hypothesis political science examples

View sidebar

A Political Science Guide

For students, researchers, and others interested in doing the work of political science, formulating/extracting hypotheses.

Formulating hypotheses, which are defined as propositions set forth to explain a group of facts or phenomena, is a fundamental component to any research scholarship. Hypotheses lay out the central arguments that will be tested and either verified or rejected in the body of a paper. Papers may address multiple competing or supporting hypotheses in order to account for the full spectrum of explanations that could account for the phenomenon being studied. As such, hypotheses often include statements about a presumed impact of an independent variable on a dependent variable.

Hypotheses should not emanate from preconceived perceptions about a given relationship between variables, but rather should come about as a product of research. Thus, hypotheses should be formed after developing an understanding of the relevant literature to a given topic rather than before conducting research. Beginning research with a specific argument in mind can lead to discounting other evidence that could either run counter to this preconceived argument or could point to other potential explanations.

There are a number of different types of hypotheses utilized in political science research:

Null hypothesis: states that there is no relationship between two concepts
Correlative hypothesis: states that there is a relationship, between two or more concepts or variables, but doesn’t specify the nature of a relationship
Directional hypothesis: states the nature of the relationship between concepts or variables. These types of relationships can include positive, negative (inverse), high or low levels of influence, etc.
Causal hypothesis: states that one variable causes the other

A good hypothesis should be both correlative and directional and most hypotheses in political science research will also be causal, asserting the impact of an independent variable on a dependent variable.

There are a number of additional considerations that must be taken into account in order to make a hypothesis as strong as possible:

Hypotheses must be falsifiable , that is able to be empirically tested. They cannot attribute causation to something like a supernatural entity whose existence can neither be proven nor denied.
Hypotheses must be internally consistent , that is that they must be proving what they claim to be proving and must not contain any logical or analytical contradiction
Hypotheses must have clearly defined outcomes (dependent variables) that are both dependent and vary based on the dependent variable.
Hypotheses must be general and should aim to explain as much as possible with as little as possible. As such, hypotheses should have as few exceptions as possible and should not rely on amorphous concepts like ‘national interest.’
Hypotheses must be empirical statements that are propositions about relationships that exist in the real world.
Hypotheses must be plausible (there must be a logical reason why they might be true) and should be specific (the relationship between variables must be expressed as explicitly as possible) and directional.
Fearon, James D. 1991. Counterfactuals and Hypothesis Testing in Political Science . World Politics 43 (2): 169-195.

Abstract : “Scholars in comparative politics and international relations routinely evaluate causal hypotheses by referring to counterfactual cases where a hypothesized causal factor is supposed to have been absent. The methodological status and the viability of this very common procedure are unclear and are worth examining. How does the strategy of counterfactual argument relate, if at all, to methods of hypothesis testing based on the comparison of actual cases, such as regression analysis or Mill’s Method of Difference? Are counterfactual thought experiments a viable means of assessing hypotheses about national and international outcomes, or are they methodologically invalid in principle? The paper addresses the first question in some detail and begins discussion of the second. Examples from work on the causes of World War I, the nonoccurrence of World War III, social revolutions, the breakdown of democratic regimes in Latin America, and the origins of fascism and corporatism in Europe illustrate the use, problems and potential of counterfactual argument in small-N-oriented political science research.” – Jstor.org

King, Gary, Robert Owen Keohane, and Sidney Verba. 1994. Designing social inquiry: scientific inference in qualitative research. Princeton, NJ: Princeton University Press.
Palazzolo, David and Dave Roberts. 2010. What is a Good Hypothesis? University of Richmond Writing Center.

Contributor: Harrison Polans

updated July 12, 2017 – MN

University Libraries

Psci 3300: introduction to political research.

Library Accounts
Selecting a Topic for Research
From Topic to Research Question
From Question to Theories, Hypotheses, and Research Design
Annotated Bibliographies
The Literature Review
Search Strategies for Ann. Bibliographies & Lit. Reviews
Find PSCI Books for Ann. Bibliographies & Lit. Reviews
Databases & Electronic Resources for Your Lit. Review
Methods, Data Analysis, Results, Limitations, and Conclusion
Finding Data and Statistics for the Data Analysis
Citing Sources for the Reference Page

Need Help with Basic and Advance Research?

Visit the Basic and Advanced Library Research Guide to learn more about the library.

If you select "no," please send me an email so I can improve this guide.

Hypothesis in Political Science

"A generalization predicting that a relationship exists between variables. Many generalizations about politics are a sort of folklore. Others proceed from earlier work carried out by social scientists. Within the social sciences most statements about behaviour relate to large groups of people. Hence, testing any hypothesis in the field of political science will involve statistical method. It will be dealing with probabilities.

To test a hypothesis one must pose a null hypothesis. If we wanted to test the validity of the common generalization, 'manual workers tend to vote for the Labour Party' we would begin by assuming the statement was untrue. The investigation would require a sample survey in which manual workers were identified and questions put to them. It would need to be done in several constituencies in different parts of the country. Having collated the data we would use the evidence to test the null hypothesis, employing statistical techniques to assess the probability of acquiring such data if the null hypothesis were correct. These techniques are known as 'significance tests'. They estimate the probability that the rejection of a null hypothesis is a mistake. If the statistical tests indicates that the odds against it being a mistake are 1000 to one, then this is stated as a '.001 level of significance'.

The fact that the research showed that it was highly likely that manual workers 'tend' to vote for the Labour vote would not satisfy most political scientists. They also want to understand those who did not. Consequently much more work would need to be done to refine the hypothesis and define the tendency with more accuracy. Whatever the case, a hypothesis in the social sciences about a group or socio-demographic category can never tell us about the behaviour of an individual in that group or category."

Hypothesis. (1999). In F. Bealey. The Blackwell Dictionary of Political Science , Oxford, United Kingdom: Blackwell Publishers.

What a Quantitative Research Design?

Quantitative research studies produce results that can be used to describe or note numerical changes in measurable characteristics of a population of interest; generalize to other, similar situations; provide explanations of predictions; and explain causal relationships. The fundamental philosophy underlying quantitative research is known as positivism, which is based on the scientific method of research. Measurement is necessary if the scientific method is to be used. The scientific method involves an empirical or theoretical basis for the investigation of populations and samples. Hypotheses must be formulated, and observable and measurable data must be gathered. Appropriate mathematical procedures must be used for the statistical analyses required for hypothesis testing.

Quantitative methods depend on the design of the study (experimental, quasi-experimental, non-experimental). Study design takes into account all those elements that surround the plan for the investigation, such as research question or problem statement, research objectives, operational definitions, scope of inferences to be made, assumptions and limitations of the study, independent and dependent variables, treatment and controls, instrumentation, systematic data collection actions, statistical analysis, time lines, and reporting procedures. The elements of a research study and experimental, quasi-experimental, and nonexperimental designs are discussed here.

Elements of Quantitative Design

Problem statement.

First, an empirical or theoretical basis for the research problem should be established. This basis may emanate from personal experiences or established theory relevant to the study. From this basis, the researcher may formulate a research question or problem statement.

Operational Definitions

Operational definitions describe the meaning of specific terms used in a study. They specify the procedures or operations to be followed in producing or measuring complex constructs that hold different meanings for different people. For example, intelligence may be defined for research purposes by scores on the Stanford-Binet Intelligence Scale.

Population and Sample

Quantitative methods include the target group (population) to which the researcher wishes to generalize and the group from which data are collected (sample). Early in the planning phase, the researcher should determine the scope of inference for results of the study. The scope of inference pertains to populations of interest, procedures used to select the sample(s), method for assigning subjects to groups, and the type of statistical analysis to be conducted.

Formulation of Hypotheses

Complex questions to compare responses of two or more groups or show relationships between two or more variables are best answered by hypothesis testing. A hypothesis is a statement of the researcher's expectations about a relationship between variables.

Hypothesis Testing

Statements of hypotheses may be written in the alternative or null form. A directional alternative hypothesis states the researcher's predicted direction of change, difference between two or more sample means, or relationship among variables. An example of a directional alternative hypothesis is as follows:

Third-grade students who use reading comprehension strategies will score higher on the State Achievement Test than their counterparts who do not use reading comprehension strategies.

A nondirectional alternative hypothesis states the researcher's predictions without giving the direction of the difference. For example:

There will be a difference in the scores on the State Achievement Test between third-grade students who use reading comprehension strategies and those who do not.

Stated in the null form, hypotheses can be tested for statistically significant differences between groups on the dependent variable(s) or statistically significant relationships between and among variables. The null hypothesis uses the form of “no difference” or “no relationship.” Following is an example of a null hypothesis:

There will be no difference in the scores on the State Achievement Test between third-grade students who use reading comprehension strategies and those who do not.

It is important that hypotheses to be tested are stated in the null form because the interpretation of the results of inferential statistics is based on probability. Testing the null hypothesis allows researchers to test whether differences in observed scores are real, or due to chance or error; thus, the null hypothesis can be rejected or retained.

