research about peer teaching

• Original article
• Open access
• Published: 09 April 2020

Why does peer instruction benefit student learning?

• Jonathan G. Tullis 1 &
• Robert L. Goldstone 2  

Cognitive Research: Principles and Implications volume  5 , Article number:  15 ( 2020 ) Cite this article

88k Accesses

52 Citations

52 Altmetric

Metrics details

In peer instruction, instructors pose a challenging question to students, students answer the question individually, students work with a partner in the class to discuss their answers, and finally students answer the question again. A large body of evidence shows that peer instruction benefits student learning. To determine the mechanism for these benefits, we collected semester-long data from six classes, involving a total of 208 undergraduate students being asked a total of 86 different questions related to their course content. For each question, students chose their answer individually, reported their confidence, discussed their answers with their partner, and then indicated their possibly revised answer and confidence again. Overall, students were more accurate and confident after discussion than before. Initially correct students were more likely to keep their answers than initially incorrect students, and this tendency was partially but not completely attributable to differences in confidence. We discuss the benefits of peer instruction in terms of differences in the coherence of explanations, social learning, and the contextual factors that influence confidence and accuracy.

Significance

Peer instruction is widely used in physics instruction across many universities. Here, we examine how peer instruction, or discussing one’s answer with a peer, affects students’ decisions about a class assignment. Across six different university classes, students answered a question, discussed their answer with a peer, and finally answered the question again. Students’ accuracy consistently improved through discussion with a peer. Our peer instruction data show that students were hesitant to switch away from their initial answer and that students did consider both their own confidence and their partner’s confidence when making their final decision, in accord with basic research about confidence in decision making. More broadly, the data reveal that peer discussion helped students select the correct answer by prompting them to create new knowledge. The benefit to student accuracy that arises when students discuss their answers with a partner is a “process gain”, in which working in a group yields better performance than can be predicted from individuals’ performance alone.

Peer instruction is specific evidence-based instructional strategy that is well-known and widely used, particularly in physics (Henderson & Dancy, 2009 ). In fact, peer instruction has been advocated as a part of best methods in science classrooms (Beatty, Gerace, Leonard, & Dufresne, 2006 ; Caldwell, 2007 ; Crouch & Mazur, 2001 ; Newbury & Heiner, 2012 ; Wieman et al., 2009 ) and over a quarter of university physics professors report using peer instruction (Henderson & Dancy, 2009 ). In peer instruction, instructors pose a challenging question to students, students answer the question individually, students discuss their answers with a peer in the class, and finally students answer the question again. There are variations of peer instruction in which instructors show the class’s distribution of answers before discussion (Nielsen, Hansen-Nygård, & Stav, 2012 ; Perez et al., 2010 ), in which students’ answers are graded for participation or for correctness (James, 2006 ), and in which instructors’ norms affect whether peer instruction offers opportunities for answer-seeking or for sense-making (Turpen & Finkelstein, 2007 ).

Despite wide variations in its implementation, peer instruction consistently benefits student learning. Switching classroom structure from didactic lectures to one centered around peer instruction improves learners’ conceptual understanding (Duncan, 2005 ; Mazur, 1997 ), reduces student attrition in difficult courses (Lasry, Mazur, & Watkins, 2008 ), decreases failure rates (Porter, Bailey-Lee, & Simon, 2013 ), improves student attendance (Deslauriers, Schelew, & Wieman, 2011 ), and bolsters student engagement (Lucas, 2009 ) and attitudes to their course (Beekes, 2006 ). Benefits of peer instruction have been found across many fields, including physics (Mazur, 1997 ; Pollock, Chasteen, Dubson, & Perkins, 2010 ), biology (Knight, Wise, & Southard, 2013 ; Smith, Wood, Krauter, & Knight, 2011 ), chemistry (Brooks & Koretsky, 2011 ), physiology (Cortright, Collins, & DiCarlo, 2005 ; Rao & DiCarlo, 2000 ), calculus (Lucas, 2009 ; Miller, Santana-Vega, & Terrell, 2007 ), computer science (Porter et al., 2013 ), entomology (Jones, Antonenko, & Greenwood, 2012 ), and even philosophy (Butchart, Handfield, & Restall, 2009 ). Additionally, benefits of peer instruction have been found at prestigious private universities, two-year community colleges (Lasry et al., 2008 ), and even high schools (Cummings & Roberts, 2008 ). Peer instruction benefits not just the specific questions posed during discussion, but also improves accuracy on later similar problems (e.g., Smith et al., 2009 ).

One of the consistent empirical hallmarks of peer instruction is that students’ answers are more frequently correct following discussion than preceding it. For example, in introductory computer science courses, post-discussion performance was higher on 70 out of 71 questions throughout the semester (Simon, Kohanfars, Lee, Tamayo, & Cutts, 2010 ). Further, gains in performance from discussion are found on many different types of questions, including recall, application, and synthesis questions (Rao & DiCarlo, 2000 ). Performance improvements are found because students are more likely to switch from an incorrect answer to the correct answer than from the correct answer to an incorrect answer. In physics, 59% of incorrect answers switched to correct following discussion, but only 13% of correct answers switched to incorrect (Crouch & Mazur, 2001 ). Other research on peer instruction shows the same patterns: 41% of incorrect answers are switched to correct ones, while only 18% of correct answers are switched to incorrect (Morgan & Wakefield, 2012 ). On qualitative problem-solving questions in physiology, 57% of incorrect answers switched to correct after discussion, and only 7% of correct answers to incorrect (Giuliodori, Lujan, & DiCarlo, 2006 ).

There are two explanations for improvements in pre-discussion to post-discussion accuracy. First, switches from incorrect to correct answers may be driven by selecting the answer from the peer who is more confident. When students discuss answers that disagree, they may choose whichever answer belongs to the more confident peer. Evidence about decision-making and advice-taking substantiates this account. First, confidence is correlated with correctness across many settings and procedures (Finley, Tullis, & Benjamin, 2010 ). Students who are more confident in their answers are typically more likely to be correct. Second, research examining decision-making and advice-taking indicates that (1) the less confident you are, the more you value others’ opinions (Granovskiy, Gold, Sumpter, & Goldstone, 2015 ; Harvey & Fischer, 1997 ; Yaniv, 2004a , 2004b ; Yaniv & Choshen-Hillel, 2012 ) and (2) the more confident the advisor is, the more strongly they influence your decision (Kuhn & Sniezek, 1996 ; Price & Stone, 2004 ; Sah, Moore, & MacCoun, 2013 ; Sniezek & Buckley, 1995 ; Van Swol & Sniezek, 2005 ; Yaniv, 2004b ). Consequently, if students simply choose their final answer based upon whoever is more confident, accuracy should increase from pre-discussion to post-discussion. This explanation suggests that switches in answers should be driven entirely by a combination of one’s own initial confidence and one’s partner’s confidence. In accord with this confidence view, Koriat ( 2015 ) shows that an individual’s confidence typically reflects the group’s most typically given answer. When the answer most often given by group members is incorrect, peer interactions amplify the selection of and confidence in incorrect answers. Correct answers have no special draw. Rather, peer instruction merely amplifies the dominant view through differences in the individual’s confidence.

In a second explanation, working with others may prompt students to verbalize explanations and verbalizations may generate new knowledge. More specifically, as students discuss the questions, they need to create a common representation of the problem and answer. Generating a common representation may compel students to identify gaps in their existing knowledge and construct new knowledge (Schwartz, 1995 ). Further, peer discussion may promote students’ metacognitive processes of detecting and correcting errors in their mental models. Students create more new knowledge and better diagnostic tests of answers together than alone. Ultimately, then, the new knowledge and improved metacognition may make the correct answer appear more compelling or coherent than incorrect options. Peer discussion would draw attention to coherent or compelling answers, more so than students’ initial confidence alone and the coherence of the correct answer would prompt students to switch away from incorrect answers. Similarly, Trouche, Sander, and Mercier ( 2014 ) argue that interactions in a group prompt argumentation and discussion of reasoning. Good arguments and reasoning should be more compelling to change individuals’ answers than confidence alone. Indeed, in a reasoning task known to benefit from careful deliberation, good arguments and the correctness of the answers change partners’ minds more than confidence in one’s answer (Trouche et al., 2014 ). This explanation predicts several distinct patterns of data. First, as seen in prior research, more students should switch from incorrect answers to correct than vice versa. Second, the intrinsic coherence of the correct answer should attract students, so the likelihood of switching answers would be predicted by the correctness of an answer above and beyond differences in initial confidence. Third, initial confidence in an answer should not be as tightly related to initial accuracy as final confidence is to final accuracy because peer discussion should provide a strong test of the coherence of students’ answers. Fourth, because the coherence of an answer is revealed through peer discussion, student confidence should increase more from pre-discussion to post-discussion when they agree on the correct answers compared to agreeing on incorrect answers.

Here, we examined the predictions of these two explanations of peer instruction across six different classes. We specifically examined whether changes in answers are driven exclusively through the confidence of the peers during discussion or whether the coherence of an answer is better constructed and revealed through peer instruction than on one’s own. We are interested in analyzing cognitive processes at work in a specific, but common, implementation of classroom-based peer instruction; we do not intend to make general claims about all kinds of peer instruction or to evaluate the long-term effectiveness of peer instruction. This research is the first to analyze how confidence in one’s answer relates to answer-switching during peer instruction and tests the impact of peer instruction in new domains (i.e., psychology and educational psychology classes).

Participants

Students in six different classes participated as part of their normal class procedures. More details about these classes are presented in Table  1 . The authors served as instructors for these classes. Across the six classes, 208 students contributed a total of 1657 full responses to 86 different questions.

The instructors of the courses developed multiple-choice questions related to the ongoing course content. Questions were aimed at testing students’ conceptual understanding, rather than factual knowledge. Consequently, questions often tested whether students could apply ideas to new settings or contexts. An example of a cognitive psychology question used is: Which is a fixed action pattern (not a reflex)?

Knee jerks up when patella is hit

Male bowerbirds building elaborate nests [correct]

Eye blinks when air is blown on it

Can play well learned song on guitar even when in conversation

The procedures for peer instruction across the six different classes followed similar patterns. Students were presented with a multiple-choice question. First, students read the question on their own, chose their answer, and reported their confidence in their answer on a scale of 1 “Not at all confident” to 10 “Highly confident”. Students then paired up with a neighbor in their class and discussed the question with their peer. After discussion, students answered the question and reported the confidence for a second time. The course instructor indicated the correct answer and discussed the reasoning for the answer after all final answers had been submitted. Instruction was paced based upon how quickly students read and answered questions. Most student responses counted towards their participation grade, regardless of the correctness of their answer (the last question in each of the cognitive psychology classes was graded for correctness).

