MONTHLY_V1_VCMSLCFG . City building volume data from Global 3D Building Structure (1 km) is available at https://doi.org/10.34894/4QAGYL . Albedo data is derived from the MODIS MCD43A3 product ( https://developers.google.com/earth-engine/datasets/catalog/MODIS_061_MCD43A3 ), and aerosol data is derived from the MODIS MCD19A2 product ( https://developers.google.com/earth-engine/datasets/catalog/MODIS_061_MCD19A2_GRANULES ). All data used for generating the results are publicly available at https://doi.org/10.6084/m9.figshare.26340592.v1 .

Code availability

The codes used for data collection and analyses are publicly available at https://doi.org/10.6084/m9.figshare.26340592.v1 .

Dosio, A., Mentaschi, L., Fischer, E. M. & Wyser, K. Extreme heat waves under 1.5 °C and 2 °C global warming. Environ. Res. Lett. 13 , 054006 (2018).

Article   ADS   Google Scholar  

Suarez-Gutierrez, L., Müller, W. A., Li, C. & Marotzke, J. Hotspots of extreme heat under global warming. Clim. Dyn. 55 , 429–447 (2020).

Article   Google Scholar  

Guo, Y. et al. Global variation in the effects of ambient temperature on mortality: a systematic evaluation. Epidemiology 25 , 781–789 (2014).

Article   PubMed   PubMed Central   Google Scholar  

Mora, C. et al. Global risk of deadly heat. Nat. Clim. Chang. 7 , 501–506 (2017).

Ebi, K. L. et al. Hot weather and heat extremes: health risks. Lancet 398 , 698–708 (2021).

Article   PubMed   Google Scholar  

Lüthi, S. et al. Rapid increase in the risk of heat-related mortality. Nat. Commun. 14 , 4894 (2023).

Article   ADS   PubMed   PubMed Central   Google Scholar  

United Nations Department of Economic Social Affairs, Population Division. in World Population Prospects 2022: Summary of Results (United Nations Fund for Population Activities, 2022).

Sachindra, D., Ng, A., Muthukumaran, S. & Perera, B. Impact of climate change on urban heat island effect and extreme temperatures: a case‐study. Q. J. R. Meteorol. Soc. 142 , 172–186 (2016).

Guo, L. et al. Evaluating contributions of urbanization and global climate change to urban land surface temperature change: a case study in Lagos, Nigeria. Sci. Rep. 12 , 14168 (2022).

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Liu, Z. et al. Surface warming in global cities is substantially more rapid than in rural background areas. Commun. Earth Environ. 3 , 219 (2022).

Mentaschi, L. et al. Global long-term mapping of surface temperature shows intensified intra-city urban heat island extremes. Glob. Environ. Change 72 , 102441 (2022).

Asseng, S., Spänkuch, D., Hernandez-Ochoa, I. M. & Laporta, J. The upper temperature thresholds of life. Lancet Planet. Health 5 , e378–e385 (2021).

Zander, K. K., Botzen, W. J., Oppermann, E., Kjellstrom, T. & Garnett, S. T. Heat stress causes substantial labour productivity loss in Australia. Nat. Clim. Chang. 5 , 647–651 (2015).

Flouris, A. D. et al. Workers’ health and productivity under occupational heat strain: a systematic review and meta-analysis. Lancet Planet. Health 2 , e521–e531 (2018).

Xu, C., Kohler, T. A., Lenton, T. M., Svenning, J.-C. & Scheffer, M. Future of the human climate niche. Proc. Natl Acad. Sci. USA 117 , 11350–11355 (2020).

Lenton, T. M. et al. Quantifying the human cost of global warming. Nat. Sustain. 6 , 1237–1247 (2023).

Harrington, L. J. et al. Poorest countries experience earlier anthropogenic emergence of daily temperature extremes. Environ. Res. Lett. 11 , 055007 (2016).

Bathiany, S., Dakos, V., Scheffer, M. & Lenton, T. M. Climate models predict increasing temperature variability in poor countries. Sci. Adv. 4 , eaar5809 (2018).

Alizadeh, M. R. et al. Increasing heat‐stress inequality in a warming climate. Earth Future 10 , e2021EF002488 (2022).

Tuholske, C. et al. Global urban population exposure to extreme heat. Proc. Natl Acad. Sci. USA 118 , e2024792118 (2021).

Article   CAS   PubMed   PubMed Central   Google Scholar  

Manoli, G. et al. Magnitude of urban heat islands largely explained by climate and population. Nature 573 , 55–60 (2019).

Article   ADS   CAS   PubMed   Google Scholar  

Wang, J. et al. Anthropogenic emissions and urbanization increase risk of compound hot extremes in cities. Nat. Clim. Chang. 11 , 1084–1089 (2021).

Article   ADS   CAS   Google Scholar  

Bowler, D. E., Buyung-Ali, L., Knight, T. M. & Pullin, A. S. Urban greening to cool towns and cities: a systematic review of the empirical evidence. Landsc. Urban Plan. 97 , 147–155 (2010).

Armson, D., Stringer, P. & Ennos, A. The effect of tree shade and grass on surface and globe temperatures in an urban area. Urban For. Urban Green. 11 , 245–255 (2012).

Wang, C., Wang, Z. H. & Yang, J. Cooling effect of urban trees on the built environment of contiguous United States. Earth Future 6 , 1066–1081 (2018).

Pataki, D. E., McCarthy, H. R., Litvak, E. & Pincetl, S. Transpiration of urban forests in the Los Angeles metropolitan area. Ecol. Appl. 21 , 661–677 (2011).

Konarska, J. et al. Transpiration of urban trees and its cooling effect in a high latitude city. Int. J. Biometeorol. 60 , 159–172 (2016).

Article   ADS   PubMed   Google Scholar  

Li, X., Zhou, W., Ouyang, Z., Xu, W. & Zheng, H. Spatial pattern of greenspace affects land surface temperature: evidence from the heavily urbanized Beijing metropolitan area, China. Landsc. Ecol. 27 , 887–898 (2012).

Yu, Z., Xu, S., Zhang, Y., Jørgensen, G. & Vejre, H. Strong contributions of local background climate to the cooling effect of urban green vegetation. Sci. Rep. 8 , 6798 (2018).

Richards, D. R., Fung, T. K., Belcher, R. & Edwards, P. J. Differential air temperature cooling performance of urban vegetation types in the tropics. Urban For. Urban Green. 50 , 126651 (2020).

Winbourne, J. B. et al. Tree transpiration and urban temperatures: current understanding, implications, and future research directions. BioScience 70 , 576–588 (2020).

Schwaab, J. et al. The role of urban trees in reducing land surface temperatures in European cities. Nat. Commun. 12 , 6763 (2021).

Vo, T. T. & Hu, L. Diurnal evolution of urban tree temperature at a city scale. Sci. Rep. 11 , 10491 (2021).

Wang, J. et al. Comparing relationships between urban heat exposure, ecological structure, and socio-economic patterns in Beijing and New York City. Landsc. Urban Plan. 235 , 104750 (2023).

Chen, B. et al. Contrasting inequality in human exposure to greenspace between cities of Global North and Global South. Nat. Commun. 13 , 4636 (2022).

Pavanello, F. et al. Air-conditioning and the adaptation cooling deficit in emerging economies. Nat. Commun. 12 , 6460 (2021).

Turner, V. K., Middel, A. & Vanos, J. K. Shade is an essential solution for hotter cities. Nature 619 , 694–697 (2023).

Hope, D. et al. Socioeconomics drive urban plant diversity. Proc. Natl Acad. Sci. USA 100 , 8788–8792 (2003).

Leong, M., Dunn, R. R. & Trautwein, M. D. Biodiversity and socioeconomics in the city: a review of the luxury effect. Biol. Lett. 14 , 20180082 (2018).