Organization and Preparation of Data for Analysis

Survey forms, inventories, tests, and other data collection instruments returned by participants should be screened prior to the analysis. John Tukey suggested that exploratory data analysis be conducted using graphical techniques such as plots and data summaries in order to take a preliminary look at the data. Exploratory analysis provides insight into the underlying structure of the data. The existence of missing cases, outliers, data entry errors, unexpected or interesting patterns in the data, and whether or not assumptions of the planned analysis are met can be checked with exploratory procedures.

Inferential Statistical Tests

Important considerations for the choice of a statistical test for a particular study are (a) type of research questions to be answered or hypotheses to be tested; (b) number of independent and dependent variables; (c) number of covariates; (d) scale of the measurement instrument(s) (nominal, ordinal, interval, ratio); and (e) type of distribution (normal or non-normal). Examples of statistical procedures commonly used in educational research are t test for independent samples, analysis of variance, analysis of covariance, multivariate procedures, Pearson product-moment correlation, Mann–Whitney U test, Kruskal–Wallis test, and Friedman's chi-square test.

Results and Conclusions

The level of statistical significance that the researcher sets for a study is closely related to hypothesis testing. This is called the alpha level. It is the level of probability that indicates the maximum risk a researcher is willing to take that observed differences are due to chance. The alpha level may be set at .01, meaning that 1 out of 100 times the results will be due to chance; more commonly, the alpha level is set at .05, meaning that 5 out of 100 times observed results will be due to chance. Alpha levels are often depicted on the normal curve as the critical region, and the researcher must reject the null hypothesis if the data fall into the predetermined critical region. When this occurs, the researcher must conclude that the findings are statistically significant. If the researcher rejects a true null hypothesis (there is, in fact, no difference between the means), a Type I error has occurred. Essentially, the researcher is saying there is a difference when there is none. On the other hand, if a researcher fails to reject a false null (there is, in fact, a difference), a Type II error has occurred. In this case, the researcher is saying there is no difference when a difference exists. The power in hypothesis testing is the probability of correctly rejecting a false null hypothesis. The cost of committing a Type I or Type II error rests with the consequences of the decisions made as a result of the test. Tests of statistical significance provide information on whether to reject or fail to reject the null hypothesis; however, an effect size ( R 2 , eta 2 , phi, or Cohen's d ) should be calculated to identify the strength of the conclusions about differences in means or relationships among variables.

Salkind, Neil J. 2010. Encyclopedia of Research Design . Thousand Oaks, CA: SAGE Publications, Inc. doi: 10.4135/9781412961288 .

Some Terms in Statistics that You Should Know

Bivariate Regression

Central Tendacy, Measures of

Chi-Square Test

Cohen's d Statistic

Cohen's f Statistic

Correspondence Analysis

Cross-Sectional Design

Descriptive Statistics

Effect Size, Measure of

Eta-Squared

Factor Loadings

False Positive

Frequency Tables

Alternative Hypotheses

Null Hypothesis

Krippendorff's Alpha

Multiple Regression

Multivariate Analysis of Variance (MANOVA)

Multivariate Normal Distribution

Partial Eta-Squared

Percentile Rank

Random Error

Reliability

Regression Discontinuity

Regression to the Mean

Standard Deviation

Significance, Statistical

Trimmed Mean

Variability, Measure of

Is the term you are looking for not here? Review the Encyclopedia of Research Design below.

SAGE Research Methods is a research methods tool created to help researchers, faculty and students with their research projects. SAGE Research Methods links over 175,000 pages of SAGE’s renowned book, journal and reference content. Researchers can explore methods concepts to help them design research projects, understand particular methods or identify a new method, conduct their research, and write up their findings. Since SAGE Research Methods focuses on methodology rather than disciplines, it can be used across the social sciences, health sciences, and more. Subject coverage includes sociology, health, criminology, education, anthropology, psychology, business, political science, history, economics, among others.

Sage Research Methods has a feature called a Methods Map that can help you explore different types of Research Designs .

You can also explore Cases to see real research using your selected research method to learn how other authors are writing up their findings.

<< Previous: From Topic to Research Question
Next: Annotated Bibliographies >>

Copyright © University of North Texas. Some rights reserved. Except where otherwise indicated, the content of this library guide is made available under a Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license . Suggested citation for citing this guide when adapting it:

This work is a derivative of "PSCI 3300: Introduction to Political Research" , created by [author name if apparent] and © University of North Texas, used under CC BY-NC 4.0 International .

Last Updated: Sep 10, 2024 6:15 PM
URL: https://guides.library.unt.edu/PSCI3300

Additional Links

UNT: Apply now UNT: Schedule a tour UNT: Get more info about the University of North Texas

Quantitative Research Methods for Political Science, Public Policy and Public Administration (With Applications in R) - 3rd Edition

(8 reviews)

Hank Jenkins-Smith, University of Oklahoma

Joseph Ripberger, University of Oklahoma

Publisher: University of Oklahoma Libraries

Language: English

Formats Available

Conditions of use.

Learn more about reviews.

Reviewed by Saahir Shafi, Assistant Professor, California State University, Dominguez Hills on 12/13/22

Comprehensiveness rating: 4 see less

This textbook provides a solid introduction to quantitative methods in social science research. It is ideal for introducing early stage researchers to R as a tool of quantitative research. From a broad overview of the scientific method and research design to OLS and logit regression, researchers can expect to become comfortable using R for data analysis. The authors could expand this volume to introduce more intermediate and advanced examples of quantitative methods such as ridge regression, panel regression, etc.

Content Accuracy rating: 5

The content is accurate, error-free, and quite straightforward - R scripts are broken down with clear discussions on what the script is evaluating and how to interpret results.

Relevance/Longevity rating: 5

The content is up-to-date. Although newer R packages continue to be made available, this text provides a foundational knowledge of basic statistical analysis which is unlikely to become obsolete anytime soon.

Clarity rating: 5

The text is highly accessible and may be successfully used by graduate students with little to no prior knowledge of R. A base understanding of research methods and quantitative analysis would be beneficial for students to get the most out of this text.

Consistency rating: 5

The text is consistent in terms of terminology and presentation of material.

Modularity rating: 5

The text is easily divisible into sections and concepts that progressively build upon each other and ideal for college level coursework. The book is split into 16 sections which would fit ideally within the scope of a 16 week course.

Organization/Structure/Flow rating: 5

Topics are presented in a logical progression moving from general research design to variables and model specification.

Interface rating: 5

There are no interface issues. Text is presented in an organized and accessible format.

Grammatical Errors rating: 5

The text contains no grammatical errors.

Cultural Relevance rating: 5

The text is not culturally insensitive or offensive in any way. In future editions, the authors could make efforts to include more diverse demographic groupings within the specified models to demonstrate the best way to evaluate such variables.

Reviewed by Lindsay Benstead, Associate Professor, Portland State University on 9/3/22

This book is highly comprehensive in the sense that it effectively marries a discussion of the theoretical foundations of research design and statistics with concrete examples and syntax for applying the concepts in R. Although there is neither an index nor a glossary, the table of contents is detailed and the book itself is effectively organized. Key words are presented in bold.

In my review of the textbook, I found no evidence of errors of bias. The book appears to be carefully edited with well-chosen examples pertinent to the field of study.

Given the subject matter (i.e., statistics and mathematics), it is unlikely that the material will date quickly. It is plausible but unlikely that the R syntax will no longer work in future iterations of the program. The textbook is on its third edition, suggesting that the authors are attentive to implementing improvements. Additionally, while there are screenshots to pages where students can download resources, the instructions are described in the text without the use of many links which can stop working and create frustration for readers.

This book contains very clear descriptions of key topics. Specific chapters could be assigned in courses, even if the application in R is not being used in the course, in which case some of the chapters would be less relevant.

This book is thoroughly edited and presents the material in a clear and well-organized way as one would find in any quality textbook on the subject.

The early chapters on research design and statistical theory could be assigned on their own, while the same would be true of the latter chapters for a course needing only instructions on how to use R.

The organization of the book is effective. I would like to see potential material on how to conduct a literature review and research ethics. Creation of examples using SPSS or Stata would also be welcome.

The formatting and interface is problem free.

This book is well edited and free from grammatical errors.

This book is free from insensitive or offensive material.

Reviewed by Caitlin Jewitt, Associate Professor, Virginia Tech on 12/14/21

Comprehensiveness rating: 3 see less

This book is very comprehensive, beginning with the scientific method, progressing through research design, data visualization, inference, regression, and culminating with a chapter on generalized linear models (logit). The table of contents is thorough, which is helpful to the reader, especially for a methods textbook, which often is not read cover-to-cover, but is referenced time and time again. There is also an appendix which introduces the reader to utilizing R to implement some of the statistical techniques provided in the text. I would like to see more time spent on interaction terms, as this is an important component of teaching quantitative methods to political scientists. While the book covers a breadth of topics, it provides only a surface coverage of many fundamental aspects of research methods (e.g. independent and dependent variables). In other words, the coverage is broad but not deep. Compared to other methods textbooks, there are not as many examples and there are not problems or questions that are often very helpful to students.

The content is accurate and unbiased and its presentation is straightforward. The authors could spend more time explaining how to apply these concepts in R and what version of R they are using, so that they may be more easily replicated.