There were small differences in procedures between classes. Students in the cognitive psychology classes input their responses using classroom clickers, but those in other classes wrote their responses on paper. Further, students in the cognitive psychology classes explicitly reported their partner’s answer and confidence, while students in other classes only reported the name of their partner (the partners’ data were aligned during data recording). The cognitive psychology students then were required to mention their own answer and their confidence to their partner during peer instruction; students in other classes were not required to tell their answer or their confidence to their peer. Finally, the questions appeared at any point during the class period for the cognitive psychology classes, while the questions typically happened at the beginning of each class for the other classes.

Analytic strategy

Data are available on the OpenScienceFramework: https://mfr.osf.io/render?url=https://osf.io/5qc46/?action=download%26mode=render .

For most of our analyses we used linear mixed-effects models (Baayen, Davidson, & Bates, 2008 ; Murayama, Sakaki, Yan, & Smith, 2014 ). The unit of analysis in a mixed-effect model is the outcome of a single trial (e.g., whether or not a particular question was answered correctly by a particular participant). We modeled these individual trial-level outcomes as a function of multiple fixed effects - those of theoretical interest - and multiple random effects - effects for which the observed levels are sampled out of a larger population (e.g., questions, students, and classes sampled out of a population of potential questions, students, and classes).

Linear mixed-effects models solve four statistical problems involved with the data of peer instruction. First, there is large variability in students’ performance and the difficulty of questions across students and classes. Mixed-effect models simultaneously account for random variation both across participants and across items (Baayen et al., 2008 ; Murayama et al., 2014 ). Second, students may miss individual classes and therefore may not provide data across every item. Similarly, classes varied in how many peer instruction questions were posed throughout the semester and the number of students enrolled. Mixed-effects models weight each response equally when drawing conclusions (rather than weighting each student or question equally) and can easily accommodate missing data. Third, we were interested in how several different characteristics influenced students’ performance. Mixed effects models can include multiple predictors simultaneously, which allows us to test the effect of one predictor while controlling for others. Finally, mixed effects models can predict the log odds (or logit) of a correct answer, which is needed when examining binary outcomes (i.e., correct or incorrect; Jaeger, 2008 ).

We fit all models in R using the lmer() function of the lme4 package (Bates, Maechler, Bolker, & Walker, 2015 ). For each mixed-effect model, we included random intercepts that capture baseline differences in difficulty of questions, in classes, and in students, in addition to multiple fixed effects of theoretical interest. In mixed-effect models with hundreds of observations, the t distribution effectively converges to the normal, so we compared the t statistic to the normal distribution for analyses involving continuous outcomes (i.e., confidence; Baayen, 2008 ). P values can be directly obtained from Wald z statistics for models with binary outcomes (i.e., correctness).

Does accuracy change through discussion?

First, we examined how correctness changed across peer discussion. A logit model predicting correctness from time point (pre-discussion to post-discussion) revealed that the odds of correctness increased by 1.57 times (95% confidence interval (conf) 1.31–1.87) from pre-discussion to post-discussion, as shown in Table  2 . In fact, 88% of students showed an increase or no change in accuracy from pre-discussion to post-discussion. Pre-discussion to post-discussion performance for each class is shown in Table  3 . We further examined how accuracy changed from pre-discussion to post-discussion for each question and the results are plotted in Fig.  1 . The data show a consistent improvement in accuracy from pre-discussion to post-discussion across all levels of initial difficulty.

The relationship between pre-discussion accuracy (x axis) and post-discussion accuracy (y axis). Each point represents a single question. The solid diagonal line represents equal pre-discussion and post-discussion accuracy; points above the line indicate improvements in accuracy and points below represent decrements in accuracy. The dashed line indicates the line of best fit for the observed data

We examined how performance increased from pre-discussion to post-discussion by tracing the correctness of answers through the discussion. Figure  2 tracks the percent (and number of items) correct from pre-discussion to post-discussion. The top row shows whether students were initially correct or incorrect in their answer; the middle row shows whether students agreed or disagreed with their partner; the last row show whether students were correct or incorrect after discussion. Additionally, Fig. 2 shows the confidence associated with each pathway. The bottow line of each entry shows the students’ average confidence; in the middle white row, the confidence reported is the average of the peer’s confidence.

The pathways of answers from pre-discussion (top row) to post-discussion (bottom row). Percentages indicate the portion of items from the category immediately above in that category, the numbers in brackets indicate the raw numbers of items, and the numbers at the bottom of each entry indicate the confidence associated with those items. In the middle, white row, confidence values show the peer’s confidence. Turquoise indicates incorrect answers and yellow indicates correct answers

Broadly, only 5% of correct answers were switched to incorrect, while 28% of incorrect answers were switched to correct following discussion. Even for the items in which students were initially correct but disagreed with their partner, only 21% of answers were changed to incorrect answers after discussion. However, out of the items where students were initially incorrect and disagreed with their partner, 42% were changed to the correct answer.

Does confidence predict switching?

Differences in the amount of switching to correct or incorrect answers could be driven solely by differences in confidence, as described in our first theory mentioned earlier. For this theory to hold, answers with greater confidence must have a greater likelihood of being correct. To examine whether initial confidence is associated with initial correctness, we calculated the gamma correlation between correctness and confidence in the answer before discussion, as shown in the first column of Table  4 . The average gamma correlation between initial confidence and initial correctness (mean (M) = 0.40) was greater than zero, t (160) = 8.59, p  < 0.001, d  = 0.68, indicating that greater confidence was associated with being correct.

Changing from an incorrect to a correct answer, then, may be driven entirely by selecting the answer from the peer with the greater confidence during discussion, even though most of the students in our sample were not required to explicitly disclose their confidence to their partner during discussion. We examined how frequently students choose the more confident answer when peers disagree. When peers disagreed, students’ final answers aligned with the more confident peer only 58% of the time. Similarly, we tested what the performance would be if peers always picked the answer of the more confident peer. If peers always chose the more confident answer during discussion, the final accuracy would be 69%, which is significantly lower than actual final accuracy (M = 72%, t (207) = 2.59, p  = 0.01, d  = 0.18). While initial confidence is related to accuracy, these results show that confidence is not the only predictor of switching answers.

Does correctness predict switching beyond confidence?

Discussion may reveal information about the correctness of answers by generating new knowledge and testing the coherence of each possible answer. To test whether the correctness of an answer added predictive power beyond the confidence of the peers involved in discussion, we analyzed situations in which students disagreed with their partner. Out of the instances when partners initially disagreed, we predicted the likelihood of keeping one’s answer based upon one’s own confidence, the partner’s confidence, and whether one’s answer was initially correct. The results of a model predicting whether students keep their answers is shown in Table  5 . For each increase in a point of one’s own confidence, the odds of keeping one’s answer increases 1.25 times (95% conf 1.13–1.38). For each decrease in a point of the partner’s confidence, the odds of keeping one’s answer increased 1.19 times (1.08–1.32). The beta weight for one’s confidence did not differ from the beta weight of the partner’s confidence, χ 2  = 0.49, p  = 0.48. Finally, if one’s own answer was correct, the odds of keeping one’s answer increased 4.48 times (2.92–6.89). In other words, the more confident students were, the more likely they were to keep their answer; the more confident their peer was, the more likely they were to change their answer; and finally, if a student was correct, they were more likely to keep their answer.

To illustrate this relationship, we plotted the probability of keeping one’s own answer as a function of the difference between one’s own and their partner’s confidence for initially correct and incorrect answers. As shown in Fig.  3 , at every confidence level, being correct led to equal or more frequently keeping one’s answer than being incorrect.

The probability of keeping one’s answer in situations where one’s partner initially disagreed as a function of the difference between partners’ levels of confidence. Error bars indicate the standard error of the proportion and are not shown when the data are based upon a single data point

As another measure of whether discussion allows learners to test the coherence of the correct answer, we analyzed how discussion impacted confidence when partners’ answers agreed. We predicted confidence in answers by the interaction of time point (i.e., pre-discussion versus post-discussion) and being initially correct for situations in which peers initially agreed on their answer. The results, displayed in Table  6 , show that confidence increased from pre-discussion to post-discussion by 1.08 points and that confidence was greater for initially correct answers (than incorrect answers) by 0.78 points. As the interaction between time point and initial correctness shows, confidence increased more from pre-discussion to post-discussion when students were initially correct (as compared to initially incorrect). To illustrate this relationship, we plotted pre-confidence against post-confidence for initially correct and initially incorrect answers when peers agreed (Fig.  4 ). Each plotted point represents a student; the diagonal blue line indicates no change between pre-confidence and post-confidence. The graph reflects that confidence increases more from pre-discussion to post-discussion for correct answers than for incorrect answers, even when we only consider cases where peers agreed.

The relationship between pre-discussion and post-discussion confidence as a function of the accuracy of an answer when partners agreed. Each dot represents a student

If students engage in more comprehensive answer testing during discussion than before, the relationship between confidence in their answer and the accuracy of their answer should be stronger following discussion than it is before. We examined whether confidence accurately reflected correctness before and after discussion. To do so, we calculated the gamma correlation between confidence and accuracy, as is typically reported in the literature on metacognitive monitoring (e.g., Son & Metcalfe, 2000 ; Tullis & Fraundorf, 2017 ). Across all students, the resolution of metacognitive monitoring increases from pre-discussion to post-discussion ( t (139) = 2.98, p  = 0.003, d  = 0.24; for a breakdown of gamma calculations for each class, see Table 4 ). Confidence was more accurately aligned with accuracy following discussion than preceding it. The resolution between student confidence and correctness increases through discussion, suggesting that discussion offers better coherence testing than answering alone.

To examine why peer instruction benefits student learning, we analyzed student answers and confidence before and after discussion across six psychology classes. Discussing a question with a partner improved accuracy across classes and grade levels with small to medium-sized effects. Questions of all difficulty levels benefited from peer discussion; even questions where less than half of students originally answered correctly saw improvements from discussion. Benefits across the spectrum of question difficulty align with prior research showing improvements when even very few students initially know the correct answer (Smith et al., 2009 ). More students switched from incorrect answers to correct answers than vice versa, leading to an improvement in accuracy following discussion. Answer switching was driven by a student’s own confidence in their answer and their partner’s confidence. Greater confidence in one’s answer indicated a greater likelihood of keeping the answer; a partner’s greater confidence increased the likelihood of changing to their answer.