Schwarz, K. et al. Trees grow on money: urban tree canopy cover and environmental justice. PloS ONE 10 , e0122051 (2015).

Chakraborty, T., Hsu, A., Manya, D. & Sheriff, G. Disproportionately higher exposure to urban heat in lower-income neighborhoods: a multi-city perspective. Environ. Res. Lett. 14 , 105003 (2019).

Wang, J. et al. Significant effects of ecological context on urban trees’ cooling efficiency. ISPRS J. Photogramm. Remote Sens. 159 , 78–89 (2020).

Marando, F. et al. Urban heat island mitigation by green infrastructure in European Functional Urban Areas. Sust. Cities Soc. 77 , 103564 (2022).

Cheng, X., Peng, J., Dong, J., Liu, Y. & Wang, Y. Non-linear effects of meteorological variables on cooling efficiency of African urban trees. Environ. Int. 169 , 107489 (2022).

Yang, Q. et al. Global assessment of urban trees’ cooling efficiency based on satellite observations. Environ. Res. Lett. 17 , 034029 (2022).

Yin, Y., He, L., Wennberg, P. O. & Frankenberg, C. Unequal exposure to heatwaves in Los Angeles: Impact of uneven green spaces. Sci. Adv. 9 , eade8501 (2023).

Fantom N., Serajuddin U. The World Bank’s Classification of Countries by Income (The World Bank, 2016).

Iungman, T. et al. Cooling cities through urban green infrastructure: a health impact assessment of European cities. Lancet 401 , 577–589 (2023).

He, C. et al. The inequality labor loss risk from future urban warming and adaptation strategies. Nat. Commun. 13 , 3847 (2022).

Kii, M. Projecting future populations of urban agglomerations around the world and through the 21st century. npj Urban Sustain 1 , 10 (2021).

Paschalis, A., Chakraborty, T., Fatichi, S., Meili, N. & Manoli, G. Urban forests as main regulator of the evaporative cooling effect in cities. AGU Adv. 2 , e2020AV000303 (2021).

Hunte, N., Roopsind, A., Ansari, A. A. & Caughlin, T. T. Colonial history impacts urban tree species distribution in a tropical city. Urban For. Urban Green. 41 , 313–322 (2019).

Kabano, P., Harris, A. & Lindley, S. Sensitivity of canopy phenology to local urban environmental characteristics in a tropical city. Ecosystems 24 , 1110–1124 (2021).

Frank, S. D. & Backe, K. M. Effects of urban heat islands on temperate forest trees and arthropods. Curr. Rep. 9 , 48–57 (2023).

Esperon-Rodriguez, M. et al. Climate change increases global risk to urban forests. Nat. Clim. Chang. 12 , 950–955 (2022).

Stewart, I. D. & Oke, T. R. Local climate zones for urban temperature studies. Bull. Am. Meteorol. Soc. 93 , 1879–1900 (2012).

Biardeau, L. T., Davis, L. W., Gertler, P. & Wolfram, C. Heat exposure and global air conditioning. Nat. Sustain. 3 , 25–28 (2020).

Davis, L., Gertler, P., Jarvis, S. & Wolfram, C. Air conditioning and global inequality. Glob. Environ. Change 69 , 102299 (2021).

Colelli, F. P., Wing, I. S. & Cian, E. D. Air-conditioning adoption and electricity demand highlight climate change mitigation–adaptation tradeoffs. Sci. Rep. 13 , 4413 (2023).

Sun, L., Chen, J., Li, Q. & Huang, D. Dramatic uneven urbanization of large cities throughout the world in recent decades. Nat. Commun. 11 , 5366 (2020).

Liu, D., Kwan, M.-P. & Kan, Z. Analysis of urban green space accessibility and distribution inequity in the City of Chicago. Urban For. Urban Green. 59 , 127029 (2021).

Hsu, A., Sheriff, G., Chakraborty, T. & Manya, D. Disproportionate exposure to urban heat island intensity across major US cities. Nat. Commun. 12 , 2721 (2021).

Zhao, L., Lee, X., Smith, R. B. & Oleson, K. Strong contributions of local background climate to urban heat islands. Nature 511 , 216–219 (2014).

Wu, S., Chen, B., Webster, C., Xu, B. & Gong, P. Improved human greenspace exposure equality during 21st century urbanization. Nat. Commun. 14 , 6460 (2023).

Zhao, J., Zhao, X., Wu, D., Meili, N. & Fatichi, S. Satellite-based evidence highlights a considerable increase of urban tree cooling benefits from 2000 to 2015. Glob. Chang. Biol. 29 , 3085–3097 (2023).

Article   CAS   PubMed   Google Scholar  

Nice, K. A., Coutts, A. M. & Tapper, N. J. Development of the VTUF-3D v1. 0 urban micro-climate model to support assessment of urban vegetation influences on human thermal comfort. Urban Clim. 24 , 1052–1076 (2018).

Meili, N. et al. An urban ecohydrological model to quantify the effect of vegetation on urban climate and hydrology (UT&C v1. 0). Geosci. Model Dev. 13 , 335–362 (2020).

Nesbitt, L., Meitner, M. J., Sheppard, S. R. & Girling, C. The dimensions of urban green equity: a framework for analysis. Urban For. Urban Green. 34 , 240–248 (2018).

Hedblom, M., Prévot, A.-C. & Grégoire, A. Science fiction blockbuster movies—a problem or a path to urban greenery? Urban For. Urban Green. 74 , 127661 (2022).

Norton, B. A. et al. Planning for cooler cities: a framework to prioritise green infrastructure to mitigate high temperatures in urban landscapes. Landsc. Urban Plan 134 , 127–138 (2015).

Medl, A., Stangl, R. & Florineth, F. Vertical greening systems—a review on recent technologies and research advancement. Build. Environ. 125 , 227–239 (2017).

Chen, B., Lin, C., Gong, P. & An, J. Optimize urban shade using digital twins of cities. Nature 622 , 242–242 (2023).

Pamukcu-Albers, P. et al. Building green infrastructure to enhance urban resilience to climate change and pandemics. Landsc. Ecol. 36 , 665–673 (2021).

Haaland, C. & van Den Bosch, C. K. Challenges and strategies for urban green-space planning in cities undergoing densification: a review. Urban For. Urban Green. 14 , 760–771 (2015).

Shafique, M., Kim, R. & Rafiq, M. Green roof benefits, opportunities and challenges—a review. Renew. Sust. Energ. Rev. 90 , 757–773 (2018).

Wang, J., Zhou, W. & Jiao, M. Location matters: planting urban trees in the right places improves cooling. Front. Ecol. Environ. 20 , 147–151 (2022).

Lan, T., Liu, Y., Huang, G., Corcoran, J. & Peng, J. Urban green space and cooling services: opposing changes of integrated accessibility and social equity along with urbanization. Sust. Cities Soc. 84 , 104005 (2022).

Wood, S. & Dupras, J. Increasing functional diversity of the urban canopy for climate resilience: Potential tradeoffs with ecosystem services? Urban For. Urban Green. 58 , 126972 (2021).

Wong, N. H., Tan, C. L., Kolokotsa, D. D. & Takebayashi, H. Greenery as a mitigation and adaptation strategy to urban heat. Nat. Rev. Earth Environ. 2 , 166–181 (2021).

United Nations. Department of economic and social affairs, population division. in The World’s Cities in 2018—Data Booklet (UN, 2018).

United Nations Development Programme (UNDP). Human Development Report 2019: Beyond Income, Beyond Averages, Beyond Today: Inequalities in Human Development in the 21st Century (United Nations Development Programme (UNDP), 2019)

Li, X. et al. Mapping global urban boundaries from the global artificial impervious area (GAIA) data. Environ. Res. Lett. 15 , 094044 (2020).