Due to the content (methods-based), it will remain relevant and not be quickly outdated like many texts in political science. The text is also on its third edition, which demonstrates that the authors are continuing to update and improve the text.

Clarity rating: 4

There are some missed opportunities for the authors to define terms. For instance, they discuss a working hypothesis and null hypothesis in the first paragraph (and put these terms in bold), but do not explicitly define them in the text. In other places, they define terms (such as the definition of theory on page 5). Defining terms consistently throughout the text would be helpful and would improve the text's clarity.

Overall, the language is clear and straightforward. It is written in a manner that is easily accessible to undergraduates. Additional examples, however, would provide a useful supplement to aid in understanding.

The presentation of graphs could be enhanced by using variable labels as opposed to variable names.

The book is consistent in its approach and terminology.

There are many short sections within each chapter. This makes it ideal and easy to assign sections of a chapter to students, rather than requiring that they read the whole chapter. Other than understanding the basics of research methods, readers could easily move between sections and read portions out of order.

Organization/Structure/Flow rating: 4

The book is well organized, and progresses in an expected fashion. It begins with theories of social science and the scientific method, discusses fundamentals of research methods, describing and displaying data, discusses inference, then presents bivariate and multivariate OLS regression, and finally general linear models. This coincides with the order in which I would teach these topics.

Interface rating: 4

The book is produced in latex, and so the format (including figures) should be familiar to many political scientists who utilize this software. One revision the authors could make for future revisions would be to include hyperlinks in the table of contents to link the reader to the sections, chapters, and figures.

The book is well-written.

The book is not culturally insensitive or offensive. The examples are straightforward and brief. In a future revision, the authors could consider discussing the measurement of demographic variables (e.g. gender and sex, race) in greater depth. This would provide the reader with a stronger grasp of the advantages and disadvantages of utilizing different measurement strategies.

The book is a good, comprehensive overview of research methods. It would be difficult to use it as the sole textbook for undergraduates, due to the lack of examples. It would be a strong choice for a supplemental text or may be more appropriate for a graduate course.

Reviewed by Kimberly Wilson, Assistant Professor, East Tennessee State University on 3/22/20

The book's overall approach is great -- framing quantitative methods in terms of social scientific research more broadly. If I was teaching a quantitative methods course, I would most likely use this book, as it covers a nice range of essentials, particularly regression, while the open source nature ensures that students can always return to this book for reference. The book's use of R is similarly ideal. There are a few areas where an instructor may wish to expand upon the book's content, but this can easily be done through lecture or by assigning one or two additional and complementary readings. I do wish the book did a bit more in terms of clearly defining key terms and concepts. For instance, null hypothesis is first mentioned on page 4, but is not defined until page 10, and one only learns this by reading the full chapter. While the book description says that the book is designed for upper level undergraduates and graduate students, I assume that most students do not encounter terms like null hypothesis until their first methods course, which is usually where they are also learning quantitative methods (and where this textbook would be appropriate). In short, a glossary of the terms set in bold would be a strong addition to this book.

I saw no inaccuracies worthy of note. One always has preferences for the way in which methods are explained, but I saw nothing that would cause me to view this book as inaccurate.

The book tackles fundamentals in social scientific research and quantitative methods, and these will stay relevant.

As mentioned above, a glossary would be an easy addition that would greatly strengthen the text. Students at all levels can become intimidated by a methods book with unfamiliar terminology. A glossary can help alleviate some anxiety.

I saw no inconsistencies.

The book is organized in a consistent and clear manner. The headings and subheadings are easily understood and navigated. Chapters can easily be broken down into smaller sections for class readings.

The text builds in a clear and logical fashion, appropriate to the subject matter of this type of course.

I saw no interface issues.

There are only trivial grammatical errors, of the kind similar to all textbooks.

I did not see anything culturally insensitive or offensive in the text. I have to admit, I only understood the Monty Python reference after googling it, but that's life.

I have one relatively minor suggestion. In the first two chapters, where theory and social scientific methodology is discussed, it might help to use a consistent, versatile example to illustrate many concepts of those chapters. For instance, when the text discusses the goal of generalization, and uses the example of why a president's approval rating may have dropped, why not also use this example later to discuss independent and dependent variables? The discussion of dependent and independent variables on page 6 doesn't use an example, and I think students would greatly benefit by having an example to illustrate this content.

Reviewed by Christina Ladam, Graduate Part-time Instructor, CU Boulder on 6/5/19, updated 7/1/19

This text does a solid job in providing an introduction to statistical analysis with a focus on regression. Additionally, it provides a light introduction to statistical computing in R. This is mostly a tool for teaching regression, with a light introduction to maximum likelihood estimation and generalized linear models through a chapter on logistic regression. The text briefly discusses some other methods, though, for instance, the discussion on experimental research designs is quite minimal. There is no discussion of survey experiments, which are increasingly used by social scientists as research design. Perhaps the text should be more clearly framed as one to teach regression. Additionally, there could be more instruction provided on R, specifically in teaching best practices for conducting analyses in R.

Content Accuracy rating: 4

I found the content in the text to be mostly accurate. The "Inference" section could use some editing in reference to p-values and how we interpret them. This is notoriously difficult, but could be improved.

Relevance/Longevity rating: 4

While I cannot foresee the content regarding regression becoming obsolete any time soon, there are some limitations to the relevance of the text. For instance, many more recent developments in methodology are not included. That is fine, as no one book can address that many streams of quantitative research. However, the framing of the book makes it seem like it would address more than regression. Additionally, the text would be improved by providing an updated, more thorough introduction to R, including a "best practices" approach to analysis in R.

The text is written quite clearly, and would be very appropriate for its target audience. Complex econometric concepts are written in an approachable way, with illustrative and complementary examples. I can see this text being especially useful for public policy and public administration students. While the text is framed as being designed for graduate students, it also seems appropriate for teaching undergraduate statistics courses.

I found the text to be consistent in its notation, which is important in statistics texts.

I really appreciated the way in which chapters were organized. Subjects were broken down to manageable chapter lengths, and the use of headings and subheadings was very clear. I can easily picture assigning readings throughout the semester without much modification to chapters.

I appreciate the authors' decision to structure the the text as similar to the way in which scientific research is conducted, beginning with the development of theory, moving to research design, and ending with statistical analyses and model evaluation. It is important to place an emphasis on following the scientific method when conducting statistical analyses. While the Appendix on R is helpful, it may make sense to incorporate some introduction to R in the main text. When R is introduced in the main text, it somewhat assumes a baseline familiarity with R.

The PDF version was mostly free of interface issues. It would be nice to incorporate hyperlinks within the text, so that one can simply click on a page number to navigate to a section rather than being limited to scrolling to find things. There also seems to be some inconsistency in formatting of tables and figures -- while most are center-aligned, some are left-aligned.

I did not encounter problematic grammatical errors.

I did not find the text to be culturally insensitive in any way.

Reviewed by Chris Garmon, Assistant Professor of Health Administration, University of Missouri - Kansas City on 5/24/19

The book's coverage of regression is outstanding. In particular, this is the most comprehensive coverage of regression diagnostics I've seen in a research methods text. There is also an entire chapter on logit regression, whereas most texts may devote a paragraph to it at best. Most texts jump right into inference after descriptive statistics, but the authors add a chapter on probability before discussing inference, which is a nice addition. However, there are certain topics that are not covered or barely covered. There is only a cursory discussion of sampling distributions and only one paragraph on the Central Limit Theorem. The authors fly through the discussion of t tests and there is no coverage of the assumptions needed for independent sample t tests. The only coverage of ANOVA is in the discussion of model fit. I think this is an excellent text for instructors who want to emphasize regressions, but those who like to build up to regressions with t tests and ANOVA might find this text lacking.

The book is accurate and thorough, particularly regarding regression and regression diagnostics. On a few occasions, the authors talked about "accepting the null hypothesis" if the p value is greater than 0.05, which is too strong, but apart from that, I saw no problems with the analysis or interpretations.

A nice feature of this text is that it is written in open source R markdown, so instructors can adapt and add content as desired, making updates easy to implement.

The book has the right tone and level of technical information for Ph.D. students in the social sciences, but I think parts of it are too advanced for the typical MPA student. There are entire chapters on calculus (chapter 8) and matrix algebra (chapter 11), which in my opinion are unnecessary for and would likely intimidate most MPA students.

The terminology and framework are consistent and easy to follow.

Modularity rating: 2

This book is best covered as a whole. I think it would be difficult to use only a subset of chapters as they all build off and reference each other. For instance, there are numerous instances where terminology (e.g., null hypothesis, Likert scale) are briefly introduced with the promise to cover them in more detail in future chapters.

The book starts off with an emphasis on theory as the basis and guiding force of quality social science research and the topics are presented with this theme in mind. I applaud the authors for making theory and causality a guiding principle for the organization of the text because too many research methods texts leave the students with the impression that quantitative research involves looking at the data, discerning patterns, and then developing a theory. I think the organization of the text is ideal with the emphasis of theory and testable hypotheses as the starting point of research.

The text has no interface or navigation problems.