Switching answers depended on more than just confidence: even when accounting for students’ confidence levels, the correctness of the answer impacted switching behavior. Across several measures, our data showed that the correctness of an answer carried weight beyond confidence. For example, the correctness of the answer predicted whether students switched their initial answer during peer disagreements, even after taking the confidence of both partners into account. Further, students’ confidence increased more when partners agreed on the correct answer compared to when they agreed on an incorrect answer. Finally, although confidence increased from pre-discussion to post-discussion when students changed their answers from incorrect to the correct ones, confidence decreased when students changed their answer away from the correct one. A plausible interpretation of this difference is that when students switch from a correct answer to an incorrect one, their decrease in confidence reflects the poor coherence of their final incorrect selection.

Whether peer instruction resulted in optimal switching behaviors is debatable. While accuracy improved through discussion, final accuracy was worse than if students had optimally switched their answers during discussion. If students had chosen the correct answer whenever one of the partners initially chose it, the final accuracy would have been significantly higher (M = 0.80 (SD = 0.19)) than in our data (M = 0.72 (SD = 0.24), t (207) = 6.49, p  < 0.001, d  = 0.45). While this might be interpreted as “process loss” (Steiner, 1972 ; Weldon & Bellinger, 1997 ), that would assume that there is sufficient information contained within the dyad to ascertain the correct answer. One individual selecting the correct answer is inadequate for this claim because they may not have a compelling justification for their answer. When we account for differences in initial confidence, students’ final accuracy was better than expected. Students’ final accuracy was better than that predicted from a model in which students always choose the answer of the more confident peer. This over-performance, often called “process gain”, can sometimes emerge when individuals collaborate to create or generate new knowledge (Laughlin, Bonner, & Miner, 2002 ; Michaelsen, Watson, & Black, 1989 ; Sniezek & Henry, 1989 ; Tindale & Sheffey, 2002 ). Final accuracy reveals that students did not simply choose the answer of the more confident student during discussion; instead, students more thoroughly probed the coherence of answers and mental models during discussion than they could do alone.

Students’ final accuracy emerges from the interaction between the pairs of students, rather than solely from individuals’ sequestered knowledge prior to discussion (e.g. Wegner, Giuliano, & Hertel, 1985 ). Schwartz ( 1995 ) details four specific cognitive products that can emerge through working in dyads. Specifically, dyads force verbalization of ideas through discussion, and this verbalization facilitates generating new knowledge. Students may not create a coherent explanation of their answer until they engage in discussion with a peer. When students create a verbal explanation of their answer to discuss with a peer, they can identify knowledge gaps and construct new knowledge to fill those gaps. Prior research examining the content of peer interactions during argumentation in upper-level biology classes has shown that these kinds of co-construction happen frequently; over three quarters of statements during discussion involve an exchange of claims and reasoning to support those claims (Knight et al., 2013 ). Second, dyads have more information processing resources than individuals, so they can solve more complex problems. Third, dyads may foster greater motivation than individuals. Finally, dyads may stimulate the creation of new, abstract representations of knowledge, above and beyond what one would expect from the level of abstraction created by individuals. Students need to communicate with their partner; to create common ground and facilitate discourse, dyads negotiate common representations to coordinate different perspectives. The common representations bridge multiple perspectives, so they lose idiosyncratic surface features of individuals’ representation. Working in pairs generates new knowledge and tests of answers that could not be predicted from individuals’ performance alone.

More broadly, teachers often put students in groups so that they can learn from each other by giving and receiving help, recognizing contradictions between their own and others’ perspectives, and constructing new understandings from divergent ideas (Bearison, Magzamen, & Filardo, 1986 ; Bossert, 1988-1989 ; Brown & Palincsar, 1989 ; Webb & Palincsar, 1996 ). Giving explanations to a peer may encourage explainers to clarify or reorganize information, recognize and rectify gaps in understandings, and build more elaborate interpretations of knowledge than they would have alone (Bargh & Schul, 1980 ; Benware & Deci, 1984 ; King, 1992 ; Yackel, Cobb, & Wood, 1991 ). Prompting students to explain why and how problems are solved facilitates conceptual learning more than reading the problem solutions twice without self-explanations (Chi, de Leeuw, Chiu, & LaVancher, 1994 ; Rittle-Johnson, 2006 ; Wong, Lawson, & Keeves, 2002 ). Self-explanations can prompt students to retrieve, integrate, and modify their knowledge with new knowledge; self-explanations can also help students identify gaps in their knowledge (Bielaczyc, Pirolli, & Brown, 1995 ; Chi & Bassock, 1989 ; Chi, Bassock, Lewis, Reimann, & Glaser, 1989 ; Renkl, Stark, Gruber, & Mandl, 1998 ; VanLehn, Jones, & Chi, 1992 ; Wong et al., 2002 ), detect and correct errors, and facilitate deeper understanding of conceptual knowledge (Aleven & Koedinger, 2002 ; Atkinson, Renkl, & Merrill, 2003 ; Chi & VanLehn, 2010 ; Graesser, McNamara, & VanLehn, 2005 ). Peer instruction, while leveraging these benefits of self-explanation, also goes beyond them by involving what might be called “other-explanation” processes - processes recruited not just when explaining a situation to oneself but to others. Mercier and Sperber ( 2019 ) argue that much of human reason is the result of generating explanations that will be convincing to other members of one’s community, thereby compelling others to act in the way that one wants.

Conversely, students receiving explanations can fill in gaps in their own understanding, correct misconceptions, and construct new, lasting knowledge. Fellow students may be particularly effective explainers because they can better take the perspective of their peer than the teacher (Priniski & Horne, 2019 ; Ryskin, Benjamin, Tullis, & Brown-Schmidt, 2015 ; Tullis, 2018 ). Peers may be better able than expert teachers to explain concepts in familiar terms and direct peers’ attention to the relevant features of questions that they do not understand (Brown & Palincsar, 1989 ; Noddings, 1985 ; Vedder, 1985 ; Vygotsky, 1981 ).

Peer instruction may benefit from the generation of explanations, but social influences may compound those benefits. Social interactions may help students monitor and regulate their cognition better than self-explanations alone (e.g., Jarvela et al., 2015 ; Kirschner, Kreijns, Phielix, & Fransen, 2015 ; Kreijns, Kirschner, & Vermeulen, 2013 ; Phielix, Prins, & Kirschner, 2010 ; Phielix, Prins, Kirschner, Erkens, & Jaspers, 2011 ). Peers may be able to judge the quality of the explanation better than the explainer. In fact, recent research suggests that peer instruction facilitates learning even more than self-explanations (Versteeg, van Blankenstein, Putter, & Steendijk, 2019 ).

Not only does peer instruction generate new knowledge, but it may also improve students’ metacognition. Our data show that peer discussion prompted more thorough testing of the coherence of the answers. Specifically, students’ confidences were better aligned with accuracy following discussion than before. Improvements in metacognitive resolution indicate that discussion provides more thorough testing of answers and ideas than does answering questions on one’s own. Discussion facilitates the metacognitive processes of detecting errors and assessing the coherence of an answer.

Agreement among peers has important consequences for final behavior. For example, when peers agreed, students very rarely changed their answer (less than 3% of the time). Further, large increases in confidence occurred when students agreed (as compared to when they disagreed). Alternatively, disagreements likely engaged different discussion processes and prompted students to combine different answers. Whether students weighed their initial answer more than their partner’s initial answer remains debatable. When students disagreed with their partner, they were more likely to stick with their own answer than switch; they kept their own answer 66% of the time. Even when their partner was more confident, students only switched to their partner’s answer 50% of the time. The low rate of switching during disagreements suggests that students weighed their own answer more heavily than their partner’s answer. In fact, across prior research, deciders typically weigh their own thoughts more than the thoughts of an advisor (Harvey, Harries, & Fischer, 2000 ; Yaniv & Kleinberger, 2000 ).

Interestingly, peers agreed more frequently than expected by chance. When students were initially correct (64% of the time), 78% of peers agreed. When students were initially incorrect (36% of the time), peers agreed 43% of the time. Pairs of students, then, agree more than expected by a random distribution of answers throughout the classroom. These data suggest that students group themselves into pairs based upon likelihood of sharing the same answer. Further, these data suggest that student understanding is not randomly distributed throughout the physical space of the classroom. Across all classes, students were instructed to work with a neighbor to discuss their answer. Given that neighbors agreed more than predicted by chance, students seem to tend to sit near and pair with peers that share their same levels of understanding. Our results from peer instruction reveal that students physically locate themselves near students of similar abilities. Peer instruction could potentially benefit from randomly pairing students together (i.e. not with a physically close neighbor) to generate the most disagreements and generative activity during discussion.

Learning through peer instruction may involve deep processing as peers actively challenge each other, and this deep processing may effectively support long-term retention. Future research can examine the persistence of gains in accuracy from peer instruction. For example, whether errors that are corrected during peer instruction stay corrected on later retests of the material remains an open question. High and low-confidence errors that are corrected during peer instruction may result in different long-term retention of the correct answer; more specifically, the hypercorrection effect suggests that errors committed with high confidence are more likely to be corrected on subsequent tests than errors with low confidence (e.g., Butler, Fazio, & Marsh, 2011 ; Butterfield & Metcalfe, 2001 ; Metcalfe, 2017 ). Whether hypercorrection holds for corrections from classmates during peer instruction (rather than from an absolute authority) could be examined in the future.

The influence of partner interaction on accuracy may depend upon the domain and kind of question posed to learners. For simple factual or perceptual questions, partner interaction may not consistently benefit learning. More specifically, partner interaction may amplify and bolster wrong answers when factual or perceptual questions lead most students to answer incorrectly (Koriat, 2015 ). However, for more “intellective tasks,” interactions and arguments between partners can produce gains in knowledge (Trouche et al., 2014 ). For example, groups typically outperform individuals for reasoning tasks (Laughlin, 2011 ; Moshman & Geil, 1998 ), math problems (Laughlin & Ellis, 1986 ), and logic problems (Doise & Mugny, 1984; Perret-Clermont, 1980 ). Peer instruction questions that allow for student argumentation and reasoning, therefore, may have the best benefits in student learning.

The underlying benefits of peer instruction extend beyond the improvements in accuracy seen from pre-discussion to post-discussion. Peer instruction prompts students to retrieve information from long-term memory, and these practice tests improve long-term retention of information (Roediger III & Karpicke, 2006 ; Tullis, Fiechter, & Benjamin, 2018 ). Further, feedback provided by instructors following peer instruction may guide students to improve their performance and correct misconceptions, which should benefit student learning (Bangert-Drowns, Kulik, & Kulik, 1991 ; Thurlings, Vermeulen, Bastiaens, & Stijnen, 2013 ). Learners who engage in peer discussion can use their new knowledge to solve new, but similar problems on their own (Smith et al., 2009 ). Generating new knowledge and revealing gaps in knowledge through peer instruction, then, effectively supports students’ ability to solve novel problems. Peer instruction can be an effective tool to generate new knowledge through discussion between peers and improve student understanding and metacognition.