Stevens, F. R., Gaughan, A. E., Linard, C. & Tatem, A. J. Disaggregating census data for population mapping using random forests with remotely-sensed and ancillary data. PloS ONE 10 , e0107042 (2015).

Buchhorn, M. et al. Copernicus global land cover layers—collection 2. Remote Sens 12 , 1044 (2020).

Gillerot, L. et al. Forest structure and composition alleviate human thermal stress. Glob. Change Biol. 28 , 7340–7352 (2022).

Article   CAS   Google Scholar  

Hamada, S., Tanaka, T. & Ohta, T. Impacts of land use and topography on the cooling effect of green areas on surrounding urban areas. Urban For. Urban Green. 12 , 426–434 (2013).

Sun, X. et al. Quantifying landscape-metrics impacts on urban green-spaces and water-bodies cooling effect: the study of Nanjing, China. Urban For . Urban Green. 55 , 126838 (2020).

Zhang, Q., Zhou, D., Xu, D. & Rogora, A. Correlation between cooling effect of green space and surrounding urban spatial form: Evidence from 36 urban green spaces. Build. Environ. 222 , 109375 (2022).

Pesaresi, M., Politis, P. GHS-BUILT-H R2023A - GHS building height, derived from AW3D30, SRTM30, and Sentinel2 composite (2018) . European Commission, Joint Research Centre (JRC) https://doi.org/10.2905/85005901-3A49-48DD-9D19-6261354F56FE (2023).

Yamazaki, D. et al. A high‐accuracy map of global terrain elevations. Geophys. Res. Lett. 44 , 5844–5853 (2017).

Wessel, P. & Smith, W. H. A global, self‐consistent, hierarchical, high‐resolution shoreline database. J. Geophys. Res. Solid Earth 101 , 8741–8743 (1996).

Ren et al. climatic map studies: a review. Int. J. Climatol. 31 , 2213–2233 (2011).

Zhou, X. et al. Evaluation of urban heat islands using local climate zones and the influence of sea-land breeze. Sust. Cities Soc. 55 , 102060 (2020).

Zhou, W., Huang, G. & Cadenasso, M. L. Does spatial configuration matter? Understanding the effects of land cover pattern on land surface temperature in urban landscapes. Landsc. Urban Plan 102 , 54–63 (2011).

Muñoz Sabater, J. ERA5-Land monthly averaged data from 1981 to present . Copernicus Climate Change Service (C3S) Climate Data Store (CDS) https://doi.org/10.24381/cds.68d2bb30 (2019).

Abatzoglou, J. T., Dobrowski, S. Z., Parks, S. A. & Hegewisch, K. C. TerraClimate, a high-resolution global dataset of monthly climate and climatic water balance from 1958–2015. Sci. Data 5 , 1–12 (2018).

Kummu, M., Taka, M. & Guillaume, J. H. Gridded global datasets for gross domestic product and Human Development Index over 1990–2015. Sci. Data 5 , 1–15 (2018).

Zanaga, D. et al. ESA WorldCover 10 m 2020 v100. https://doi.org/10.5281/zenodo.5571936 (2021).

McNally, A. et al. A land data assimilation system for sub-Saharan Africa food and water security applications. Sci. Data 4 , 1–19 (2017).

Schaaf C., & Wang Z. MODIS/Terra+Aqua BRDF/Albedo Daily L3 Global - 500m V061 . NASA EOSDIS Land Processes Distributed Active Archive Center. https://doi.org/10.5067/MODIS/MCD43A3.061 (2021).

Lyapustin A., & Wang Y. MODIS/Terra+Aqua Land Aerosol Optical Depth Daily L2G Global 1km SIN Grid V061 . NASA EOSDIS Land Processes Distributed Active Archive Center. https://doi.org/10.5067/MODIS/MCD19A2.061 (2022).

Li, M., Wang, Y., Rosier, J. F., Verburg, P. H. & Vliet, J. V. Global maps of 3D built-up patterns for urban morphological analysis. Int. J. Appl. Earth Obs. Geoinf. 114 , 103048 (2022).

Google Scholar  

Elvidge, C. D., Baugh, K., Zhizhin, M., Hsu, F. C. & Ghosh, T. VIIRS night-time lights. Int. J. Remote Sens. 38 , 5860–5879 (2017).

Zhou, W. et al. Urban tree canopy has greater cooling effects in socially vulnerable communities in the US. One Earth 4 , 1764–1775 (2021).

Beck, H. E. et al. Present and future Köppen-Geiger climate classification maps at 1-km resolution. Sci. Data 5 , 1–12 (2018).

R. Core Team. R: A Language and Environment for Statistical Computing . R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/ (2023).

Fox J., & Weisberg S. An R Companion to Applied Regression 3rd edn (Sage, 2019). https://socialsciences.mcmaster.ca/jfox/Books/Companion/ .

Lefcheck, J. S. piecewiseSEM: Piecewise structural equation modelling in r for ecology, evolution, and systematics. Methods Ecol. Evol. 7 , 573–579 (2016).

Zeileis, A. _ineq: Measuring Inequality, Concentration, and Poverty_ . R package version 0.2-13. https://CRAN.R-project.org/package=ineq (2014).

Download references

Acknowledgements

We thank all the data providers. We thank Marten Scheffer for valuable discussion. C.X. is supported by the National Natural Science Foundation of China (Grant No. 32061143014). J.-C.S. was supported by Center for Ecological Dynamics in a Novel Biosphere (ECONOVO), funded by Danish National Research Foundation (grant DNRF173), and his VILLUM Investigator project “Biodiversity Dynamics in a Changing World”, funded by VILLUM FONDEN (grant 16549). W.Z. was supported by the National Science Foundation of China through Grant No. 42225104. T.M.L. and J.F.A. are supported by the Open Society Foundations (OR2021-82956). W.J.R. is supported by the funding received from Roger Worthington.

Author information

Authors and affiliations.

School of Life Sciences, Nanjing University, Nanjing, China

Yuxiang Li, Shuqing N. Teng & Chi Xu

Center for Ecological Dynamics in a Novel Biosphere (ECONOVO), Department of Biology, Aarhus University, Aarhus, Denmark

Jens-Christian Svenning

State Key Laboratory of Urban and Regional Ecology, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing, China

University of Chinese Academy of Sciences, Beijing, China

Beijing Urban Ecosystem Research Station, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing, China

School for Environment and Sustainability, University of Michigan, Ann Arbor, MI, USA

Global Systems Institute, University of Exeter, Exeter, UK

Jesse F. Abrams & Timothy M. Lenton

Department of Forest Ecosystems and Society, Oregon State University, Corvallis, OR, USA

William J. Ripple

Department of Environmental Science and Engineering, Fudan University, Shanghai, China

Department of Applied Ecology, North Carolina State University, Raleigh, NC, USA

Robert R. Dunn

You can also search for this author in PubMed   Google Scholar

Contributions

Y.L., S.N.T., R.R.D., and C.X. designed the study. Y.L. collected the data, generated the code, performed the analyses, and produced the figures with inputs from J.-C.S., W.Z., K.Z., J.F.A., T.M.L., W.J.R., Z.Y., S.N.T., R.R.D. and C.X. Y.L., S.N.T., R.R.D. and C.X. wrote the first draft with inputs from J.-C.S., W.Z., K.Z., J.F.A., T.M.L., W.J.R., and Z.Y. All coauthors interpreted the results and revised the manuscript.

Corresponding authors

Correspondence to Shuqing N. Teng , Robert R. Dunn or Chi Xu .

Ethics declarations

Competing interests.

The authors declare no competing interests.

Peer review

Peer review information.

Nature Communications thanks Chris Webster and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary information, peer review file, description of additional supplementary files, supplementary data 1, reporting summary, rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ .

Reprints and permissions

About this article

Cite this article.