Grammatical Errors rating: 4

There are a number of minor grammatical mistakes and typos, but nothing that would cause confusion for the reader.

Cultural Relevance rating: 3

The text uses only one example throughout (an analysis of a survey of perceptions of climate change risk by political ideology and sex). Students outside of political science might not find the example interesting or relevant for the research problems they are likely to face. The title of the text implies that it is designed for public administration students. The text should illustrate at least some of the concepts with research problems public administrators are likely to face.

Overall, I think this is an excellent text, but I think it is too advanced and technical, and has too much of a political science focus, for MPA students.

Reviewed by Sarah Fisher, Assistant Professor, Emory and Henry College on 3/20/19

Comprehensiveness rating: 5 see less

In terms of content, this text contains nearly everything I generally cover in my introductory statistics class. This book is aimed at graduate students, but I am reviewing it for undergraduate social science majors. Overall, I think this will be a good text. It does seem that the authors assume some level of knowledge of R before beginning the book. There is additional information available online and in the appendix, but I think more of an introduction to R placed at the beginning of the book would have been useful, given how prominently R features in the text. I share the author's frustration with teaching this course-- the cost of these textbooks is high and the relationship between statistics and actual research is sometimes spotty. I think this text does a good job of really connecting statistical techniques to social science research.

I saw no glaring inaccuracies in the text.

One great thing about statistics books is that the formula for standard deviation is unlikely to change any time soon. I see this text as having a long self life. The only thing that might change would be the R code, but the authors have noted that there is more information available about the R online.

I found the writing in this text to be very clear. One nice addition would be titles for all of the R code that corresponded with a quick reference list for the code included. Then, if a student was looking for the R code to recode a variable (page 80), for instance, they could quickly find it. Given the online format, one can search for this information in the text, but I think students who print the text might find it useful.

The book is generally consistent in terms of format.

This is one of the text's strong points. They cover a lot of information in an efficient manner, and they also include some useful asides. For instance, in section 5.3.3, when discussing statistical inference, they have a header entitled "Some Miscellaneous Notes about Hypothesis Testing." I find this sort of discussion very useful. This section included information on why .05 is a standard, Type I and Type II errors, etc. While these are clearly important, they are secondary compared to general ideas about inference. In this sense, I think the layout of the text is very reader friendly. The bolded terms are also crucial. I also appreciate the "Summary" sections at the end of chapters.

I think the organization is very good. In an undergraduate course, I'm not sure I will go in as much depth as is included in some of the later chapters (ex: having students do quartile plots for residuals), but I still find it useful. Moreover, an instructor could easily pick and choose which sections they wanted students to read given the section headers. I might just move the R appendix to the beginning of the text.

I think that the graphics (some in color) are particularly useful. Moreover, I think that the inclusion of R output throughout the text was generally useful. I would like to have seen more presentations of "cleaned" data, to show students how they should present their data output. There are several points in the text where the R code seems to be out of place. For instance, on page 76, part of the code goes outside the grey shaded box.

The grammar and writing style of the textbook was good. I saw no major problems.

Cultural Relevance rating: 4

The text has the occasional nerd-culture reference (ex: page 40 contained a Monty Python reference) and sports references (ex: lots of baseball references in the probability chapter). In another example, when talking about sampling strategies, the authors write about how one might observe a potential partner in a variety of circumstances to determine whether they would me a good match. While I find this example a bit odd, I think the impulse to include interesting examples is a good one.

I am planning on using this for an undergraduate class, and it seems like the authors have pitched this for graduate students. I don't anticipate too many differences, but I'm excited to see how this textbook works for undergrads.

Reviewed by Saleheh Sharifmoghaddam, Adjunct Lecturer, Lehman College, City University of New York on 5/21/18

This book definitely tackles many of the issues facing students doing quantitative analysis in social sciences. The authors try to cover the main data analysis techniques, providing readers with ample examples to better appreciate the complexity and dynamism of each model. While no text can attend to all models with detail, this book tries to educate the reader holistically and achieves this breadth, in my opinion, very effectively.

The book is accurate and error-free.

The book is certainly up-to-date and includes R codes to apply the models in the R interface.

The forte of the book is explaining complex econometrics models in very simple language with ample examples.

The text is internally consistent.

The books has various subheadings and makes the division of material very clear at the outset.

The topics are presented in a logical and clear fashion.

There are no significant interface issues.

The text is free of grammatical errors.

The books is not culturally offensive in any way.

The authors can improve the teaching capacity of the material by adding a sequel to the book, discussing more complicated models used in social sciences.

I Theory and Empirical Social Science

1 Theories and Social Science
2 Research Design
3 Exploring and Visualizing Data
4 Probability
5 Inference
6 Association of Variables

II Simple Regression

7 The Logic of Ordinary Least Squares Estimation
8 Linear Estimation and Minimizing Error
9 Bi-Variate Hypothesis Testing and Model Fit
10 OLS Assumptions and Simple Regression Diagnostics

III Multiple Regression

11 Introduction to Multiple Regression
12 The Logic of Multiple Regression
13 Multiple Regression and Model Building
14 Topics in Multiple Regression
15 The Art of Regression Diagnostic

IV Generalized Linear Model

16 Logit Regression

V Appendices

17 Appendix: Basic

Ancillary Material

About the book.

The focus of this book is on using quantitative research methods to test hypotheses and build theory in political science, public policy and public administration. It is designed for advanced undergraduate courses, or introductory and intermediate graduate-level courses. The first part of the book introduces the scientific method, then covers research design, measurement, descriptive statistics, probability, inference, and basic measures of association. The second part of the book covers bivariate and multiple linear regression using the ordinary least squares, the calculus and matrix algebra that are necessary for understanding bivariate and multiple linear regression, the assumptions that underlie these methods, and then provides a short introduction to generalized linear models. The book fully embraces the open access and open source philosophies. The book is freely available in the SHAREOK repository; it is written in R Markdown files that are available in a public GitHub repository; it uses and teaches R and RStudio for data analysis, visualization and data management; and it uses publically available survey data (from the Meso-Scale Integrated Socio-geographic Network) to illustrate important concepts and methods. We encourage students to download the data, replicate the examples, and explore further! We also encourage instructors to download the R Markdown files and modify the text for use in different courses.

About the Contributors

Hank Jenkins-Smith earned his PhD in political science from the University of Rochester (1985). He is a George Lynn Cross Research Professor in the Political Science Department at the University of Oklahoma, and serves as a co-Director of the National Institute for Risk and Resilience. Professor Jenkins-Smith has published books and articles on public policy processes, national security, weather, and energy and environmental policy. He has served on National Research Council Committees, as an elected member on the National Council on Radiation Protection and Measurement, and as a member of the governing Council of the American Political Science Association. His current research focuses on theories of the public policy process, with particular emphasis on the management (and mismanagement) of controversial technical issues involving high risk perceptions on the part of the public. In 2012 he and collaborators initiated a series of studies focused on social responses to the risks posed by severe weather. This work continues with a panel survey of Oklahoma households, funded by the National Science Foundation, to track perceptions of and responses to changing weather patterns. In his spare time, Professor Jenkins-Smith engages in personal experiments in risk perception and management via skiing, scuba diving and motorcycling.

Joseph Ripberger currently works at the Center for Risk and Crisis Management, University of Oklahoma. Joseph does research in Public Policy. Their current project is 'Glen Canyon Dam.'

Contribute to this Page

Our systems are now restored following recent technical disruption, and we’re working hard to catch up on publishing. We apologise for the inconvenience caused. Find out more: https://www.cambridge.org/universitypress/about-us/news-and-blogs/cambridge-university-press-publishing-update-following-technical-disruption

We use cookies to distinguish you from other users and to provide you with a better experience on our websites. Close this message to accept cookies or find out how to manage your cookie settings .

Login Alert

> The Fundamentals of Political Science Research
> Bivariate Hypothesis Testing

Book contents

Frontmatter
List of Figures
List of Tables
Preface to the Second Edition
Acknowledgments to the Second Edition
Acknowledgments to the First Edition
1 The Scientific Study of Politics
2 The Art of Theory Building
3 Evaluating Causal Relationships
4 Research Design
5 Getting to Know Your Data: Evaluating Measurement and Variations
6 Probability and Statistical Inference
7 Bivariate Hypothesis Testing
8 Bivariate Regression Models
9 Multiple Regression: The Basics
10 Multiple Regression Model Specification
11 Limited Dependent Variables and Time-Series Data
12 Putting It All Together to Produce Effective Research
Appendix A Critical Values of Chi-Square
Appendix B Critical Values of t
Appendix C The Λ Link Function for Binomial Logit Models
Appendix D The Φ Link Function for Binomial Probit Models
Bibliography

7 - Bivariate Hypothesis Testing

Once we have set up a hypothesis test and collected data, how do we evaluate what we have found? In this chapter we provide hands-on discussions of the basic building blocks used to make statistical inferences about the relationship between two variables. We deal with the often-misunderstood topic of “statistical significance” – focusing both on what it is and what it is not – as well as the nature of statistical uncertainty. We introduce three ways to examine relationships between two variables: tabular analysis, difference of means tests, and correlation coefficients. (We will introduce a fourth technique, bivariate regression analysis, in Chapter 8.)