Availability of data and materials

As described below, data and materials are available on the OpenScienceFramework: https://mfr.osf.io/render?url=https://osf.io/5qc46/?action=download%26mode=render .

Aleven, V., & Koedinger, K. R. (2002). An effective metacognitive strategy: Learning by doing and explaining with a computer based cognitive tutor. Cognitive Science , 26 , 147–179.

Article   Google Scholar  

Atkinson, R. K., Renkl, A., & Merrill, M. M. (2003). Transitioning from studying examples to solving problems: Effects of self-explanation prompts and fading worked-out steps. Journal of Educational Psychology , 95 , 774–783.

Baayen, R. H. (2008). Analyzing linguistic data: A practical introduction to statistics . Cambridge: Cambridge University Press.

Baayen, R. H., Davidson, D. J., & Bates, D. M. (2008). Mixed-effects modeling with crossed random effects for subjects and items. Journal of Memory and Language , 59 , 390–412.

Bangert-Drowns, R. L., Kulik, J. A., & Kulik, C.-L. C. (1991). Effects of frequent classroom testing. Journal of Educational Research , 85 , 89–99.

Bargh, J. A., & Schul, Y. (1980). On the cognitive benefit of teaching. Journal of Educational Psychology , 72 , 593–604.

Bates, D., Maechler, M., Bolker, B., & Walker, S. (2015). Fitting linear mixed-effects models using lme4. Journal of Statistical Software , 67 , 1–48.

Bearison, D. J., Magzamen, S., & Filardo, E. K. (1986). Sociocognitive conflict and cognitive growth in young children. Merrill-Palmer Quarterly , 32 (1), 51–72.

Google Scholar  

Beatty, I. D., Gerace, W. J., Leonard, W. J., & Dufresne, R. J. (2006). Designing effective questions for classroom response system teaching. American Journal of Physics , 74 (1), 31e39.

Beekes, W. (2006). The “millionaire” method for encouraging participation. Active Learning in Higher Education , 7 , 25–36.

Benware, C. A., & Deci, E. L. (1984). Quality of learning with an active versus passive motivational set. American Educational Research Journal , 21 , 755–765.

Bielaczyc, K., Pirolli, P., & Brown, A. L. (1995). Training in self-explanation and self regulation strategies: Investigating the effects of knowledge acquisition activities on problem solving. Cognition and Instruction , 13 , 221–251.

Bossert, S. T. (1988-1989). Cooperative activities in the classroom. Review of Research in Education , 15 , 225–252.

Brooks, B. J., & Koretsky, M. D. (2011). The influence of group discussion on students’ responses and confidence during peer instruction. Journal of Chemistry Education , 88 , 1477–1484.

Brown, A. L., & Palincsar, A. S. (1989). Guided, cooperative learning and individual knowledge acquisition. In L. B. Resnick (Ed.), Knowing, learning, and instruction: essays in honor of Robert Glaser , (pp. 393–451). Hillsdale: Erlbaum.

Butchart, S., Handfield, T., & Restall, G. (2009). Using peer instruction to teach philosophy, logic and critical thinking. Teaching Philosophy , 32 , 1–40.

Butler, A. C., Fazio, L. K., & Marsh, E. J. (2011). The hypercorrection effect persists over a week, but high-confidence errors return. Psychonomic Bulletin & Review , 18 (6), 1238–1244.

Butterfield, B., & Metcalfe, J. (2001). Errors committed with high confidence are hypercorrected. Journal of Experimental Psychology: Learning, Memory, and Cognition , 27 (6), 1491.

PubMed   Google Scholar  

Caldwell, J. E. (2007). Clickers in the large classroom: current research and best-practice tips. CBE-Life Sciences Education , 6 (1), 9–20.

Article   PubMed   PubMed Central   Google Scholar  

Chi, M., & VanLehn, K. A. (2010). Meta-cognitive strategy instruction in intelligent tutoring systems: How, when and why. Journal of Educational Technology and Society , 13 , 25–39.

Chi, M. T. H., & Bassock, M. (1989). Learning from examples via self-explanations. In L. B. Resnick (Ed.), Knowing, learning, and instruction: Essays in honor of Robert Glaser , (pp. 251–282). Hillsdale: Erlbaum.

Chi, M. T. H., Bassock, M., Lewis, M., Reimann, P., & Glaser, R. (1989). Self-explanations: How students study and use examples in learning to solve problems. Cognitive Science , 13 , 145–182.

Chi, M. T. H., de Leeuw, N., Chiu, M. H., & LaVancher, C. (1994). Eliciting self-explanations improves understanding. Cognitive Science , 18 , 439–477.

Cortright, R. N., Collins, H. L., & DiCarlo, S. E. (2005). Peer instruction enhanced meaningful learning: Ability to solve novel problems. Advances in Physiology Education , 29 , 107–111.

Article   PubMed   Google Scholar  

Crouch, C. H., & Mazur, E. (2001). Peer instruction: Ten years of experience and results. American Journal of Physics , 69 , 970–977.

Cummings, K., & Roberts, S. (2008). A study of peer instruction methods with school physics students. In C. Henderson, M. Sabella, & L. Hsu (Eds.), Physics education research conference , (pp. 103–106). College Park: American Institute of Physics.

Deslauriers, L., Schelew, E., & Wieman, C. (2011). Improved learning in a large-enrollment physics class. Science , 332 , 862–864.

Duncan, D. (2005). Clickers in the classroom: How to enhance science teaching using classroom response systems . San Francisco: Pearson/Addison-Wesley.

Finley, J. R., Tullis, J. G., & Benjamin, A. S. (2010). Metacognitive control of learning and remembering. In M. S. Khine, & I. M. Saleh (Eds.), New science of learning: Cognition, computers and collaborators in education . New York: Springer Science & Business Media, LLC.

Giuliodori, M. J., Lujan, H. L., & DiCarlo, S. E. (2006). Peer instruction enhanced student performance on qualitative problem solving questions. Advances in Physiology Education , 30 , 168–173.

Graesser, A. C., McNamara, D., & VanLehn, K. (2005). Scaffolding deep comprehension strategies through AutoTutor and iSTART. Educational Psychologist , 40 , 225–234.

Granovskiy, B., Gold, J. M., Sumpter, D., & Goldstone, R. L. (2015). Integration of social information by human groups. Topics in Cognitive Science , 7 , 469–493.

Harvey, N., & Fischer, I. (1997). Taking advice: Accepting help, improving judgment, and sharing responsibility. Organizational Behavior and Human Decision Processes , 70 , 117–133.

Harvey, N., Harries, C., & Fischer, I. (2000). Using advice and assessing its quality. Organizational Behavior and Human Decision Processes , 81 , 252–273.

Henderson, C., & Dancy, M. H. (2009). The impact of physics education research on the teaching of introductory quantitative physics in the United States. Physical Review Special Topics: Physics Education Research , 5 (2), 020107.

Jaeger, T. F. (2008). Categorical data analysis: away from ANOVAs (transformation or not) and towards logit mixed models. Journal of Memory and Language , 59 , 434–446.

James, M. C. (2006). The effect of grading incentive on student discourse in peer instruction. American Journal of Physics , 74 (8), 689–691.

Jarvela, S., Kirschner, P., Panadero, E., Malmberg, J., Phielix, C., Jaspers, J., … Jarvenoja, H. (2015). Enhancing socially shared regulation in collaborative learning groups: Designing for CSCL regulation tools. Educational Technology Research and Development , 63 (1), 125e142.

Jones, M. E., Antonenko, P. D., & Greenwood, C. M. (2012). The impact of collaborative and individualized student response system strategies on learner motivation, metacognition, and knowledge transfer. Journal of Computer Assisted Learning , 28 (5), 477–487.

King, A. (1992). Facilitating elaborative learning through guided student-generated questioning. Educational Psychologist , 27 , 111–126.

Kirschner, P. A., Kreijns, K., Phielix, C., & Fransen, J. (2015). Awareness of cognitive and social behavior in a CSCL environment. Journal of Computer Assisted Learning , 31 (1), 59–77.

Knight, J. K., Wise, S. B., & Southard, K. M. (2013). Understanding clicker discussions: student reasoning and the impact of instructional cues. CBE-Life Sciences Education , 12 , 645–654.

Koriat, A. (2015). When two heads are better than one and when they can be worse: The amplification hypothesis. Journal of Experimental Psychology: General , 144 , 934–950. https://doi.org/10.1037/xge0000092 .

Kreijns, K., Kirschner, P. A., & Vermeulen, M. (2013). Social aspects of CSCL environments: A research framework. Educational Psychologist , 48 (4), 229e242.

Kuhn, L. M., & Sniezek, J. A. (1996). Confidence and uncertainty in judgmental forecasting: Differential effects of scenario presentation. Journal of Behavioral Decision Making , 9 , 231–247.

Lasry, N., Mazur, E., & Watkins, J. (2008). Peer instruction: From Harvard to the two-year college. American Journal of Physics , 76 (11), 1066–1069.

Laughlin, P. R. (2011). Group problem solving. Princeton: Princeton University Press.

Book   Google Scholar  

Laughlin, P. R., Bonner, B. L., & Miner, A. G. (2002). Groups perform better than individuals on letters-to-numbers problems. Organisational Behaviour and Human Decision Processes , 88 , 605–620.

Laughlin, P. R., & Ellis, A. L. (1986). Demonstrability and social combination processes on mathematical intellective tasks. Journal of Experimental Social Psychology, 22, 177–189.

Lucas, A. (2009). Using peer instruction and i-clickers to enhance student participation in calculus. Primus , 19 (3), 219–231.

Mazur, E. (1997). Peer instruction: A user’s manual . Upper Saddle River: Prentice Hall.

Mercier, H., & Sperber, D. (2019). The enigma of reason . Cambridge: Harvard University Press.

Metcalfe, J. (2017). Learning from errors. Annual Review of Psychology , 68 , 465–489.

Michaelsen, L. K., Watson, W. E., & Black, R. H. (1989). Realistic test of individual versus group decision making. Journal of Applied Psychology , 64 , 834–839.

Miller, R. L., Santana-Vega, E., & Terrell, M. S. (2007). Can good questions and peer discussion improve calculus instruction? Primus , 16 (3), 193–203.

Morgan, J. T., & Wakefield, C. (2012). Who benefits from peer conversation? Examining correlations of clicker question correctness and course performance. Journal of College Science Teaching , 41 (5), 51–56.

Moshman, D., & Geil, M. (1998). Collaborative reasoning: Evidence for collective rationality. Thinking and Reasoning, 4, 231–248.