Li, Y., Svenning, JC., Zhou, W. et al. Green spaces provide substantial but unequal urban cooling globally. Nat Commun 15 , 7108 (2024). https://doi.org/10.1038/s41467-024-51355-0

Download citation

Received : 06 December 2023

Accepted : 05 August 2024

Published : 02 September 2024

DOI : https://doi.org/10.1038/s41467-024-51355-0

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

• Open access
• Published: 03 September 2024

CT-based whole lung radiomics nomogram for identification of PRISm from non-COPD subjects

• TaoHu Zhou 1 , 2   na1 ,
• Yu Guan 1   na1 ,
• XiaoQing Lin 1 , 3   na1 ,
• XiuXiu Zhou 1 ,
• Liang Mao 4 ,
• YanQing Ma 5 ,
• Bing Fan 6 ,
• Jie Li 1 , 3 ,
• ShiYuan Liu 1 &

Respiratory Research volume  25 , Article number:  329 ( 2024 ) Cite this article

Metrics details

Preserved Ratio Impaired Spirometry (PRISm) is considered to be a precursor of chronic obstructive pulmonary disease. Radiomics nomogram can effectively identify the PRISm subjects from non-COPD subjects, especially when during large-scale CT lung cancer screening.

Totally 1481 participants (864, 370 and 247 in training, internal validation, and external validation cohorts, respectively) were included. Whole lung on thin-section computed tomography (CT) was segmented with a fully automated segmentation algorithm. PyRadiomics was adopted for extracting radiomics features. Clinical features were also obtained. Moreover, Spearman correlation analysis, minimum redundancy maximum relevance (mRMR) feature ranking and least absolute shrinkage and selection operator (LASSO) classifier were adopted to analyze whether radiomics features could be used to build radiomics signatures. A nomogram that incorporated clinical features and radiomics signature was constructed through multivariable logistic regression. Last, calibration, discrimination and clinical usefulness were analyzed using validation cohorts.

The radiomics signature, which included 14 stable features, was related to PRISm of training and validation cohorts ( p  < 0.001). The radiomics nomogram incorporating independent predicting factors (radiomics signature, age, BMI, and gender) well discriminated PRISm from non-COPD subjects compared with clinical model or radiomics signature alone for training cohort (AUC 0.787 vs. 0.675 vs. 0.778), internal (AUC 0.773 vs. 0.682 vs. 0.767) and external validation cohorts (AUC 0.702 vs. 0.610 vs. 0.699). Decision curve analysis suggested that our constructed radiomics nomogram outperformed clinical model.

Conclusions

The CT-based whole lung radiomics nomogram could identify PRISm to help decision-making in clinic.

Key message

Identifying PRISm subjects among non-COPD subjects, especially in the context of large-scale CT lung cancer screening, is currently a challenge.

In this retrospective, and multicentric study that included 1481 subjects, radiomics nomogram developed by integrating radiomics signature and clinical features achieved good performance for the identification of PRISm, with AUC of 0.787, 0.773 and 0.702 in training, internal and external validation cohort.

Radiomics nomogram, as a promising tool for identifying the PRISm from non-COPD subjects, hold great potential for guiding timely treatment and showing the added value of chest CT to evaluate the lung function status besides the morphological evaluation, especially during large-scale CT lung cancer screening.

Introduction

Chronic obstructive pulmonary disease (COPD) is a primarily factor causing morbidity and mortality globally, ranking the third leading cause of death globally and resulting in tremendous health care, social and economic burdens [ 1 , 2 , 3 ]. The COPD burden may be significantly increased in the future decades owing to rapid aging of Chinese population. According to recent demographic data from the National Bureau of Statistics of China, the proportion of the population aged 65 and above is expected to increase from 12.6% in 2020 to a projected 28.1% by 2050 [ 4 ]. This rapid aging trend could significantly heighten the COPD burden in China compared to other populations. A recent meta-analytic study revealed that the prevalence of COPD increases notably with age with a marked rise from 4.37% (95% CI 2.76% − 6.33%) among individuals aged 40–49 years to 24.03% (95% CI 20.04%-28.26%) in those aged 70 years and older [ 5 ]. Moreover, it has been reported China had the largest absolute economic burden of COPD in the world, China alone accounts for 83.5% of the economic losses in upper-middle-income countries [ 6 ]. Screening and identifying COPD early can prevent disease progression and reduce health and economic burdens.

Preserved ratio impaired spirometry (PRISm), also known as restrictive pattern or unclassified spirometry, is defined as a FEV1 of less than 80% predicted, despite a normal or preserved FEV1/forced vital capacity (FVC) ratio (≥ 0.70) [ 7 ]. PRISm can transition to normal, obstructive or restrictive spirometry over time [ 8 ]. Therefore, PRISm has been increasingly identified with prognostic significance [ 9 , 10 , 11 , 12 ]. Based on some population-based studies performed in Western populations, PRISm subjects are associated with an increased airflow limitation (AFL) rate and an increased mortality risk relative to subjects having normal lung function [ 9 , 10 , 11 , 12 ]. These observations are similar to those reported from the Asian region [ 13 ]. According to the community survey involving 3032 Japanese people during the 5-year follow-up [ 13 ], 31 with PRISm at the first visit showed an increased overall mortality and a higher probability of COPD progression compared with people showing normal spirometry. As a result, finding markers that can accurately identify PRISm and offer a foundation for early prognosis prediction is of great significance for enhancing the management of clinical subjects.

Despite PFTs remaining the gold standard for diagnosing PRISm, the utilization rate is relatively low. In China, only 6.7% of individuals over 40 have undergone PFTs, resulting in a significant number of undiagnosed cases due to limited accessibility [ 14 ]. Imaging, particularly chest CT scans, offer considerable advantages and potential. With the widespread adoption of lung cancer screening programs, not only the use of CT scans are increased but also CT could provide more detailed anatomical information. Results from large-scale screening studies such as NELSON and NLST have shown reduced mortality rates of lung cancer undergoing lung cancer CT screenings [ 15 ]. These advancements offer greater survival chances for cancer patients. If we can simultaneously screen for PRISm during lung cancer screenings could provide more timely medical interventions and more benefit for the population [ 16 ].

In recent years, radiomics has aroused increasing attention, which refers to the process in which medical images are converted in high-dimensional, mineable data through high-throughput quantitative feature extraction and data analysis to support decision-making [ 17 ]. Radiomics is adopted for identifying chest diseases and evaluating prognosis [ 18 , 19 , 20 , 21 , 22 ]. Recently, several imaging studies have focused on exploring those imaging features of PRISm and the significance of quantitative HRCT in early diagnosis. However, to our best knowledge, studies have not yet investigated the relationship between radiomics and PRISm. The purpose of the study was to investigate the performance of CT radiomics in identifying PRISm subjects among non-COPD subjects with one-stop CT screening.

Totally 1513 subjects with PFT in five hospitals were retrospectively recruited between February 2013 and December 2022. The trial was registered in Chinese Clinical Trial Registry on 29 March 2023 (Number: ChiCTR2300069929, URL: https://www.chictr.org.cn/showproj.html?proj=192439 ). Subjects were enrolled based on the following inclusion criteria: (1) both chest CT and PFT were performed in the same hospital; (2) the PFT to chest CT interval less than 2 weeks; (3) complete thin-section (< 2 mm) chest CT images; (4) the postbronchodilator FEV1/FVC ≥ 0.7. The exclusion criteria as follows: (1) co-morbid other thoracic disease (e.g., pneumonia, pulmonary atelectasis, lung nodules larger than 6 mm or masses, asthma, and pleural effusion); (2) concomitant malignant neoplasms; and (3) artifacts. Finally, 1481 subjects were included in the study. Among them, 1234 subjects from one hospital were randomly divided into training ( n  = 864) and internal validation cohort ( n  = 370) with the ratio of 7:3, using “caret” R package. Those from other four hospitals were classified in independent external validation cohort ( n  = 247). Figure  1 displays the subjects screening workflow. In the meanwhile, clinical basic information of the subjects such as age, sex, height, weight, BMI, and smoking status, was obtained based on electronic medical records system. The approval of the retrospective study was provided by the institutional review board of the leading hospital. Due to the retrospective nature, the informed consent was waived.