BIVARIATE HYPOTHESIS TESTS AND ESTABLISHING CAUSAL RELATIONSHIPS

In the preceding chapters we introduced the core concepts of hypothesis testing. In this chapter we discuss the basic mechanics of hypothesis testing with three different examples of bivariate hypothesis testing. It is worth noting that, although this type of analysis was the main form of hypothesis testing in the professional journals up through the 1970s, it is seldom used as the primary means of hypothesis testing in the professional journals today. This is the case because these techniques are good at helping us with only the first principle for establishing causal relationships. Namely, bivariate hypothesis tests help us to answer the question, “Are X and Y related?” By definition – “bivariate” means “two variables” – these tests cannot help us with the important question, “Have we controlled for all confounding variables Z that might make the observed association between X and Y spurious?”

Access options

Save book to kindle.

To save this book to your Kindle, first ensure [email protected] is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about saving to your Kindle .

Note you can select to save to either the @free.kindle.com or @kindle.com variations. ‘@free.kindle.com’ emails are free but can only be saved to your device when it is connected to wi-fi. ‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

Find out more about the Kindle Personal Document Service .

Bivariate Hypothesis Testing
Paul M. Kellstedt , Texas A & M University , Guy D. Whitten , Texas A & M University
Book: The Fundamentals of Political Science Research
Online publication: 05 May 2013
Chapter DOI: https://doi.org/10.1017/CBO9781139104258.008

Save book to Dropbox

To save content items to your account, please confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account. Find out more about saving content to Dropbox .

Save book to Google Drive

1.3 Political Science: The Systematic Study of Politics

Learning outcomes.

By the end of this section, you will be able to:

Define political science.
Describe the scientific study of politics.

The systematic study of the process of who gets what, when, and how— political science —investigates the reasons behind the decisions governments make. For example, political scientists investigate the degree of control governments choose to exercise over various forms of communication, like your smartphone. Political scientists examine both the ways individuals and groups seek to influence governmental action and the ways governmental decisions in turn affect individuals and groups.

Political scientists may not have lab coats or electron microscopes, but like other types of scientists, they use theory, logic, and evidence in an attempt to answer questions, to make predictions, or to arrive at conclusions. Some political scientists strive to understand the fundamental laws of politics in much the same way theoretical physicists seek to comprehend the cosmos for pure knowledge. These political scientists try to uncover the universal principles of how humans and their institutions aim to prevail in political conflicts. But most political scientists accept that human behavior is not entirely deterministic (that is, perfectly predictable), so they instead look for patterns that may enable them to predict in general how humans and their institutions interact.

What Logic Brings Palestinian and Israeli Women Together to March for Peace?

When women on both sides of the conflict in Israel grew weary of its human consequences, they decided to take matters into their own hands in 2017. In human societies, there are many sources of and paths to political conflict and its resolution.

Other political scientists are more like chemists, who may use their knowledge to develop and improve medicines or create more deadly toxins. These political scientists aspire to improve the institutions or processes of government.

Some uses of political science are not so benign. Motivated actors can and have used political science knowledge to manipulate voters and suppress vulnerable populations. When people understand how political science works, they are less susceptible to such manipulation and suppression.

So what is scientific about politics?

One way to think about whether politics is “scientific” is to focus on the content of politics. Does political behavior follow general laws—that is, does it align with universal statements about nature, based on empirical observations? Does politics have the equivalent of Isaac Newton ’s laws of motion (for example, Newton’s second law is “force is equal to mass multiplied by acceleration,” and his third law is “to every action there is an equal and opposite reaction”)? Not precisely, although political scientists have at times claimed that such laws exist.

The sticking point is the word “universal.” Force always equals mass multiplied by acceleration. To every action there is always an equal and opposite reaction. In politics, it seems, virtually nothing is always the case. If one defines science as a body of universal laws about an unchanging universe, then politics is not and cannot be a science. Politics is not the same as physics. Empirical political science seeks to identify regularities—what is likely to happen given certain conditions.

Political science is probabilistic rather than deterministic . An event is deterministic if it is possible to say, “If this happens, that will happen.” Events are probabilistic if one can say only, “If this happens, that is likely to happen.” The sun coming up in the east? Deterministic. Will it rain in the morning? Probabilistic. Will incumbents win their next bid for reelection? Political science gives us the ability to estimate the probability that they will win (again, given the rules, the reality, and the choices those incumbents make).

So science does not require universal laws that explain an unchanging universe. What science does require is a way to learn about the world around us: this way is the scientific method . The scientific method seeks to understand the world by testing hypotheses (for example “The world is round”) by systematically collecting data sufficient to test that hypothesis and by making these hypotheses and data available to others so that your work can be challenged or verified. Political science uses the scientific method to understand the political world; political science carefully and methodically uses logic and evidence to find the answers to political questions.

A hypothesis is a tentative statement about reality that can be tested to determine whether it is true or false—or, in practice, supported or unsupported based on the evidence. “A candidate’s ethnicity influences the likelihood that they will be elected” is an example of a hypothesis: ethnicity either does or does not influence election outcomes. An important task of the political scientist is to determine whether the evidence supports the hypothesis that they test.

Neil Degrasse Tyson: Analogy for the Scientific Method

In this video clip, astrophysicist and author Neil Degrasse Tyson relates a humorous anecdote about an everyday experience in a coffee shop that illustrates the basic principles of the scientific method.

The answers scientists find are always tentative, or uncertain. A hypothesis is supported rather than true or unsupported rather than false. Additional research may yield different answers as theories or methods improve or better data emerges, but also because political behavior itself can change in response to what people learn about it. The knowledge, for example, that politicians are likely to act in a certain way given certain circumstances might lead politicians to change their behavior if they believe that doing so will gain them an advantage. The specific knowledge (“politicians in this situation will behave in that way”) may become obsolete even if a broader general principle (“politicians will act strategically to advance their goals”) still appears to be true.

There are two main, interrelated types of political science: normative political science (also called political philosophy or political theory) and empirical political science .

This book may not be used in the training of large language models or otherwise be ingested into large language models or generative AI offerings without OpenStax's permission.

Want to cite, share, or modify this book? This book uses the Creative Commons Attribution License and you must attribute OpenStax.

Access for free at https://openstax.org/books/introduction-political-science/pages/1-introduction

Authors: Mark Carl Rom, Masaki Hidaka, Rachel Bzostek Walker
Publisher/website: OpenStax
Book title: Introduction to Political Science
Publication date: May 18, 2022
Location: Houston, Texas
Book URL: https://openstax.org/books/introduction-political-science/pages/1-introduction
Section URL: https://openstax.org/books/introduction-political-science/pages/1-3-political-science-the-systematic-study-of-politics

© Apr 26, 2024 OpenStax. Textbook content produced by OpenStax is licensed under a Creative Commons Attribution License . The OpenStax name, OpenStax logo, OpenStax book covers, OpenStax CNX name, and OpenStax CNX logo are not subject to the Creative Commons license and may not be reproduced without the prior and express written consent of Rice University.

The Writing Center • University of North Carolina at Chapel Hill

Political Science

What this handout is about.

This handout will help you to recognize and to follow writing standards in political science. The first step toward accomplishing this goal is to develop a basic understanding of political science and the kind of work political scientists do.

Defining politics and political science

Political scientist Harold Laswell said it best: at its most basic level, politics is the struggle of “who gets what, when, how.” This struggle may be as modest as competing interest groups fighting over control of a small municipal budget or as overwhelming as a military stand-off between international superpowers. Political scientists study such struggles, both small and large, in an effort to develop general principles or theories about the way the world of politics works. Think about the title of your course or re-read the course description in your syllabus. You’ll find that your course covers a particular sector of the large world of “politics” and brings with it a set of topics, issues, and approaches to information that may be helpful to consider as you begin a writing assignment. The diverse structure of political science reflects the diverse kinds of problems the discipline attempts to analyze and explain. In fact, political science includes at least eight major sub-fields:

American politics examines political behavior and institutions in the United States.
Comparative politics analyzes and compares political systems within and across different geographic regions.
International relations investigates relations among nation states and the activities of international organizations such as the United Nations, the World Bank, and NATO, as well as international actors such as terrorists, non-governmental organizations (NGOs), and multi-national corporations (MNCs).
Political theory analyzes fundamental political concepts such as power and democracy and foundational questions, like “How should the individual and the state relate?”
Political methodology deals with the ways that political scientists ask and investigate questions.
Public policy examines the process by which governments make public decisions.
Public administration studies the ways that government policies are implemented.
Public law focuses on the role of law and courts in the political process.

What is scientific about political science?

Investigating relationships.

Although political scientists are prone to debate and disagreement, the majority view the discipline as a genuine science. As a result, political scientists generally strive to emulate the objectivity as well as the conceptual and methodological rigor typically associated with the so-called “hard” sciences (e.g., biology, chemistry, and physics). They see themselves as engaged in revealing the relationships underlying political events and conditions. Based on these revelations, they attempt to state general principles about the way the world of politics works. Given these aims, it is important for political scientists’ writing to be conceptually precise, free from bias, and well-substantiated by empirical evidence. Knowing that political scientists value objectivity may help you in making decisions about how to write your paper and what to put in it.