Murayama, K., Sakaki, M., Yan, V. X., & Smith, G. M. (2014). Type I error inflation in the traditional by-participant analysis to metamemory accuracy: A generalized mixed-effects model perspective. Journal of Experimental Psychology: Learning, Memory, and Cognition , 40 , 1287–1306.

Newbury, P., & Heiner, C. (2012). Ready, set, react! getting the most out of peer instruction using clickers. Retrieved October 28, 2015, from http://www.cwsei.ubc.ca/Files/ReadySetReact_3fold.pdf .

Nielsen, K. L., Hansen-Nygård, G., & Stav, J. B. (2012). Investigating peer instruction: how the initial voting session affects students’ experiences of group discussion. ISRN Education , 2012 , article 290157.

Noddings, N. (1985). Small groups as a setting for research on mathematical problem solving. In E. A. Silver (Ed.), Teaching and learning mathematical problem solving , (pp. 345–360). Hillsdale: Erlbaum.

Perret-Clermont, A. N. (1980). Social Interaction and Cognitive Development in Children. London: Academic Press.

Perez, K. E., Strauss, E. A., Downey, N., Galbraith, A., Jeanne, R., Cooper, S., & Madison, W. (2010). Does displaying the class results affect student discussion during peer instruction? CBE Life Sciences Education , 9 , 133–140.

Phielix, C., Prins, F. J., & Kirschner, P. A. (2010). Awareness of group performance in a CSCL-environment: Effects of peer feedback and reflection. Computers in Human Behavior , 26 (2), 151–161.

Phielix, C., Prins, F. J., Kirschner, P. A., Erkens, G., & Jaspers, J. (2011). Group awareness of social and cognitive performance in a CSCL environment: Effects of a peer feedback and reflection tool. Computers in Human Behavior , 27 (3), 1087–1102.

Pollock, S. J., Chasteen, S. V., Dubson, M., & Perkins, K. K. (2010). The use of concept tests and peer instruction in upper-division physics. In M. Sabella, C. Singh, & S. Rebello (Eds.), AIP conference proceedings , (vol. 1289, p. 261). New York: AIP Press.

Porter, L., Bailey-Lee, C., & Simon, B. (2013). Halving fail rates using peer instruction: A study of four computer science courses. In SIGCSE ‘13: Proceedings of the 44th ACM technical symposium on computer science education , (pp. 177–182). New York: ACM Press.

Price, P. C., & Stone, E. R. (2004). Intuitive evaluation of likelihood judgment producers. Journal of Behavioral Decision Making , 17 , 39–57.

Priniski, J. H., & Horne, Z. (2019). Crowdsourcing effective educational interventions. In A. K. Goel, C. Seifert, & C. Freska (Eds.), Proceedings of the 41st annual conference of the cognitive science society . Austin: Cognitive Science Society.

Rao, S. P., & DiCarlo, S. E. (2000). Peer instruction improves performance on quizzes. Advances in Physiological Education , 24 , 51–55.

Renkl, A., Stark, R., Gruber, H., & Mandl, H. (1998). Learning from worked-out examples: The effects of example variability and elicited self-explanations. Contemporary Educational Psychology , 23 , 90–108.

Rittle-Johnson, B. (2006). Promoting transfer: Effects of self-explanation and direct instruction. Child Development , 77 , 1–15.

Roediger III, H. L., & Karpicke, J. D. (2006). Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science , 17 , 249–255.

Ryskin, R., Benjamin, A. S., Tullis, J. G., & Brown-Schmidt, S. (2015). Perspective-taking in comprehension, production, and memory: An individual differences approach. Journal of Experimental Psychology: General , 144 , 898–915.

Sah, S., Moore, D. A., & MacCoun, R. J. (2013). Cheap talk and credibility: The consequences of confidence and accuracy on advisor credibility and persuasiveness. Organizational Behavior and Human Decision Processes , 121 , 246–255.

Schwartz, D. L. (1995). The emergence of abstract representations in dyad problem solving. The Journal of the Learning Sciences , 4 , 321–354.

Simon, B., Kohanfars, M., Lee, J., Tamayo, K., & Cutts, Q. (2010). Experience report: peer instruction in introductory computing. In Proceedings of the 41st SIGCSE technical symposium on computer science education .

Smith, M. K., Wood, W. B., Adams, W. K., Wieman, C., Knight, J. K., Guild, N., & Su, T. T. (2009). Why peer discussion improves student performance on in-class concept questions. Science , 323 , 122–124.

Smith, M. K., Wood, W. B., Krauter, K., & Knight, J. K. (2011). Combining peer discussion with instructor explanation increases student learning from in-class concept questions. CBE-Life Sciences Education , 10 , 55–63.

Sniezek, J. A., & Buckley, T. (1995). Cueing and cognitive conflict in judge–Advisor decision making. Organizational Behavior and Human Decision Processes , 62 , 159–174.

Sniezek, J. A., & Henry, R. A. (1989). Accuracy and confidence in group judgment. Organizational Behavior and Human Decision Processes , 43 , 1–28.

Son, L. K., & Metcalfe, J. (2000). Metacognitive and control strategies in study-time allocation. Journal of Experimental Psychology: Learning, Memory, and Cognition , 26 , 204–221.

Steiner, I. D. (1972). Group processes and productivity . New York: Academic Press.

Thurlings, M., Vermeulen, M., Bastiaens, T., & Stijnen, S. (2013). Understanding feedback: A learning theory perspective. Educational Research Review , 9 , 1–15.

Tindale, R. S., & Sheffey, S. (2002). Shared information, cognitive load, and group memory. Group Processes & Intergroup Relations , 5 (1), 5–18.

Trouche, E., Sander, E., & Mercier, H. (2014). Arguments, more than confidence, explain the good performance of reasoning groups. Journal of Experimental Psychology: General , 143 , 1958–1971.

Tullis, J. G. (2018). Predicting others’ knowledge: Knowledge estimation as cue-utilization. Memory & Cognition , 46 , 1360–1375.

Tullis, J. G., Fiechter, J. L., & Benjamin, A. S. (2018). The efficacy of learners’ testing choices. Journal of Experimental Psychology: Learning, Memory, and Cognition , 44 , 540–552.

Tullis, J. G., & Fraundorf, S. H. (2017). Predicting others’ memory performance: The accuracy and bases of social metacognition. Journal of Memory and Language , 95 , 124–137.

Turpen, C., & Finkelstein, N. (2007). Understanding how physics faculty use peer instruction. In L. Hsu, C. Henderson, & L. McCullough (Eds.), Physics education research conference , (pp. 204–209). College Park: American Institute of Physics.

Van Swol, L. M., & Sniezek, J. A. (2005). Factors affecting the acceptance of expert advice. British Journal of Social Psychology , 44 , 443–461.

VanLehn, K., Jones, R. M., & Chi, M. T. H. (1992). A model of the self-explanation effect. Journal of the Learning Sciences , 2 (1), 1–59.

Vedder, P. (1985). Cooperative learning: A study on processes and effects of cooperation between primary school children . Westerhaven: Rijkuniversiteit Groningen.

Versteeg, M., van Blankenstein, F. M., Putter, H., & Steendijk, P. (2019). Peer instruction improves comprehension and transfer of physiological concepts: A randomized comparison with self-explanation. Advances in Health Sciences Education , 24 , 151–165.

Vygotsky, L. S. (1981). The genesis of higher mental functioning. In J. V. Wertsch (Ed.), The concept of activity in Soviet psychology , (pp. 144–188). Armonk: Sharpe.

Webb, N. M., & Palincsar, A. S. (1996). Group processes in the classroom. In D. C. Berliner, & R. C. Calfee (Eds.), Handbook of educational psychology , (pp. 841–873). New York: Macmillan Library Reference USA: London: Prentice Hall International.

Wegner, D. M., Giuliano, T., & Hertel, P. (1985). Cognitive interdependence in close relationships. In W. J. Ickes (Ed.), Compatible and incompatible relationships , (pp. 253–276). New York: Springer-Verlag.

Chapter   Google Scholar  

Weldon, M. S., & Bellinger, K. D. (1997). Collective memory: Collaborative and individual processes in remembering. Journal of Experimental Psychology: Learning, Memory, and Cognition , 23 , 1160–1175.

Wieman, C., Perkins, K., Gilbert, S., Benay, F., Kennedy, S., Semsar, K., et al. (2009). Clicker resource guide: An instructor’s guide to the effective use of personalresponse systems (clickers) in teaching . Vancouver: University of British Columbia Available from http://www.cwsei.ubc.ca/resources/files/Clicker_guide_CWSEI_CU-SEI.pdf .

Wong, R. M. F., Lawson, M. J., & Keeves, J. (2002). The effects of self-explanation training on students’ problem solving in high school mathematics. Learning and Instruction , 12 , 23.

Yackel, E., Cobb, P., & Wood, T. (1991). Small-group interactions as a source of learning opportunities in second-grade mathematics. Journal for Research in Mathematics Education , 22 , 390–408.

Yaniv, I. (2004a). The benefit of additional opinions. Current Directions in Psychological Science , 13 , 75–78.

Yaniv, I. (2004b). Receiving other people’s advice: Influence and benefit. Organizational Behavior and Human Decision Processes , 93 , 1–13.

Yaniv, I., & Choshen-Hillel, S. (2012). Exploiting the wisdom of others to make better decisions: Suspending judgment reduces egocentrism and increases accuracy. Journal of Behavioral Decision Making , 25 , 427–434.

Yaniv, I., & Kleinberger, E. (2000). Advice taking in decision making: Egocentric discounting and reputation formation. Organizational Behavior and Human Decision Processes , 83 , 260–281.

Download references

Acknowledgements

Not applicable.

No funding supported this manuscript.

Author information

Authors and affiliations.

Department of Educational Psychology, University of Arizona, 1430 E. Second St., Tucson, AZ, 85721, USA

Jonathan G. Tullis

Department of Psychology, Indiana University, Bloomington, IN, USA

Robert L. Goldstone

You can also search for this author in PubMed   Google Scholar

Contributions

JGT collected some data, analyzed the data, and wrote the first draft of the paper. RLG collected some data, contributed significantly to the framing of the paper, and edited the paper. The authors read and approved the final manuscript.

Authors’ information

JGT: Assistant Professor in Educational Psychology at University of Arizona. RLG: Chancellor’s Professor in Psychology at Indiana University.

Corresponding author

Correspondence to Jonathan G. Tullis .

Ethics declarations

Ethics approval and consent to participate.

The ethics approval was waived by the Indiana University Institutional Review Board (IRB) and the University of Arizona IRB, given that these data are collected as part of normal educational settings and processes.