Flowchart for the selection of the study population

CT image acquisition and pulmonary function test

Table S1 shows the CT acquisition parameters and pulmonary function test apparatus in detail. Lung function was categorized in line with modified GOLD criteria and prior studies [ 12 , 23 ]. In this study, based on the PFT results, the non-COPD subjects were classified into the PRISm and normal spirometry groups for the training, internal validation and the independent external validation cohorts. Normal spirometry was defined as FEV1/FVC ≥ 0.70 and FEV1 ≥ 80% predicted; PRISm was defined as FEV1/FVC ≥ 0.70 and FEV1 < 80% predicted.

Whole lung CT segmentation, radiomics features extraction and selection

The pretrained CNN of U-Net structure was used to process chest CT images of each subject [ 24 ]. Firstly, the right and left lungs were automatically segmented using a publicly accessed deep-learning model, U-net (R231) ( https://github.com/JoHof/lungmask ), which has been trained based on different large-scale datasets including broad visual variability. Secondly, we merged the right and left lung into a combined region of interest (ROI) (Figure S1 ). Thirdly, an experienced chest radiologist with 8-year experience examined the segmentation outcome visually with the use of ITK-SNAP software (version 3.8.0, www.itksnap.org ). Inaccurate segmentation could be corrected manually with ITK-SNAP.

Prior to the extraction of radiomics features, three steps were utilized for image preprocessing. Firstly, linear interpolation was employed to resample images to 1 mm*1 mm*1 mm. Secondly, gray-level discretization was used for converting continuous images to discrete integer values and a bin width of 25 was used to reduce the effect of imaging noise. At last, wavelet and log image filters were adopted for eliminating mixed noise during image digitization and obtaining high- or low-frequency features. Pyradiomics software (version 3.0.1, https://pyradiomics.readthedocs.io/en/latest/ ) was adopted for extracting lung radiomics features. Totally 1218 features were obtained in each volume of interest with open-source package (pyradiomics), including first-order, gray-level cooccurrence matrix (GLCM), gray-level run length matrix (GLRLM), gray-level size zone matrix (GLSZM), gray-level dependence matrix (GLDM), and shape features. Radiomics features comply with the image biomarker standardization initiative (IBSI) [ 25 ], including standardization of acquisition and feature extraction, guidelines for annotation, segmentation, feature selection, model building and validation and clinical implementation. For the purpose of normalizing the features, the Z score method was adopted. In addition, the difference in the numerical scale was removed.

Optimal radiomics features were selected. Firstly, by eliminating redundancy with correlation coefficient > 0.90, we selected the optimal radiomic feature. Secondly, maximal redundancy minimal relevance (mRMR) algorithm was adopted for eliminating irrelevant or redundant features. mRMR has been demonstrated to be an efficient and reliable feature selection method for radiomics which can discover the optimal subset of features through considering the importance of features and the correlation between them. Least absolute shrinkage and selection operator (LASSO), which is the embedded approach, has been extensively applied to select high-dimensional radiomics features [ 26 ]. In addition, 10-fold cross-validation was carried out using penalty parameter and LASSO regression algorithm. The best feature dataset that had the lowest cross-validation binomial deviation was chosen, while nonzero coefficient was defined as selected feature weight, representing relation of features with PRISm. We used a linear combination of selected feature and coefficient vectors to calculate the Radscore of each subject. In addition, the construction of radiomics model was made.

Model, nomogram construction and performance evaluation

Three models were built including clinical, radiomic and the combined models. Statistically significant risk variables were identified through univariable logistic regression, which were then incorporated for multivariable regression to construct the clinical and combined models. Last, we constructed a nomogram to visualize the combined model, graphically evaluated the variable importance and calculate the prediction accuracy. AUCs (areas under the ROC curve) of those three models were compared using Delong test. Nomogram calibration was assessed through calibration curves (Hosmer-Lemeshow test). Nomogram clinical practicability was assessed by decision curves Analysis (DCA).

Statistical analysis

In statistical analysis, R software (version 4.2.2; http://www.Rproject.org ) and IBM SPSS Statistics (Version 26.0; IBM Corp., New York, USA) were used. Measurement data were represented through mean ± standard deviation. Continuous variables of normal distribution were evaluated with the use of Mann-Whitney/Wilcoxon nonparametric test. Categorical data were explored by chi-square test between groups. Independent predicting factors were determined from diverse clinical variables by multivariate logistic regression. P  < 0.05 represented statistical significance. “glmnet” package was used for LASSO analysis. In addition, we employed “caret” package for random division. “rms” package was applied for drawing calibration plot and conducting multivariate logistic regression. “pROC” package was used for drawing ROC (receiver operating characteristic curve) of radiomics signature. “rmda” package for DCA to evaluate net benefit, which is defined as the differential value of true positives proportion and false positives proportion, weighted by the relative harm of false positive and false negative results.

Demographic data and clinical model establishment

Table  1 displays basic demographic features of subjects. A total of 864 subjects with normal spirometry were in training cohort (500 males, 364 females; average age, 60.3 ± 12.7 years), 370 subjects were in internal validation cohort (207 males, 163 females; average age, 60.9 ± 11.8 years) and 247 subjects were in external validation cohort (163 males, 84 females; average age, 61.6 ± 12.4 years). The rates of PRISm subjects were 37.5% (324 of 864), 42.2% (156 of 370), and 40.5% (100 of 247) of training, internal, and external validation cohorts, respectively. Three cohorts differed significantly regarding gender and BMI ( p  < 0.05). Significant differences were found in training and internal validation cohorts with respect to age and smoking status ( p  < 0.001), but not in external cohort. In addition, age distribution was not of significant difference between two validation cohorts. Table  1 displays results of univariate and multivariate logistic regression. Age, BMI, and gender identified from multivariable regression were included to develop the clinical model. Figure  2 presents ROC curves for clinical model. The corresponding AUCs were 0.675 (95% CI: 0.637, 0.712), 0.682 (95% CI: 0.627, 0.737) and 0.610 (95% CI: 0.538, 0.682) in training, internal and external validation cohorts, respectively.

Diagnostic performance of the clinical factors model, radiomics signature, and radiomics nomogram was assessed and compared through ROC curves in the training ( A ), internal validation ( B ) and external validation ( C ) cohorts. ROC = receiver operating characteristics; AUC = area under the receiver operating characteristic curve

Radiomics feature selection and radiomics signature

Among 1218 radiomics features extracted from chest CT images, 245 features exhibited high stability, and then were decreased to 30 features through minimum redundancy maximum relevancy. Finally, LASSO was conducted to select features (Fig.  3 A, B), among which, those 14 features with highest importance were retained, as shown in Fig.  3 C. The calculation formulas for Radscore are listed in Supplementary Results .

Radiomics feature selection by using the least absolute shrinkage and selection operator (LASSO) logistic regression. ( A ) Selection of the tuning parameter (λ) in the LASSO model via 10-fold cross-validation based on minimum criteria. Binomial deviances from the LASSO regression cross-validation model are plotted as a function of log(λ). The y-axis shows binomial deviances and the lower x-axis the log(λ). Numbers along the upper x-axis indicate the average number of predictors. Red dots indicate average deviance values for each model with a given λ, and vertical bars through the red dots indicate the upper and lower values of the deviances. The vertical black lines define the optimal values of λ, where the model provides its best fit to the data. ( B ) The coefficients have been plotted vs. log(λ). ( C ) The 14 features with nonzero coefficients are shown in the plot

Figure  2 displays ROC curves for radiomics signature. The AUCs of our radiomics signature were 0.778 (95% CI: 0.746, 0.810), 0.767 (95% CI: 0.718, 0.815) and 0.699 (95% CI: 0.633, 0.766) for training, internal and external validation cohorts, respectively.