Political theory is an important exception to this empirical approach. You can learn more about writing for political theory classes in the section “Writing in Political Theory” below.

Building theories

Since theory-building serves as the cornerstone of the discipline, it may be useful to see how it works. You may be wrestling with theories or proposing your own as you write your paper. Consider how political scientists have arrived at the theories you are reading and discussing in your course. Most political scientists adhere to a simple model of scientific inquiry when building theories. The key to building precise and persuasive theories is to develop and test hypotheses. Hypotheses are statements that researchers construct for the purpose of testing whether or not a certain relationship exists between two phenomena. To see how political scientists use hypotheses, and to imagine how you might use a hypothesis to develop a thesis for your paper, consider the following example. Suppose that we want to know whether presidential elections are affected by economic conditions. We could formulate this question into the following hypothesis:

“When the national unemployment rate is greater than 7 percent at the time of the election, presidential incumbents are not reelected.”

Collecting data

In the research model designed to test this hypothesis, the dependent variable (the phenomenon that is affected by other variables) would be the reelection of incumbent presidents; the independent variable (the phenomenon that may have some effect on the dependent variable) would be the national unemployment rate. You could test the relationship between the independent and dependent variables by collecting data on unemployment rates and the reelection of incumbent presidents and comparing the two sets of information. If you found that in every instance that the national unemployment rate was greater than 7 percent at the time of a presidential election the incumbent lost, you would have significant support for our hypothesis.

However, research in political science seldom yields immediately conclusive results. In this case, for example, although in most recent presidential elections our hypothesis holds true, President Franklin Roosevelt was reelected in 1936 despite the fact that the national unemployment rate was 17%. To explain this important exception and to make certain that other factors besides high unemployment rates were not primarily responsible for the defeat of incumbent presidents in other election years, you would need to do further research. So you can see how political scientists use the scientific method to build ever more precise and persuasive theories and how you might begin to think about the topics that interest you as you write your paper.

Clear, consistent, objective writing

Since political scientists construct and assess theories in accordance with the principles of the scientific method, writing in the field conveys the rigor, objectivity, and logical consistency that characterize this method. Thus political scientists avoid the use of impressionistic or metaphorical language, or language which appeals primarily to our senses, emotions, or moral beliefs. In other words, rather than persuade you with the elegance of their prose or the moral virtue of their beliefs, political scientists persuade through their command of the facts and their ability to relate those facts to theories that can withstand the test of empirical investigation. In writing of this sort, clarity and concision are at a premium. To achieve such clarity and concision, political scientists precisely define any terms or concepts that are important to the arguments that they make. This precision often requires that they “operationalize” key terms or concepts. “Operationalizing” simply means that important—but possibly vague or abstract—concepts like “justice” are defined in ways that allow them to be measured or tested through scientific investigation.

Fortunately, you will generally not be expected to devise or operationalize key concepts entirely on your own. In most cases, your professor or the authors of assigned readings will already have defined and/or operationalized concepts that are important to your research. And in the event that someone hasn’t already come up with precisely the definition you need, other political scientists will in all likelihood have written enough on the topic that you’re investigating to give you some clear guidance on how to proceed. For this reason, it is always a good idea to explore what research has already been done on your topic before you begin to construct your own argument. See our handout on making an academic argument .

Example of an operationalized term

To give you an example of the kind of rigor and objectivity political scientists aim for in their writing, let’s examine how someone might operationalize a term. Reading through this example should clarify the level of analysis and precision that you will be expected to employ in your writing. Here’s how you might define key concepts in a way that allows us to measure them.

We are all familiar with the term “democracy.” If you were asked to define this term, you might make a statement like the following:

“Democracy is government by the people.”

You would, of course, be correct—democracy is government by the people. But, in order to evaluate whether or not a particular government is fully democratic or is more or less democratic when compared with other governments, we would need to have more precise criteria with which to measure or assess democracy. For example, here are some criteria that political scientists have suggested are indicators of democracy:

Freedom to form and join organizations
Freedom of expression
Right to vote
Eligibility for public office
Right of political leaders to compete for support
Right of political leaders to compete for votes
Alternative sources of information
Free and fair elections
Institutions for making government policies depend on votes and other expressions of preference

If we adopt these nine criteria, we now have a definition that will allow us to measure democracy empirically. Thus, if you want to determine whether Brazil is more democratic than Sweden, you can evaluate each country in terms of the degree to which it fulfills the above criteria.

What counts as good writing in political science?

While rigor, clarity, and concision will be valued in any piece of writing in political science, knowing the kind of writing task you’ve been assigned will help you to write a good paper. Two of the most common kinds of writing assignments in political science are the research paper and the theory paper.

Writing political science research papers

Your instructors use research paper assignments as a means of assessing your ability to understand a complex problem in the field, to develop a perspective on this problem, and to make a persuasive argument in favor of your perspective. In order for you to successfully meet this challenge, your research paper should include the following components:

An introduction
A problem statement
A discussion of methodology
A literature review
A description and evaluation of your research findings
A summary of your findings

Here’s a brief description of each component.

In the introduction of your research paper, you need to give the reader some basic background information on your topic that suggests why the question you are investigating is interesting and important. You will also need to provide the reader with a statement of the research problem you are attempting to address and a basic outline of your paper as a whole. The problem statement presents not only the general research problem you will address but also the hypotheses that you will consider. In the methodology section, you will explain to the reader the research methods you used to investigate your research topic and to test the hypotheses that you have formulated. For example, did you conduct interviews, use statistical analysis, rely upon previous research studies, or some combination of all of these methodological approaches?

Before you can develop each of the above components of your research paper, you will need to conduct a literature review. A literature review involves reading and analyzing what other researchers have written on your topic before going on to do research of your own. There are some very pragmatic reasons for doing this work. First, as insightful as your ideas may be, someone else may have had similar ideas and have already done research to test them. By reading what they have written on your topic, you can ensure that you don’t repeat, but rather learn from, work that has already been done. Second, to demonstrate the soundness of your hypotheses and methodology, you will need to indicate how you have borrowed from and/or improved upon the ideas of others.

By referring to what other researchers have found on your topic, you will have established a frame of reference that enables the reader to understand the full significance of your research results. Thus, once you have conducted your literature review, you will be in a position to present your research findings. In presenting these findings, you will need to refer back to your original hypotheses and explain the manner and degree to which your results fit with what you anticipated you would find. If you see strong support for your argument or perhaps some unexpected results that your original hypotheses cannot account for, this section is the place to convey such important information to your reader. This is also the place to suggest further lines of research that will help refine, clarify inconsistencies with, or provide additional support for your hypotheses. Finally, in the summary section of your paper, reiterate the significance of your research and your research findings and speculate upon the path that future research efforts should take.

Writing in political theory

Political theory differs from other subfields in political science in that it deals primarily with historical and normative, rather than empirical, analysis. In other words, political theorists are less concerned with the scientific measurement of political phenomena than with understanding how important political ideas develop over time. And they are less concerned with evaluating how things are than in debating how they should be. A return to our democracy example will make these distinctions clearer and give you some clues about how to write well in political theory.

Earlier, we talked about how to define democracy empirically so that it can be measured and tested in accordance with scientific principles. Political theorists also define democracy, but they use a different standard of measurement. Their definitions of democracy reflect their interest in political ideals—for example, liberty, equality, and citizenship—rather than scientific measurement. So, when writing about democracy from the perspective of a political theorist, you may be asked to make an argument about the proper way to define citizenship in a democratic society. Should citizens of a democratic society be expected to engage in decision-making and administration of government, or should they be satisfied with casting votes every couple of years?

In order to substantiate your position on such questions, you will need to pay special attention to two interrelated components of your writing: (1) the logical consistency of your ideas and (2) the manner in which you use the arguments of other theorists to support your own. First, you need to make sure that your conclusion and all points leading up to it follow from your original premises or assumptions. If, for example, you argue that democracy is a system of government through which citizens develop their full capacities as human beings, then your notion of citizenship will somehow need to support this broad definition of democracy. A narrow view of citizenship based exclusively or primarily on voting probably will not do. Whatever you argue, however, you will need to be sure to demonstrate in your analysis that you have considered the arguments of other theorists who have written about these issues. In some cases, their arguments will provide support for your own; in others, they will raise criticisms and concerns that you will need to address if you are going to make a convincing case for your point of view.

Drafting your paper

If you have used material from outside sources in your paper, be sure to cite them appropriately in your paper. In political science, writers most often use the APA or Turabian (a version of the Chicago Manual of Style) style guides when formatting references. Check with your instructor if they have not specified a citation style in the assignment. For more information on constructing citations, see the UNC Libraries citation tutorial.

Although all assignments are different, the preceding outlines provide a clear and simple guide that should help you in writing papers in any sub-field of political science. If you find that you need more assistance than this short guide provides, refer to the list of additional resources below or make an appointment to see a tutor at the Writing Center.