Consent for publication

No individual data are presented in the manuscript.

Competing interests

The authors declare that they have no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Tullis, J.G., Goldstone, R.L. Why does peer instruction benefit student learning?. Cogn. Research 5 , 15 (2020). https://doi.org/10.1186/s41235-020-00218-5

Download citation

Received : 08 October 2019

Accepted : 25 February 2020

Published : 09 April 2020

DOI : https://doi.org/10.1186/s41235-020-00218-5

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Group decisions
• Peer instruction
• Metacognition
• Decision making

The Definition Of Peer Teaching: A Sampling Of Existing Research

Peer teaching occurs when students, by design, teach other students.

Peer Teaching: A Definition

by TeachThought Staff

What is peer teaching?

In short, peer teaching occurs when students, by design, teach other students.

But teaching what? And how? Austin Community College provided an overview of some of the existing (though decades old) research in a collection of resources for teachers in training, which provides a nice context for peer teaching.

“There is a wealth of evidence that peer teaching is extremely effective for a wide range of goals, content, and students of different levels and personalities (McKeachie et al., 1986). Peer teaching involves one or more students teaching other students in a particular subject area and builds on the belief that “to teach is to learn twice” (Whitman, 1998).”

“Peer teaching can enhance learning by enabling learners to take responsibility for reviewing, organizing, and consolidating existing knowledge and material; understanding its basic structure; filling in the gaps; finding additional meanings; and reformulating knowledge into new conceptual frameworks’ (Dueck, 1993).”

“Help from peers increases learning both for the students being helped as well as for those giving the help. For the students being helped, the assistance from their peers enables them to move away from dependence on teachers and gain more opportunities to enhance their learning. For the students giving the help, the cooperative learning groups serve as opportunities to increase their own performance. They have the chance to experience and learn that “teaching is the best teacher” (Farivar and Webb, 1994).”

In lieu of the benefits peer teaching and learning provide, it has a mixed reputation in education to its abuse via ‘let the ‘high’ students teaching the ‘low’ students’ which, done poorly, fails to meet the needs of both.

Peer Learning

David Boud of Stanford University explored the concepts of peer teaching, learning, and reciprocal peer learning in a short overview of existing research–which is limited. Though the context he discusses is primarily in the higher-ed domain where peer teaching is a literal component of most university learning models, the concepts transfer to K-12 as well.

According to Boud, peer learning is obviously closely related,

“We define peer learning in its broadest sense, then, as ‘students learning from and with each other in both formal and informal ways’. The emphasis is on the learning process, including the emotional support that learners offer each other, as much as the learning task itself. In peer teaching the roles of teacher and learner are fixed, whereas in peer learning they are either undefined or may shift during the course of the learning experience. Staff may be actively involved as group facilitators or they may simply initiate student-directed activities such as workshops or learning partnerships.”

As for the limited research data, Boud continues,

‘According to Topping’s review of literature, surprisingly little research has been done into either dyadic reciprocal peer tutoring or same-year group tutoring (Topping, 1996). He identified only 10 studies, all with a very narrow, empirical focus. This suggests that the teaching model, rather than the learning model, is still the most common way of understanding how students assist each other. Although the teaching model has value, we must also consider the learning process itself if we want to make the best use of peers as resources for learning.”

Whitman and Fife (1989)  summarize research that was to that point current, below.

“Recommendations from current literature include the following: learning may occur when students work cooperatively, both peer teachers and peer learners learn, and learning may increase with a blend of situations in which professors are present and are not present.”

A significant portion of existing discourse on peer teaching relate to its application in the medical field, or language learning. A study published at Oxford Academic’s ELT Journal in 2017 added little new information, with the abstract concluding, “The use of peer teaching in the language classroom offers a creative way for students to participate more fully in the learning process,” and alluding to “(p)revious studies (that) have reported that peer taught lessons bring benefits such as improved motivation, enhanced learning, and authentic communication.”

TeachThought is an organization dedicated to innovation in education through the growth of outstanding teachers.

An official website of the United States government

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

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

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Cogn Res Princ Implic
• v.5; 2020 Dec

Why does peer instruction benefit student learning?

Jonathan g. tullis.

1 Department of Educational Psychology, University of Arizona, 1430 E. Second St., Tucson, AZ 85721 USA

Robert L. Goldstone

2 Department of Psychology, Indiana University, Bloomington, IN USA

Associated Data

As described below, data and materials are available on the OpenScienceFramework: https://mfr.osf.io/render?url=https://osf.io/5qc46/?action=download%26mode=render .

In peer instruction, instructors pose a challenging question to students, students answer the question individually, students work with a partner in the class to discuss their answers, and finally students answer the question again. A large body of evidence shows that peer instruction benefits student learning. To determine the mechanism for these benefits, we collected semester-long data from six classes, involving a total of 208 undergraduate students being asked a total of 86 different questions related to their course content. For each question, students chose their answer individually, reported their confidence, discussed their answers with their partner, and then indicated their possibly revised answer and confidence again. Overall, students were more accurate and confident after discussion than before. Initially correct students were more likely to keep their answers than initially incorrect students, and this tendency was partially but not completely attributable to differences in confidence. We discuss the benefits of peer instruction in terms of differences in the coherence of explanations, social learning, and the contextual factors that influence confidence and accuracy.

Significance

Peer instruction is widely used in physics instruction across many universities. Here, we examine how peer instruction, or discussing one’s answer with a peer, affects students’ decisions about a class assignment. Across six different university classes, students answered a question, discussed their answer with a peer, and finally answered the question again. Students’ accuracy consistently improved through discussion with a peer. Our peer instruction data show that students were hesitant to switch away from their initial answer and that students did consider both their own confidence and their partner’s confidence when making their final decision, in accord with basic research about confidence in decision making. More broadly, the data reveal that peer discussion helped students select the correct answer by prompting them to create new knowledge. The benefit to student accuracy that arises when students discuss their answers with a partner is a “process gain”, in which working in a group yields better performance than can be predicted from individuals’ performance alone.

Peer instruction is specific evidence-based instructional strategy that is well-known and widely used, particularly in physics (Henderson & Dancy, 2009 ). In fact, peer instruction has been advocated as a part of best methods in science classrooms (Beatty, Gerace, Leonard, & Dufresne, 2006 ; Caldwell, 2007 ; Crouch & Mazur, 2001 ; Newbury & Heiner, 2012 ; Wieman et al., 2009 ) and over a quarter of university physics professors report using peer instruction (Henderson & Dancy, 2009 ). In peer instruction, instructors pose a challenging question to students, students answer the question individually, students discuss their answers with a peer in the class, and finally students answer the question again. There are variations of peer instruction in which instructors show the class’s distribution of answers before discussion (Nielsen, Hansen-Nygård, & Stav, 2012 ; Perez et al., 2010 ), in which students’ answers are graded for participation or for correctness (James, 2006 ), and in which instructors’ norms affect whether peer instruction offers opportunities for answer-seeking or for sense-making (Turpen & Finkelstein, 2007 ).

Despite wide variations in its implementation, peer instruction consistently benefits student learning. Switching classroom structure from didactic lectures to one centered around peer instruction improves learners’ conceptual understanding (Duncan, 2005 ; Mazur, 1997 ), reduces student attrition in difficult courses (Lasry, Mazur, & Watkins, 2008 ), decreases failure rates (Porter, Bailey-Lee, & Simon, 2013 ), improves student attendance (Deslauriers, Schelew, & Wieman, 2011 ), and bolsters student engagement (Lucas, 2009 ) and attitudes to their course (Beekes, 2006 ). Benefits of peer instruction have been found across many fields, including physics (Mazur, 1997 ; Pollock, Chasteen, Dubson, & Perkins, 2010 ), biology (Knight, Wise, & Southard, 2013 ; Smith, Wood, Krauter, & Knight, 2011 ), chemistry (Brooks & Koretsky, 2011 ), physiology (Cortright, Collins, & DiCarlo, 2005 ; Rao & DiCarlo, 2000 ), calculus (Lucas, 2009 ; Miller, Santana-Vega, & Terrell, 2007 ), computer science (Porter et al., 2013 ), entomology (Jones, Antonenko, & Greenwood, 2012 ), and even philosophy (Butchart, Handfield, & Restall, 2009 ). Additionally, benefits of peer instruction have been found at prestigious private universities, two-year community colleges (Lasry et al., 2008 ), and even high schools (Cummings & Roberts, 2008 ). Peer instruction benefits not just the specific questions posed during discussion, but also improves accuracy on later similar problems (e.g., Smith et al., 2009 ).

One of the consistent empirical hallmarks of peer instruction is that students’ answers are more frequently correct following discussion than preceding it. For example, in introductory computer science courses, post-discussion performance was higher on 70 out of 71 questions throughout the semester (Simon, Kohanfars, Lee, Tamayo, & Cutts, 2010 ). Further, gains in performance from discussion are found on many different types of questions, including recall, application, and synthesis questions (Rao & DiCarlo, 2000 ). Performance improvements are found because students are more likely to switch from an incorrect answer to the correct answer than from the correct answer to an incorrect answer. In physics, 59% of incorrect answers switched to correct following discussion, but only 13% of correct answers switched to incorrect (Crouch & Mazur, 2001 ). Other research on peer instruction shows the same patterns: 41% of incorrect answers are switched to correct ones, while only 18% of correct answers are switched to incorrect (Morgan & Wakefield, 2012 ). On qualitative problem-solving questions in physiology, 57% of incorrect answers switched to correct after discussion, and only 7% of correct answers to incorrect (Giuliodori, Lujan, & DiCarlo, 2006 ).

There are two explanations for improvements in pre-discussion to post-discussion accuracy. First, switches from incorrect to correct answers may be driven by selecting the answer from the peer who is more confident. When students discuss answers that disagree, they may choose whichever answer belongs to the more confident peer. Evidence about decision-making and advice-taking substantiates this account. First, confidence is correlated with correctness across many settings and procedures (Finley, Tullis, & Benjamin, 2010 ). Students who are more confident in their answers are typically more likely to be correct. Second, research examining decision-making and advice-taking indicates that (1) the less confident you are, the more you value others’ opinions (Granovskiy, Gold, Sumpter, & Goldstone, 2015 ; Harvey & Fischer, 1997 ; Yaniv, 2004a , 2004b ; Yaniv & Choshen-Hillel, 2012 ) and (2) the more confident the advisor is, the more strongly they influence your decision (Kuhn & Sniezek, 1996 ; Price & Stone, 2004 ; Sah, Moore, & MacCoun, 2013 ; Sniezek & Buckley, 1995 ; Van Swol & Sniezek, 2005 ; Yaniv, 2004b ). Consequently, if students simply choose their final answer based upon whoever is more confident, accuracy should increase from pre-discussion to post-discussion. This explanation suggests that switches in answers should be driven entirely by a combination of one’s own initial confidence and one’s partner’s confidence. In accord with this confidence view, Koriat ( 2015 ) shows that an individual’s confidence typically reflects the group’s most typically given answer. When the answer most often given by group members is incorrect, peer interactions amplify the selection of and confidence in incorrect answers. Correct answers have no special draw. Rather, peer instruction merely amplifies the dominant view through differences in the individual’s confidence.