Radiomics nomogram construction and model performance evaluation

Radscore and clinical features were incorporated to develop the radiomics nomogram for training cohort ( Fig.  4 A). The calibration curve of the radiomics nomogram showed good consistence between the predicted and expected probabilities for PRISm (Fig.  4 B, C, D). Meanwhile, upon Hosmer–Lemeshow test, their P -values were 0.9995, 0.4521, and 0.1049 for training, internal, and external validation cohorts, respectively, which revealed relatively excellent agreement between the nomogram prediction and the actual observation. Figure  2 shows ROC curves for radiomics nomogram. AUCs were used as an index of diagnostic accuracy; a higher AUC reflects greater accuracy. Its AUC, sensitivity, specificity, and accuracy were 0.787 (95% CI: 0.756, 0.818), 75.0%, 68.5%, and 70.9%; 0.773 (95%CI: 0.725, 0.821), 64.1%, 78.8%, and 71.6%; and 0.702 (95%CI: 0.636,0.767), 63.9%, 68.0%, and 65.6% in training, internal and external validation cohorts, separately.

Radiomics nomogram, calibration curves and DCA curves. ( A ) The radiomics nomogram, combining age, BMI, gender and Radscore, was developed in the training cohort. (B–D) The nomogram calibration curves in training ( B ), internal validation ( C ), and external validation ( D ) cohorts. Calibration curves indicate the goodness-of-ft of the model. ( E ) Decision curve analysis for different models

Table  2 displays model diagnostic accuracy. Comparison between ROC curves was performed by the DeLong test. The Delong test showed that there was a statistically significant difference in AUCs between the radiomics nomogram and the clinical model (Z = 6.64, p  < 0.001; Z = 3.72, p  < 0.001; and Z = 2.46, p  = 0.014 in the training, internal validation, and external validation cohort, respectively). There is a difference between the radiomics signature and radiomics nomogram (Z = -2.01, p  = 0.044) in the training cohort. But no significant difference (Z = -0.84, p  = 0.401 and Z = 0.18, p  = 0.855 in internal validation, and external validation cohort, respectively) between the radiomics nomogram and radiomics signature. The correlation between radiomics signature and clinical model was the moderate in the training and internal validation cohorts ( R  = 0.4). Correlations in the external validation cohort was weak ( R  = 0.2).

Decision curve analysis was used to evaluate the clinical practicability of the nomogram prediction model ( Fig.  4 E ) . The results showed that the nomogram obtained more benefit than the “treat all,” “treat none,” and the clinical model when the threshold probability was in the range of 4–70%. An example of the nomogram in use is shown in Fig.  5 . Similar to the points scoring system, we assigned points for each predictor of PRISm and then equated these predictors with the risk of PRISm. We can read the top score scale upward from the predictors to determine the points score associated with patient BMI, age, gender and the Radscore. Once a score has been assigned to each predictor, an overall score is calculated. Then, the total score is converted to the probability of PRISm by reading the associated probability of PRISm from the total point scale.

An example of the nomogram in clinical practice. The nomogram was used to calculate the scoring process of risk of PRISm. ( A ) Thin-section chest CT image of a 49-year-old normal female subject. Her Clinical features were analyzed as follows: BMI = 19.5 kg/m2, Radscore = -2.64. The nomogram showed that this patient had a total of 174 points after summing all points, which corresponds to a close to 4.00% probability of PRISm. ( B ) Thin-section chest CT image of a 43-year-old male subject. His clinical features were analyzed as follows: BMI = 30.80 kg/m2, Radscore = 4.30. The nomogram showed that this patient had a total of 225 points after summing all points, which corresponds to a close to 96.9% probability of PRISm

Identifying subjects with PRISm is of great importance to verify the early, effective, and individualized decision-making in the prevention of COPD, because many PRISm would progress into COPD. In the present study, the radiomics nomogram incorporating clinical factors and radiomics signature was established and verified to identify PRISm subjects based on whole lung CT radiomics. The radiomics nomogram proposed in the current work exhibited favorable discrimination in training cohort (AUC, 0.787), internal validation cohort (AUC, 0.773) and external validation cohort (AUC, 0.702), outperforming radiomics signature (training, 0.778; internal validation, 0.767; external validation, 0.699) and clinical factor model (training, 0.675; internal validation, 0.682; external validation, 0.610).

The incidence and disease burden of COPD is high in China, the overall pulmonary function detection rate was still at a low level, and many people have been underdiagnosed. In contrast, the popularity of chest CT is very high, especially with the large-scale chest CT screening for lung cancer. Moreover, more and more community health service centers will be equipped with CT. Therefore, the most important clinical scenario is for the large-scale lung cancer screening population that usually does not perform PFT, and many people who were high risk (most likely to develop COPD) can be found through our model prediction, which can help enhance the early intervention of PRISm, reduce the social-economic burden and improve the patient’s life quality. Many clinical factors have been explored in PRISm. It has been found female sex, old age, smoking, and extreme weight were related to PRISm [ 27 ]. The former and current smokers were examined in one cross-sectional and follow-up study of COPDGene [ 10 , 28 ], as a result, PRISm patients showed the increased BMI compared with COPD patients and normal subjects, while persistent smoking independently predicted the reduced life quality of COPD patients. Many studies suggest that age and BMI are imperative risk factors of PRISm [ 7 ]. The older cohorts may show an increased impaired spirometry rate, particularly through the longitudinal follow-up. BMI can induce the risk of PRISm risk through the distinct pathway, including inflammatory and metabolic effects of adipose tissue [ 29 ]. Previous population-based studies suggest increased risk of restrictive pattern among females [ 30 , 31 ]. In our study, age, sex and BMI were selected as independent predictors for PRISm subjects, Table  3 showed that female subjects with increased age and BMI are more likely to be PRISm, which was consistent with previous studies.

To the best of our knowledge, relevant studies are few on the identification of PRISm with CT-based methods. Wei et al. [ 32 ] evaluated the CT-based quantitative features with an in-house system, and found that lung capacity, emphysema index, and airway wall area did not predict intermediate-stage chronic bronchitis that progresses from normal lung function to early COPD. Moreover, their study did not evaluate CT textural features and relevant clinical factors. To date, there are rare study for the identification of PRISm population using radiomics. The CT-based radiomics nomogram is established by integrating clinical factors and radiomics signature for identifying PRISm in our study. It is very difficult to identify the proper margin of the diffuse lung lesions. Thus, the full-automatic lung lobe segmentation method was performed using U-Net, which has been proven efficacy in pulmonary disease, especially pulmonary diffuse disease [ 33 , 34 , 35 , 36 ]. This radiomics signature included 14 radiomics features, which well distinguished PRISm subjects from normal spirometry subjects, and the performances were high in training (0.776 [95%CI, 0.746–0.810]), internal (0.767 [95%CI, 0.718–0.815]) and external validation (0.699 [95%CI, 0.633–0.766]) cohorts. In our radiomics signature, the majority of the features were transformed by wavelet filter, splitting imaging data into various different frequency components on three axis of the whole lung region [ 37 ]. This suggests that wavelet features probably interpret spatial heterogeneity in whole lung regions at multiple scales. In addition, the constructed radiomics signature model was combined with the clinical factors. Lu et al. predicted PRISm from the normal by the combination of CT quantitative parameters, as well as clinical features with an AUC of 0.786 [ 38 ]. Our study made a greater performance sightly (AUC = 0.787).