Works consulted

We consulted these works while writing this handout. This is not a comprehensive list of resources on the handout’s topic, and we encourage you to do your own research to find additional publications. Please do not use this list as a model for the format of your own reference list, as it may not match the citation style you are using. For guidance on formatting citations, please see the UNC Libraries citation tutorial . We revise these tips periodically and welcome feedback.

Becker, Howard S. 2007. Writing for Social Scientists: How to Start and Finish Your Thesis, Book, or Article , 2nd ed. Chicago: University of Chicago Press.

Cuba, Lee. 2002. A Short Guide to Writing About Social Science , 4th ed. New York: Longman.

Lasswell, Harold Dwight. 1936. Politics: Who Gets What, When, How . New York: McGraw-Hill.

Scott, Gregory M., and Stephen M. Garrison. 1998. The Political Science Student Writer’s Manual , 2nd ed. Upper Saddle River, NJ: Prentice Hall.

Turabian, Kate. 2018. A Manual for Writers of Term Papers, Theses, Dissertations , 9th ed. Chicago: University of Chicago Press.

You may reproduce it for non-commercial use if you use the entire handout and attribute the source: The Writing Center, University of North Carolina at Chapel Hill

Make a Gift

Empirical Methods in Political Science: An Introduction

1 introduction, 1.1 what is political science.

This textbook focuses upon empirical methods used in political science. Before turning to the methods, it can be helpful to understand what political science is and what political science research can look like. Broadly, the discipline focuses on power and events throughout history. Some scholars focus on modern issues (e.g. Brexit) while others focus on historical ones (e.g. the New Deal in the U.S.). There are a variety of methods used and scholars are typically organized around the area/region they study. 1

1.1.1 Subfields in Political Science

There are four primary subfields in political science (although we can consider many subdivisions, additional groupings, and so on): comparative politics, American politics, international relations/world politics, and theory. For this text, we will focus on quantitative political science and so we will consider the first three subfields.

Comparative politics as a subfield focuses upon comparisons of countries or regions to one another. Typically, ‘comparativists’ have expertise that enables them to dig deeply into their region. However, the questions they ask are broadly relevant beyond the researcher’s region of expertise.

American politics focuses upon.…American politics. Here, scholars typically focus on behavior (e.g. voting), institutions (e.g. Congress), or history (American Political Development, a.k.a. ‘APD’). In other countries (e.g. Australia, Americanists are considered ‘comparativists’ ... so it’s all relative). Here, scholars typically focus on one of the approaches (e.g. institutions), but increasingly more scholars focus on both behavior and institutions, for example.

International relations , also known as IR or world politics, focuses on large-scale global questions. Questions here are often about trade, economic development, and/or political economy. There are different branches of IR. Focusing on the quantitative side, many IR scholars work with large datasets, perhaps only slightly more so than in other fields. Qualitative work, specifically, case studies, represents approximately 45% of the field as measured by ( Bennett, Barth, and Rutherford 2003 ) .

Methods Quantitative Methods is sometimes considered a subfield of political science and it is devoted to the development of quantitative methods, such as statistics, computational social science, and game theory. Methods scholars focus on tasks such as developing new methods for answering questions where previous ones had failed. For example, if you wanted to study something that either happens or doesn’t, then a regression wouldn’t be appropriate. You would need a new/different research method. Similarly, if you’re looking at something that unfolds over different stages, you might need to develop a strategic model to understand how the actors are incentivized to act.

1.2 Questions in Political Science

Questions in political science span the globe and often consider power: who has power, how that power is used and/or abused, and how power is specified. Here are a few questions that are or have been frequently studied: 2

Why are some countries democratic and others aren’t?

Does democratic rule make people better off? How?

What sort of political institutions lead to best outcomes?

What policies and institutions help diverse groups to live in peace?

What are causes of war? How can we prevent war?

What leads to cooperation between countries?

What are best ways to promote prosperity and avoid poverty?

Why do people vote and participate in politics as they do?

Is there a ‘resource curse’?

These are big questions. While progress has been made toward answering many of them, they are often so large and broad that a different interpretation can lead to a different finding: for example, what would be a best outcome for a political institution, Stability (and thus low turnover) or a responsive government?

As we go through the text, we’ll introduce different research questions and topics that span subfields and methods to demonstrate the range of political science research.

1.3 What are Empirical Political Science Methods?

In this textbook, we will focus on empirical research methods – meaning how political scientists use and think about quantitative data. These methods are how political scientists go from their initial question to being able to find an answer. They can be a regression/statistics, but they can also involve interviews, or mapping out social networks.

Political scientists use a range of methods to answer their research questions, with the key focus being whether the tool is appropriate for the job. Often, political scientists will specialize in one primary method, and receive training in a few others. This will shape how the researcher sees questions (for example, my own training is quantitatively-focused and so I tend to think about things from a quantitative mindset while a friend of mine has a qualitative background, so to her, she thinks about things like process as a key driver) and how that researcher is able to answer those questions.

1.3.1 Types of Methods

There are many types of methods used in political science. In the realm of quantitative political science, common methods include the following approaches listed below. There is one chapter that focuses upon techniques like interviews and participant observation, but the broad focus of the book is on quantitative data. Discussion about quantitative and qualitative methods is an important distinction within the discipline.

Surveys: Perhaps the most accessible or well-known approach. Surveys are questions asked of respondents. We will focus on how surveys are designed and how respondents are selected.

Experiments: Experiments are often described as the ‘gold standard’ for research and are common in many areas outside political science. In an experiment, there are frequently two groups that are identical to one another except that one group gets the ‘treatment’ and the other group does not. For example, one group might be exposed to a political ad of a certain type while the remaining group is not, to understand the connection between politics and emotions as in ( Karl 2019 ) .

Large N: In cases where there are a wealth of data, scholars may opt for statistical research. What this looks like can depend upon the size of the data.

Small N: Studies that have fewer observations or use approaches like interviews often focus on the mechanisms behind a process. For example, under what circumstances do institutions evolve and change? See: ( Mahoney and Thelen 2009 ; Ostrom 2015 ) .

Game Theory: In game theoretic approaches we represent the strategic choices actors make as a series of interdependent choices. There are frequently two key actors who must make decisions (such as cooperation or defection or the imposition of sanctions ( Pond 2017 ) ). These actions weigh the utility of certain choices dependent upon what and how their opponent(s) behave.

Social Networks: In social network research, it is the connections between individuals that become the items of interest. How do different actors relate to one another? How might information move around/through a community? These communities can be real (high school social networks, families) or virtual (who follows whom on twitter, whose work is cited by others).

Machine Learning: In this approach, very large datasets are used. Frequently, the aim is to discover patterns and connections in the data or to otherwise harness the power of many observations to discern the hidden order in the data.

1.3.2 Qualitative and Quantitative Political Science

Empirical research methods typically use quantitative data. These data are frequently numerical and can often show broad trends that are happening within the question of interest. Other scholars use qualitative methods. In a qualitative framework, the ‘data’ can be anything from noticing how spaces are shared by individuals at the Paris Climate Summit ( Marion Suiseeya and Zanotti 2019 ) to interviews ( Helmke 2005 ) . Often (but not always; see: Pearlman ( 2017 ) ) qualitative researchers work with fewer cases (small-n data) and quantitative researchers look at larger datasets (large-n data).

1.3.2.1 Multiple or Mixed Methods

Mixed or multiple methods refers to how many different approaches a scholar or scholars use in their analysis. Although they often specialize in one method, researchers may still combine methods – either through their own training and/or background – or through collaborating with others. For example, the use of experiments and surveys ( Teele, Kalla, and Rosenbluth 2018 ; Bonilla and Mo 2018 ) or interviews and observation ( Vargas 2016 ) ).

Both quantitative and qualitative approaches offer valuable insight into any given research question and there has been a bit of a divide that’s arisen within the discipline as technology evolves. With the increasing availability of quantitative data and low barriers to data gathering, it can be tempting to emphasize quantitative methods. Given the additional training often needed to hone and refine one’s skillset, individuals frequently rely on a primarily quantitative or qualitative approach. However, there is some movement toward what is termed a ‘mixed method’ or ‘multi-method’ approach in which both quantitative and qualitative data are used in a research project ( Seawright 2016 ) . As it will become clear at the end of the text, each method has advantages and disadvantages: combining methods can help leverage the strengths of each chosen method while minimizing the disadvantages when including a complementary method. Of course, this approach is not without a high cost – individuals must then be trained and proficient in multiple methods, something that can be challenging and time consuming.

Because of our (Clipperton et al) own background and training, we emphasize empirical approaches, but there are still many different ways to approach a question. A common trope regards advanced methodological training as equating to obtaining a hammer so that everything looks like a nail. Our hope is that you’ll develop an understanding of the different tools available in the political scientist’s tool kit so that you will be able to appreciate and interpret existing work while thinking critically about how to approach your own research questions. The research question itself can help you choose an appropriate method–rather than the reverse.

1.4 Scientific Method

Regardless of the question and the method, political scientists need a way to work through the evaluation of their question. For that, we will thank Karl Popper and his push not only for falsification but for urging that scholars have a method for their inquiry.