In a second explanation, working with others may prompt students to verbalize explanations and verbalizations may generate new knowledge. More specifically, as students discuss the questions, they need to create a common representation of the problem and answer. Generating a common representation may compel students to identify gaps in their existing knowledge and construct new knowledge (Schwartz, 1995 ). Further, peer discussion may promote students’ metacognitive processes of detecting and correcting errors in their mental models. Students create more new knowledge and better diagnostic tests of answers together than alone. Ultimately, then, the new knowledge and improved metacognition may make the correct answer appear more compelling or coherent than incorrect options. Peer discussion would draw attention to coherent or compelling answers, more so than students’ initial confidence alone and the coherence of the correct answer would prompt students to switch away from incorrect answers. Similarly, Trouche, Sander, and Mercier ( 2014 ) argue that interactions in a group prompt argumentation and discussion of reasoning. Good arguments and reasoning should be more compelling to change individuals’ answers than confidence alone. Indeed, in a reasoning task known to benefit from careful deliberation, good arguments and the correctness of the answers change partners’ minds more than confidence in one’s answer (Trouche et al., 2014 ). This explanation predicts several distinct patterns of data. First, as seen in prior research, more students should switch from incorrect answers to correct than vice versa. Second, the intrinsic coherence of the correct answer should attract students, so the likelihood of switching answers would be predicted by the correctness of an answer above and beyond differences in initial confidence. Third, initial confidence in an answer should not be as tightly related to initial accuracy as final confidence is to final accuracy because peer discussion should provide a strong test of the coherence of students’ answers. Fourth, because the coherence of an answer is revealed through peer discussion, student confidence should increase more from pre-discussion to post-discussion when they agree on the correct answers compared to agreeing on incorrect answers.

Here, we examined the predictions of these two explanations of peer instruction across six different classes. We specifically examined whether changes in answers are driven exclusively through the confidence of the peers during discussion or whether the coherence of an answer is better constructed and revealed through peer instruction than on one’s own. We are interested in analyzing cognitive processes at work in a specific, but common, implementation of classroom-based peer instruction; we do not intend to make general claims about all kinds of peer instruction or to evaluate the long-term effectiveness of peer instruction. This research is the first to analyze how confidence in one’s answer relates to answer-switching during peer instruction and tests the impact of peer instruction in new domains (i.e., psychology and educational psychology classes).

Participants

Students in six different classes participated as part of their normal class procedures. More details about these classes are presented in Table  1 . The authors served as instructors for these classes. Across the six classes, 208 students contributed a total of 1657 full responses to 86 different questions.

Descriptions of classes used

The instructors of the courses developed multiple-choice questions related to the ongoing course content. Questions were aimed at testing students’ conceptual understanding, rather than factual knowledge. Consequently, questions often tested whether students could apply ideas to new settings or contexts. An example of a cognitive psychology question used is: Which is a fixed action pattern (not a reflex)?

• Knee jerks up when patella is hit
• Male bowerbirds building elaborate nests [correct]
• Eye blinks when air is blown on it
• Can play well learned song on guitar even when in conversation

The procedures for peer instruction across the six different classes followed similar patterns. Students were presented with a multiple-choice question. First, students read the question on their own, chose their answer, and reported their confidence in their answer on a scale of 1 “Not at all confident” to 10 “Highly confident”. Students then paired up with a neighbor in their class and discussed the question with their peer. After discussion, students answered the question and reported the confidence for a second time. The course instructor indicated the correct answer and discussed the reasoning for the answer after all final answers had been submitted. Instruction was paced based upon how quickly students read and answered questions. Most student responses counted towards their participation grade, regardless of the correctness of their answer (the last question in each of the cognitive psychology classes was graded for correctness).

There were small differences in procedures between classes. Students in the cognitive psychology classes input their responses using classroom clickers, but those in other classes wrote their responses on paper. Further, students in the cognitive psychology classes explicitly reported their partner’s answer and confidence, while students in other classes only reported the name of their partner (the partners’ data were aligned during data recording). The cognitive psychology students then were required to mention their own answer and their confidence to their partner during peer instruction; students in other classes were not required to tell their answer or their confidence to their peer. Finally, the questions appeared at any point during the class period for the cognitive psychology classes, while the questions typically happened at the beginning of each class for the other classes.

Analytic strategy

Data are available on the OpenScienceFramework: https://mfr.osf.io/render?url=https://osf.io/5qc46/?action=download%26mode=render .

For most of our analyses we used linear mixed-effects models (Baayen, Davidson, & Bates, 2008 ; Murayama, Sakaki, Yan, & Smith, 2014 ). The unit of analysis in a mixed-effect model is the outcome of a single trial (e.g., whether or not a particular question was answered correctly by a particular participant). We modeled these individual trial-level outcomes as a function of multiple fixed effects - those of theoretical interest - and multiple random effects - effects for which the observed levels are sampled out of a larger population (e.g., questions, students, and classes sampled out of a population of potential questions, students, and classes).

Linear mixed-effects models solve four statistical problems involved with the data of peer instruction. First, there is large variability in students’ performance and the difficulty of questions across students and classes. Mixed-effect models simultaneously account for random variation both across participants and across items (Baayen et al., 2008 ; Murayama et al., 2014 ). Second, students may miss individual classes and therefore may not provide data across every item. Similarly, classes varied in how many peer instruction questions were posed throughout the semester and the number of students enrolled. Mixed-effects models weight each response equally when drawing conclusions (rather than weighting each student or question equally) and can easily accommodate missing data. Third, we were interested in how several different characteristics influenced students’ performance. Mixed effects models can include multiple predictors simultaneously, which allows us to test the effect of one predictor while controlling for others. Finally, mixed effects models can predict the log odds (or logit) of a correct answer, which is needed when examining binary outcomes (i.e., correct or incorrect; Jaeger, 2008 ).

We fit all models in R using the lmer() function of the lme4 package (Bates, Maechler, Bolker, & Walker, 2015 ). For each mixed-effect model, we included random intercepts that capture baseline differences in difficulty of questions, in classes, and in students, in addition to multiple fixed effects of theoretical interest. In mixed-effect models with hundreds of observations, the t distribution effectively converges to the normal, so we compared the t statistic to the normal distribution for analyses involving continuous outcomes (i.e., confidence; Baayen, 2008 ). P values can be directly obtained from Wald z statistics for models with binary outcomes (i.e., correctness).

Does accuracy change through discussion?

First, we examined how correctness changed across peer discussion. A logit model predicting correctness from time point (pre-discussion to post-discussion) revealed that the odds of correctness increased by 1.57 times (95% confidence interval (conf) 1.31–1.87) from pre-discussion to post-discussion, as shown in Table  2 . In fact, 88% of students showed an increase or no change in accuracy from pre-discussion to post-discussion. Pre-discussion to post-discussion performance for each class is shown in Table  3 . We further examined how accuracy changed from pre-discussion to post-discussion for each question and the results are plotted in Fig.  1 . The data show a consistent improvement in accuracy from pre-discussion to post-discussion across all levels of initial difficulty.

The effect of time point (pre-discussion to post-discussion) on accuracy using a mixed effect logit model

Accuracy before and after discussion by class

The relationship between pre-discussion accuracy (x axis) and post-discussion accuracy (y axis). Each point represents a single question. The solid diagonal line represents equal pre-discussion and post-discussion accuracy; points above the line indicate improvements in accuracy and points below represent decrements in accuracy. The dashed line indicates the line of best fit for the observed data

We examined how performance increased from pre-discussion to post-discussion by tracing the correctness of answers through the discussion. Figure  2 tracks the percent (and number of items) correct from pre-discussion to post-discussion. The top row shows whether students were initially correct or incorrect in their answer; the middle row shows whether students agreed or disagreed with their partner; the last row show whether students were correct or incorrect after discussion. Additionally, Fig. ​ Fig.2 2 shows the confidence associated with each pathway. The bottow line of each entry shows the students’ average confidence; in the middle white row, the confidence reported is the average of the peer’s confidence.

The pathways of answers from pre-discussion (top row) to post-discussion (bottom row). Percentages indicate the portion of items from the category immediately above in that category, the numbers in brackets indicate the raw numbers of items, and the numbers at the bottom of each entry indicate the confidence associated with those items. In the middle, white row, confidence values show the peer’s confidence. Turquoise indicates incorrect answers and yellow indicates correct answers

Broadly, only 5% of correct answers were switched to incorrect, while 28% of incorrect answers were switched to correct following discussion. Even for the items in which students were initially correct but disagreed with their partner, only 21% of answers were changed to incorrect answers after discussion. However, out of the items where students were initially incorrect and disagreed with their partner, 42% were changed to the correct answer.

Does confidence predict switching?

Differences in the amount of switching to correct or incorrect answers could be driven solely by differences in confidence, as described in our first theory mentioned earlier. For this theory to hold, answers with greater confidence must have a greater likelihood of being correct. To examine whether initial confidence is associated with initial correctness, we calculated the gamma correlation between correctness and confidence in the answer before discussion, as shown in the first column of Table  4 . The average gamma correlation between initial confidence and initial correctness (mean (M) = 0.40) was greater than zero, t (160) = 8.59, p  < 0.001, d  = 0.68, indicating that greater confidence was associated with being correct.

The gamma correlation between accuracy and confidence before and after discussion for each class

a Gamma correlation requires that learners have variance in both confidence and correctness before and after discussion. Degrees of freedom are reduced because many students did not have requisite variation

Changing from an incorrect to a correct answer, then, may be driven entirely by selecting the answer from the peer with the greater confidence during discussion, even though most of the students in our sample were not required to explicitly disclose their confidence to their partner during discussion. We examined how frequently students choose the more confident answer when peers disagree. When peers disagreed, students’ final answers aligned with the more confident peer only 58% of the time. Similarly, we tested what the performance would be if peers always picked the answer of the more confident peer. If peers always chose the more confident answer during discussion, the final accuracy would be 69%, which is significantly lower than actual final accuracy (M = 72%, t (207) = 2.59, p  = 0.01, d  = 0.18). While initial confidence is related to accuracy, these results show that confidence is not the only predictor of switching answers.