The constructed novel PRISm prediction nomogram was further evaluated by a decision curve to clarify the clinical utility, which could offer insight into clinical outcomes on the basis of threshold probability, from which the net benefit could be derived [ 39 , 40 ]. The results showed that if the threshold probability of a patient is > 4%, the application of our constructed radiomics nomogram in predicting PRISm was more beneficial in relative to the treat-none or treat-all-patients scheme. The present novel nomogram provides an important quantitative indicator and reference for the decision-making and management of treatment regimens for PRISm subjects. The new approach sheds more lights on clinical outcomes according to threshold probability, and net benefits were obtained on this basis [ 41 ].

However, this study still has the following limitations. First, owing to the retrospective and multi-institutional nature of the current work, CT acquisition parameters and reconstruction techniques were not in consistence. But, we use techniques such as regularization, normalization, and resampling to improve the performance of CT images, thereby enhancing the accuracy and reliability of diagnosis. Second, common CT quantitative parameters can be measured to provide more information on pulmonary lesions, such as air trapping, pulmonary vascular disease and so on. Therefore, in the future, we will incorporate these common quantitative features to optimize our prediction model. Thirdly, to keep pace with the advances in technology, other advanced deep-learning algorithm should be applied in our further studies. Fourth, we have excluded lung cancer patients in this study; however, in future research, we will include lung cancer patients and apply whole lung radiomics to distinguish whether they have PRISm or not.

To sum up, the CT radiomics model incorporating clinical factors and radiomics signature is established and validated to identify PRISm in non-COPD subjects. The radiomics approach may be helpful to delay initiation COPD progression.

Data availability

No datasets were generated or analysed during the current study.

Abbreviations

Chronic obstructive pulmonary disease

Global initiative for chronic obstructive lung disease

Pulmonary function test

The ratio of forced expiratory volume in 1 s to forced vital capacity

Preserved Ratio Impaired Spirometry

Airflow limitation

• Computed tomography

Region of interest

Body mass index

Gray level cooccurrence matrix

Gray level size zone matrix

Gray level dependence matrix

Maximal redundancy minimal relevance

Least absolute shrinkage and selection operator

Decision curves analysis

The area under the curve

The receiver operating characteristic curve

Yin P, Wu J, Wang L, et al. The Burden of COPD in China and its provinces: findings from the global burden of Disease Study 2019. Front Public Health. 2022;10:859499. https://doi.org/10.3389/fpubh.2022.859499 .

Article   PubMed   PubMed Central   Google Scholar  

Yadav AK, Gu W, Zhang T, Xu X, Yu L. Current perspectives on Biological Therapy for COPD. Copd. 2023;20(1):197–209. https://doi.org/10.1080/15412555.2023.2187210 .

Article   Google Scholar  

Wang C, Xu J, Yang L, et al. Prevalence and risk factors of chronic obstructive pulmonary disease in China (the China Pulmonary Health [CPH] study): a national cross-sectional study. Lancet. 2018;28(10131):1706–17. https://doi.org/10.1016/s0140-6736(18)30841-9 .

China. https://www.stats.gov.cn/

Al Wachami N, Guennouni M, Iderdar Y, et al. Estimating the global prevalence of chronic obstructive pulmonary disease (COPD): a systematic review and meta-analysis. BMC Public Health. 2024;25(1):297. https://doi.org/10.1186/s12889-024-17686-9 .

Chen S, Kuhn M, Prettner K, et al. The global economic burden of chronic obstructive pulmonary disease for 204 countries and territories in 2020-50: a health-augmented macroeconomic modelling study. Lancet Glob Health. 2023;11(8):e1183–93. https://doi.org/10.1016/s2214-109x(23)00217-6 .

Article   CAS   Google Scholar  

Higbee DH, Granell R, Davey Smith G, Dodd JW. Prevalence, risk factors, and clinical implications of preserved ratio impaired spirometry: a UK Biobank cohort analysis. Lancet Respir Med. 2022;10(2):149–57. https://doi.org/10.1016/s2213-2600(21)00369-6 .

Wan ES. The clinical spectrum of PRISm. Am J Respir Crit Care Med. 2022;1(5):524–5. https://doi.org/10.1164/rccm.202205-0965ED .

Park HJ, Byun MK, Rhee CK, Kim K, Kim HJ, Yoo KH. Significant predictors of medically diagnosed chronic obstructive pulmonary disease in patients with preserved ratio impaired spirometry: a 3-year cohort study. Respir Res. 2018;24(1):185. https://doi.org/10.1186/s12931-018-0896-7 .

Wan ES, Fortis S, Regan EA, et al. Longitudinal phenotypes and mortality in preserved ratio impaired spirometry in the COPDGene Study. Am J Respir Crit Care Med. 2018;1(11):1397–405. https://doi.org/10.1164/rccm.201804-0663OC .

Wan ES, Hokanson JE, Regan EA, et al. Significant spirometric transitions and preserved ratio impaired Spirometry among ever smokers. Chest. 2022;161(3):651–61. https://doi.org/10.1016/j.chest.2021.09.021 .

Wijnant SRA, De Roos E, Kavousi M, et al. Trajectory and mortality of preserved ratio impaired spirometry: the Rotterdam Study. Eur Respir J. 2020;55(1). https://doi.org/10.1183/13993003.01217-2019 .

Washio Y, Sakata S, Fukuyama S, et al. Risks of mortality and airflow limitation in Japanese individuals with preserved ratio impaired spirometry. Am J Respir Crit Care Med. 2022;1(5):563–72. https://doi.org/10.1164/rccm.202110-2302OC .

Tong H, Cong S, Fang LW, et al. [Performance of pulmonary function test in people aged 40 years and above in China, 2019–2020]. Zhonghua Liu Xing Bing Xue Za Zhi. 2023;10(5):727–34. https://doi.org/10.3760/cma.j.cn112338-20230202-00051 .

Oudkerk M, Liu S, Heuvelmans MA, Walter JE, Field JK. Lung cancer LDCT screening and mortality reduction - evidence, pitfalls and future perspectives. Nat Rev Clin Oncol. 2021;18(3):135–51. https://doi.org/10.1038/s41571-020-00432-6 .

Sunyi Zheng, Peter MA, van Ooijen OM. Lung Cancer Screening and Nodule Detection: the role of Artificial Intelligence Artificial Intelligence in cardiothoracic imaging. 2020:459. https://doi.org/10.1007/978-3-030-92087-6_43

Gillies RJ, Kinahan PE, Hricak H. Radiomics: images are more than pictures, they are data. Radiol. 2016;278(2):563–77. https://doi.org/10.1148/radiol.2015151169 .

Huang W, Deng H, Li Z, et al. Baseline whole-lung CT features deriving from deep learning and radiomics: prediction of benign and malignant pulmonary ground-glass nodules. Front Oncol. 2023;13:1255007. https://doi.org/10.3389/fonc.2023.1255007 .

Huang W, Zhang H, Ge Y, et al. Radiomics-based machine learning methods for volume doubling time prediction of Pulmonary Ground-glass nodules with baseline chest computed Tomography. J Thorac Imaging. 2023;1(5):304–14. https://doi.org/10.1097/rti.0000000000000725 .

Tu W, Sun G, Fan L, et al. Radiomics signature: a potential and incremental predictor for EGFR mutation status in NSCLC patients, comparison with CT morphology. Lung Cancer. 2019;132:28–35. https://doi.org/10.1016/j.lungcan.2019.03.025 .

Wang Y, Lyu D, Fan L, Liu S. Advances in the prediction of spread through air spaces with imaging in lung cancer: a narrative review. Transl Cancer Res. 2023;31(3):624–30. https://doi.org/10.21037/tcr-22-2593 .

Zhou T, Tu W, Dong P et al. CT-Based Radiomic Nomogram for the Prediction of Chronic Obstructive Pulmonary Disease in Patients with Lung cancer. Acad Radiol. 14. 2023; https://doi.org/10.1016/j.acra.2023.03.021

Agustí A, Celli BR, Criner GJ, et al. Global Initiative for Chronic Obstructive Lung Disease 2023 Report: GOLD Executive Summary. Eur Respir J. 2023;61(4). https://doi.org/10.1183/13993003.00239-2023 .