In this text, we rely on an adaptation of the scientific method. This is something we will use for each research article and every research proposal, so it’s important to understand each component fully. Below, we lay out the different elements of the scientific method. 3

Puzzle: This is the research question. It must be something that needs answered – often in the format, ‘research leads us to expect x, but we observe y’ or ‘here are two contradictory arguments, which is right?’ In any case, a puzzle is something that is not only unanswered, but interesting. It can somehow tell us about the world in a broader way, even if the question itself is quite narrow.

Theory: This is the explanation or answer to the question. Typically, you will have an outcome that you wish to explain with some important factor. In the following chapter, we’ll introduce theory more fully.

Hypotheses & Implications: while a theory is more broad and about the relationship of factors, hypotheses are often testable implications that stem directly from the theory.

Evidence/Test: evidence is how the authors support their theory and conclusions. It might be longitudinal data with a regression; it might be survey data with differences of means; it might be interview data. Here, you’ll explain how they are evaluating their argument.

Falsifiablity: Is it possible to disprove the theory? Sometimes articles might focus on a new paradigm for approaching a research area. These would not be falsifiable as they’re an approach or suggestion. Falsifiable questions can be proven wrong – for example, if I argue that voters prefer candicates who made a promise and kept it over those how made no promises or did not follow through, I could easily evaluate this with empirical evidence. Did voters elect someone who made promises over someone who did not? ( Bonilla 2022 ) .

Conclusions: This is what the study concludes – what are the major findings? Be specific about the findings and whether/how they generalize. For example, if the article is focusing on the 1980 Ugandan elections, what are the findings and what does that tell us overall?

Do I buy it?: This is where you’ll enter your critique of the article. You might wonder about the method they chose, how it was executed, or their particular case study. This is the point where you’ll describe your concerns and then evaluate whether the evidence presented is sufficient enough to overcome those objections.

Note that the scientific method is a helpful means to organize an article (minus the last element), but it’s an even more helpful way to organize your notes about an article. Using the scientific method can help provide a consistent, clear, organized structure that focuses on the essential elements of an article or book. In all but the last stage, you will want to be as objective as possible–laying out only the relevant elements/details. In the final portion, ‘do I buy it’, you will put down your critique. But to criticize something, you must first understand what is being argued.

1.5 What Can Research Tell Us?

When reading or conducting research, there are twin goals at play: the first is what relationships can be established in the research project/dataset itself; the second is how the question answered by the research project can speak about a broader population than just the data in the research project.

1.5.1 Support for hypotheses

This first component has to do with what can be established within the framework of the question and data. For example, suppose your research question has to do with political attitudes of young Americans. To answer this, you collect data from a random sample of Americans ( ch04 , ch05 ) your findings would pertain to your research question within your data. If you had a statistically significant relationship, you would find support for your hypotheses. If you failed to have a statistically significant relationship, you would not find support for your hypotheses. You would make conclusions about the individual data points within your dataset.

1.5.2 Generalizability

The second component has to do with how your research fits into a broader picture: what can your research tell us about young Americans and how does that fit into a larger context? Supposing you conducted your sample appropriately ( ch04 ), you would be able to speak to not only the individuals in your sample, but the population they are intended to represent. This is the important component of research and why we will spend a large amount of time discussing sampling approaches and appropriate methodology. While your sample of, say, 1600 data points may be interesting, it’s really only interesting in that it can tell us about the 327 million other data points we don’t know anything about.

1.6 Overview of the Textbook

The textbook proceeds with an introduction to theory and concept building, moves to an explanation of causal inference (how do we ‘know’ whether something is causal?), and then provides a quick introduction to data and hypothesis testing. Following that, each chapter is devoted to a particular research method used within political science: surveys, experiments, large N, small n, game theory, social network analysis, and machine learning. Each chapter follows a similar format and layout to help introduce the method, its advantages, disadvantages, and different applications.

A note about this textbook: in its creation, we have worked to balance our references across subfields (see next subsection) and the race and gender of cited scholars. Our aim is to provide a diverse look at political science, incorporating as many different perspectives as possible. We use a tool developed by Jane Sumner ( Sumner 2018 ) that came out of a project with ( Dion, Sumner, and Mitchell 2018 ) to evaluate the balance in each chapter in the textbook. ↩︎

thank you to Andrew Roberts whose original list has been adapted here ↩︎

These questions adapted from ( Clark, Golder, and Golder 2017 ) ↩︎

COMMENTS

Formulating/Extracting Hypotheses
Counterfactuals and Hypothesis Testing in Political Science. World Politics 43 (2): 169-195. ... Examples from work on the causes of World War I, the nonoccurrence of World War III, social revolutions, the breakdown of democratic regimes in Latin America, and the origins of fascism and corporatism in Europe illustrate the use, problems and ...
5 Hypothesis Testing
Learn how to test hypotheses about population parameters using statistical methods and random samples. This chapter explains the Neyman-Pearson paradigm, the central limit theorem, and the idea of significance testing in political science research.
PSCI 3300: Introduction to Political Research
Hypothesis in Political Science "A generalization predicting that a relationship exists between variables. Many generalizations about politics are a sort of folklore. Others proceed from earlier work carried out by social scientists. Within the social sciences most statements about behaviour relate to large groups of people.
Empirical Methods in Political Science: An Introduction
A textbook by Jean Clipperton and other authors that covers the basics of empirical research in political science. It covers topics such as research design, data collection, analysis, and interpretation, with examples and exercises.
1.5 Empirical Political Science
Empirical political science, as described here, is not different from other applications of the scientific method, whether one is examining rocks in geology, birds in botany, or the human mind in psychology.In every science-based course you take, you will observe systematic efforts to develop knowledge by using data to test hypotheses.
PDF From Research Question to Theory to Hypothesis I distribute or post, copy
matter for your theory. The assumption about gravity, for example, is a neces-sary assumption for most theories of war, but natural science gives us reason to believe that gravity holds in all places on Earth. Since I don't know of any cur-rent research on extraplanetary political science, the assumption of continued . Do not copy, post, or
7 Experiments
Learn about the history, methods, and types of experiments in political science, a research design that allows cheating history to observe the effects of interventions. The web page does not address the query about the application of minimum force in political science, which is a concept related to the use of force in international relations.
Counterfactuals and Hypothesis Testing in Political Science
Examples from work on the causes of World War I, the nonoccurrence of World War III, social revolutions, the breakdown of democratic regimes in Latin America, and the origins of fascism and corporatism in Europe illustrate the use, problems and potential of counterfactual argument in small-N-oriented political science research.
Counterfactuals and Hypothesis Testing in Political Science
The examples make clear that counterfactuals matter both when the re-searcher is focusing on one actual case (for example, the outbreak of World War I or the Brazilian military takeover in i964) and when the researcher considers several actual cases (for example, social revolutions or interwar European regime types). The third section returns ...
Quantitative Research Methods for Political Science, Public Policy and
The focus of this book is on using quantitative research methods to test hypotheses and build theory in political science, public policy and public administration. It is designed for advanced undergraduate courses, or introductory and intermediate graduate-level courses. The first part of the book introduces the scientific method, then covers research design, measurement, descriptive ...
Bivariate Hypothesis Testing (Chapter 7)
In this chapter we discuss the basic mechanics of hypothesis testing with three different examples of bivariate hypothesis testing. It is worth noting that, although this type of analysis was the main form of hypothesis testing in the professional journals up through the 1970s, it is seldom used as the primary means of hypothesis testing in the ...
Hypotheses in Political Science Research
This video guides you to developing good hypotheses to guide your expectations when you make observations.
4 Data
This is an introductory college textbook on using empirical methods in political science research. Empirical Methods in Political Science ... 5.5 Steps of Hypothesis Testing; 5.6 Types of ... Democrat, Republican), are all examples of nominal variables. Ordinal variables, on the other hand, can be ordered in a sequence yet we cannot establish ...
PDF Writing in Political Science
Learn how to write a research paper in political science using the scientific method, theory building, and research methodology. This handout explains the key steps, terms, and tips for choosing a topic, formulating a thesis, and testing a theory.
1.3 Political Science: The Systematic Study of Politics
This web page is part of a free textbook on political science, but it has a glitch and cannot be accessed. The web page introduces the systematic study of politics and its main topics, methods, and theories.
POLSC101: Research in Political Science
Learn the basic principles and methods of political science in this survey course. Explore the concepts, ideologies, institutions, and issues of politics in the United States and the world.
3 Theory
Learn the meaning, significance, and building blocks of theory as well as theory-building and theory testing procedures in political science. A theory is a set of logically consistent statements that explain empirical social and political phenomena or relationships.
9 Small N
The field of political science has traditionally focused on the importance of hypothesis testing, causal inference, experiments and the use of large n data. Quantitative methods in all its capacities is without a doubt important, but what can be lost at times is the value of small n methods of inquiry within the field of political science.
Political Science
Learn how to write a science paper in political science, a discipline that studies the struggle of "who gets what, when, how" in different contexts. Find out how political scientists use the scientific method, hypotheses, data, and theories to build arguments and test relationships.
1 Introduction
Learn about the subfields, questions, and methods of political science research. This textbook focuses on quantitative data and empirical methods, such as surveys, experiments, statistics, and game theory.