Hofmanninger J, Prayer F, Pan J, Röhrich S, Prosch H, Langs G. Automatic lung segmentation in routine imaging is primarily a data diversity problem, not a methodology problem. Eur Radiol Exp. 2020;20(1):50. https://doi.org/10.1186/s41747-020-00173-2 .

Yang K, Yang Y, Kang Y, et al. The value of radiomic features in chronic obstructive pulmonary disease assessment: a prospective study. Clin Radiol. 2022;77(6):e466–72. https://doi.org/10.1016/j.crad.2022.02.015 .

Remeseiro B, Bolon-Canedo V. A review of feature selection methods in medical applications. Comput Biol Med. 2019;112:103375. https://doi.org/10.1016/j.compbiomed.2019.103375 .

Guerra S, Carsin AE, Keidel D, et al. Health-related quality of life and risk factors associated with spirometric restriction. Eur Respir J. 2017;49(5). https://doi.org/10.1183/13993003.02096-2016 .

Wan ES, Castaldi PJ, Cho MH, et al. Epidemiology, genetics, and subtyping of preserved ratio impaired spirometry (PRISm) in COPDGene. Respir Res. 2014;6(1):89. https://doi.org/10.1186/s12931-014-0089-y .

Maclay JD, MacNee W. Cardiovascular disease in COPD: mechanisms. Chest. 2013;143(3):798–807. https://doi.org/10.1378/chest.12-0938 .

Guerra S, Sherrill DL, Venker C, Ceccato CM, Halonen M, Martinez FD. Morbidity and mortality associated with the restrictive spirometric pattern: a longitudinal study. Thorax. 2010;65(6):499–504. https://doi.org/10.1136/thx.2009.126052 .

Mannino DM, McBurnie MA, Tan W, et al. Restricted spirometry in the Burden of Lung Disease Study. Int J Tuberc Lung Dis. 2012;16(10):1405–11. https://doi.org/10.5588/ijtld.12.0054 .

Wei X, Ding Q, Yu N, et al. Imaging Features of Chronic Bronchitis with preserved ratio and impaired spirometry (PRISm). Lung. 2018;196(6):649–58. https://doi.org/10.1007/s00408-018-0162-2 .

Yang Y, Li W, Guo Y, et al. Early COPD risk decision for adults aged from 40 to 79 years based on lung Radiomics features. Front Med (Lausanne). 2022;9:845286. https://doi.org/10.3389/fmed.2022.845286 .

Article   PubMed   Google Scholar  

Yang Y, Li W, Guo Y, et al. Lung radiomics features for characterizing and classifying COPD stage based on feature combination strategy and multi-layer perceptron classifier. Math Biosci Eng. 2022;25(8):7826–55. https://doi.org/10.3934/mbe.2022366 .

Yang Y, Li W, Kang Y, et al. A novel lung radiomics feature for characterizing resting heart rate and COPD stage evolution based on radiomics feature combination strategy. Math Biosci Eng. 2022;17(4):4145–65. https://doi.org/10.3934/mbe.2022191 .

Yang Y, Wang S, Zeng N, et al. Lung Radiomics features selection for COPD Stage classification based on Auto-Metric graph neural network. Diagnostics (Basel). 2022;20(10). https://doi.org/10.3390/diagnostics12102274 .

Wilson R, Devaraj A. Radiomics of pulmonary nodules and lung cancer. Transl Lung Cancer Res. 2017;6(1):86–91. https://doi.org/10.21037/tlcr.2017.01.04 .

Lu J, Ge H, Qi L, et al. Subtyping preserved ratio impaired spirometry (PRISm) by using quantitative HRCT imaging characteristics. Respir Res. 2022;11(1):309. https://doi.org/10.1186/s12931-022-02113-7 .

Localio AR, Goodman S. Beyond the usual prediction accuracy metrics: reporting results for clinical decision making. Ann intern med. 2012;157(4):294-5. https://doi.org/10.7326/0003-4819-157-4-201208210-00014

Van Calster B, Vickers AJ. Calibration of risk prediction models: impact on decision-analytic performance. Med Decis Mak. 2015;35(2):162–9. https://doi.org/10.1177/0272989x14547233 .

Balachandran VP, Gonen M, Smith JJ, DeMatteo RP. Nomograms in oncology: more than meets the eye. Lancet Oncol. 2015;16(4):e173–80. https://doi.org/10.1016/s1470-2045(14)71116-7 .

Download references

Acknowledgements

We thank the colleagues in our department for their help in our study.

This work was supported by the National Key Research and Development Program of China (2022YFC2010002, 2022YFC2010000 and 2022YFC2010005), the National Natural Science Foundation of China (82171926, 81930049 and 82430065), the Medical Imaging Database Construction Program of National Health Commission (YXFSC2022JJSJ002), the Clinical Innovative Project of Shanghai Changzheng Hospital (2020YLCYJY24), the Program of Science and Technology Commission of Shanghai Municipality (21DZ2202600), and the Shanghai Sailing Program (20YF1449000).

Author information

TaoHu Zhou, Yu Guan and XiaoQing Lin contributed equally to this work.

Authors and Affiliations

Department of Radiology, Second Affiliated Hospital of Naval Medical University, No. 415 Fengyang Road, Shanghai, 200003, China

TaoHu Zhou, Yu Guan, XiaoQing Lin, XiuXiu Zhou, Jie Li, ShiYuan Liu & Li Fan

School of Medical Imaging, Shandong Second Medical University, Weifang, 261053, Shandong, China

College of Health Sciences and Engineering, University of Shanghai for Science and Technology, No.516 Jungong Road, Shanghai, 200093, China

XiaoQing Lin & Jie Li

Department of Medical Imaging, Affiliated Hospital of Ji Ning Medical University, Ji Ning, 272000, China

Department of Radiology, Zhejiang Provincial People’s Hospital, Affiliated People’s Hospital of Hangzhou Medical College, Hangzhou, ZJ, China

Department of Radiology, Jiangxi Provincial People’s Hospital, The First Affiliated Hospital of Nanchang Medical College, Nanchang, China

You can also search for this author in PubMed   Google Scholar

Contributions

TaoHu Zhou and Li Fan had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Yu Guan and XiaoQing Lin contributed equally to this work. Concept and design: TaoHu Zhou and Li Fan. Acquisition, analysis, or interpretation of data: XiuXiu Zhou, Liang Mao, YanQing Ma, Bing Fan, Jie LiDrafting of the manuscript: TaoHu Zhou, Yu Guan and XiaoQing Lin. Statistical analysis: TaoHu Zhou. Obtained funding: ShiYuan Liu, Li Fan. Supervision: ShiYuan Liu, Li Fan. Image processing: TaoHu Zhou.

Corresponding author

Correspondence to Li Fan .

Ethics declarations

Ethics approval and consent to participate.

This retrospective study involving human participants were reviewed and approved by the ethics committee of Second Affiliated Hospital of Naval Medical University. Patient informed consent was waived.

Consent for publication

No individual participant data is reported that would require consent to publish from the participant (or legal parent or guardian for children).

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/ .

Reprints and permissions

About this article

Cite this article.

Zhou, T., Guan, Y., Lin, X. et al. CT-based whole lung radiomics nomogram for identification of PRISm from non-COPD subjects. Respir Res 25 , 329 (2024). https://doi.org/10.1186/s12931-024-02964-2

Download citation

Received : 11 March 2024

Accepted : 28 August 2024

Published : 03 September 2024

DOI : https://doi.org/10.1186/s12931-024-02964-2

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Respiratory Research

ISSN: 1465-993X

• General enquiries: [email protected]