Does correctness predict switching beyond confidence?

Discussion may reveal information about the correctness of answers by generating new knowledge and testing the coherence of each possible answer. To test whether the correctness of an answer added predictive power beyond the confidence of the peers involved in discussion, we analyzed situations in which students disagreed with their partner. Out of the instances when partners initially disagreed, we predicted the likelihood of keeping one’s answer based upon one’s own confidence, the partner’s confidence, and whether one’s answer was initially correct. The results of a model predicting whether students keep their answers is shown in Table  5 . For each increase in a point of one’s own confidence, the odds of keeping one’s answer increases 1.25 times (95% conf 1.13–1.38). For each decrease in a point of the partner’s confidence, the odds of keeping one’s answer increased 1.19 times (1.08–1.32). The beta weight for one’s confidence did not differ from the beta weight of the partner’s confidence, χ 2  = 0.49, p  = 0.48. Finally, if one’s own answer was correct, the odds of keeping one’s answer increased 4.48 times (2.92–6.89). In other words, the more confident students were, the more likely they were to keep their answer; the more confident their peer was, the more likely they were to change their answer; and finally, if a student was correct, they were more likely to keep their answer.

Logit mixed-level regression analysis

The results of a logit mixed level regression predicting keeping one's answer from one's own confidence, the peer's confidence, and the correctness of one's initial answer for situations in which peers initially disagreed

To illustrate this relationship, we plotted the probability of keeping one’s own answer as a function of the difference between one’s own and their partner’s confidence for initially correct and incorrect answers. As shown in Fig.  3 , at every confidence level, being correct led to equal or more frequently keeping one’s answer than being incorrect.

The probability of keeping one’s answer in situations where one’s partner initially disagreed as a function of the difference between partners’ levels of confidence. Error bars indicate the standard error of the proportion and are not shown when the data are based upon a single data point

As another measure of whether discussion allows learners to test the coherence of the correct answer, we analyzed how discussion impacted confidence when partners’ answers agreed. We predicted confidence in answers by the interaction of time point (i.e., pre-discussion versus post-discussion) and being initially correct for situations in which peers initially agreed on their answer. The results, displayed in Table  6 , show that confidence increased from pre-discussion to post-discussion by 1.08 points and that confidence was greater for initially correct answers (than incorrect answers) by 0.78 points. As the interaction between time point and initial correctness shows, confidence increased more from pre-discussion to post-discussion when students were initially correct (as compared to initially incorrect). To illustrate this relationship, we plotted pre-confidence against post-confidence for initially correct and initially incorrect answers when peers agreed (Fig.  4 ). Each plotted point represents a student; the diagonal blue line indicates no change between pre-confidence and post-confidence. The graph reflects that confidence increases more from pre-discussion to post-discussion for correct answers than for incorrect answers, even when we only consider cases where peers agreed.

Mixed-level regression analysis of predicting confidence

The results of the mixed level regression predicting confidence in one's answer from the time point (pre- or post- discussion), the correctness of one's answer, and their interaction for situations in which peers initially agreed

The relationship between pre-discussion and post-discussion confidence as a function of the accuracy of an answer when partners agreed. Each dot represents a student

If students engage in more comprehensive answer testing during discussion than before, the relationship between confidence in their answer and the accuracy of their answer should be stronger following discussion than it is before. We examined whether confidence accurately reflected correctness before and after discussion. To do so, we calculated the gamma correlation between confidence and accuracy, as is typically reported in the literature on metacognitive monitoring (e.g., Son & Metcalfe, 2000 ; Tullis & Fraundorf, 2017 ). Across all students, the resolution of metacognitive monitoring increases from pre-discussion to post-discussion ( t (139) = 2.98, p  = 0.003, d  = 0.24; for a breakdown of gamma calculations for each class, see Table ​ Table4). 4 ). Confidence was more accurately aligned with accuracy following discussion than preceding it. The resolution between student confidence and correctness increases through discussion, suggesting that discussion offers better coherence testing than answering alone.

To examine why peer instruction benefits student learning, we analyzed student answers and confidence before and after discussion across six psychology classes. Discussing a question with a partner improved accuracy across classes and grade levels with small to medium-sized effects. Questions of all difficulty levels benefited from peer discussion; even questions where less than half of students originally answered correctly saw improvements from discussion. Benefits across the spectrum of question difficulty align with prior research showing improvements when even very few students initially know the correct answer (Smith et al., 2009 ). More students switched from incorrect answers to correct answers than vice versa, leading to an improvement in accuracy following discussion. Answer switching was driven by a student’s own confidence in their answer and their partner’s confidence. Greater confidence in one’s answer indicated a greater likelihood of keeping the answer; a partner’s greater confidence increased the likelihood of changing to their answer.

Switching answers depended on more than just confidence: even when accounting for students’ confidence levels, the correctness of the answer impacted switching behavior. Across several measures, our data showed that the correctness of an answer carried weight beyond confidence. For example, the correctness of the answer predicted whether students switched their initial answer during peer disagreements, even after taking the confidence of both partners into account. Further, students’ confidence increased more when partners agreed on the correct answer compared to when they agreed on an incorrect answer. Finally, although confidence increased from pre-discussion to post-discussion when students changed their answers from incorrect to the correct ones, confidence decreased when students changed their answer away from the correct one. A plausible interpretation of this difference is that when students switch from a correct answer to an incorrect one, their decrease in confidence reflects the poor coherence of their final incorrect selection.

Whether peer instruction resulted in optimal switching behaviors is debatable. While accuracy improved through discussion, final accuracy was worse than if students had optimally switched their answers during discussion. If students had chosen the correct answer whenever one of the partners initially chose it, the final accuracy would have been significantly higher (M = 0.80 (SD = 0.19)) than in our data (M = 0.72 (SD = 0.24), t (207) = 6.49, p  < 0.001, d  = 0.45). While this might be interpreted as “process loss” (Steiner, 1972 ; Weldon & Bellinger, 1997 ), that would assume that there is sufficient information contained within the dyad to ascertain the correct answer. One individual selecting the correct answer is inadequate for this claim because they may not have a compelling justification for their answer. When we account for differences in initial confidence, students’ final accuracy was better than expected. Students’ final accuracy was better than that predicted from a model in which students always choose the answer of the more confident peer. This over-performance, often called “process gain”, can sometimes emerge when individuals collaborate to create or generate new knowledge (Laughlin, Bonner, & Miner, 2002 ; Michaelsen, Watson, & Black, 1989 ; Sniezek & Henry, 1989 ; Tindale & Sheffey, 2002 ). Final accuracy reveals that students did not simply choose the answer of the more confident student during discussion; instead, students more thoroughly probed the coherence of answers and mental models during discussion than they could do alone.

Students’ final accuracy emerges from the interaction between the pairs of students, rather than solely from individuals’ sequestered knowledge prior to discussion (e.g. Wegner, Giuliano, & Hertel, 1985 ). Schwartz ( 1995 ) details four specific cognitive products that can emerge through working in dyads. Specifically, dyads force verbalization of ideas through discussion, and this verbalization facilitates generating new knowledge. Students may not create a coherent explanation of their answer until they engage in discussion with a peer. When students create a verbal explanation of their answer to discuss with a peer, they can identify knowledge gaps and construct new knowledge to fill those gaps. Prior research examining the content of peer interactions during argumentation in upper-level biology classes has shown that these kinds of co-construction happen frequently; over three quarters of statements during discussion involve an exchange of claims and reasoning to support those claims (Knight et al., 2013 ). Second, dyads have more information processing resources than individuals, so they can solve more complex problems. Third, dyads may foster greater motivation than individuals. Finally, dyads may stimulate the creation of new, abstract representations of knowledge, above and beyond what one would expect from the level of abstraction created by individuals. Students need to communicate with their partner; to create common ground and facilitate discourse, dyads negotiate common representations to coordinate different perspectives. The common representations bridge multiple perspectives, so they lose idiosyncratic surface features of individuals’ representation. Working in pairs generates new knowledge and tests of answers that could not be predicted from individuals’ performance alone.

More broadly, teachers often put students in groups so that they can learn from each other by giving and receiving help, recognizing contradictions between their own and others’ perspectives, and constructing new understandings from divergent ideas (Bearison, Magzamen, & Filardo, 1986 ; Bossert, 1988-1989 ; Brown & Palincsar, 1989 ; Webb & Palincsar, 1996 ). Giving explanations to a peer may encourage explainers to clarify or reorganize information, recognize and rectify gaps in understandings, and build more elaborate interpretations of knowledge than they would have alone (Bargh & Schul, 1980 ; Benware & Deci, 1984 ; King, 1992 ; Yackel, Cobb, & Wood, 1991 ). Prompting students to explain why and how problems are solved facilitates conceptual learning more than reading the problem solutions twice without self-explanations (Chi, de Leeuw, Chiu, & LaVancher, 1994 ; Rittle-Johnson, 2006 ; Wong, Lawson, & Keeves, 2002 ). Self-explanations can prompt students to retrieve, integrate, and modify their knowledge with new knowledge; self-explanations can also help students identify gaps in their knowledge (Bielaczyc, Pirolli, & Brown, 1995 ; Chi & Bassock, 1989 ; Chi, Bassock, Lewis, Reimann, & Glaser, 1989 ; Renkl, Stark, Gruber, & Mandl, 1998 ; VanLehn, Jones, & Chi, 1992 ; Wong et al., 2002 ), detect and correct errors, and facilitate deeper understanding of conceptual knowledge (Aleven & Koedinger, 2002 ; Atkinson, Renkl, & Merrill, 2003 ; Chi & VanLehn, 2010 ; Graesser, McNamara, & VanLehn, 2005 ). Peer instruction, while leveraging these benefits of self-explanation, also goes beyond them by involving what might be called “other-explanation” processes - processes recruited not just when explaining a situation to oneself but to others. Mercier and Sperber ( 2019 ) argue that much of human reason is the result of generating explanations that will be convincing to other members of one’s community, thereby compelling others to act in the way that one wants.

Conversely, students receiving explanations can fill in gaps in their own understanding, correct misconceptions, and construct new, lasting knowledge. Fellow students may be particularly effective explainers because they can better take the perspective of their peer than the teacher (Priniski & Horne, 2019 ; Ryskin, Benjamin, Tullis, & Brown-Schmidt, 2015 ; Tullis, 2018 ). Peers may be better able than expert teachers to explain concepts in familiar terms and direct peers’ attention to the relevant features of questions that they do not understand (Brown & Palincsar, 1989 ; Noddings, 1985 ; Vedder, 1985 ; Vygotsky, 1981 ).

Peer instruction may benefit from the generation of explanations, but social influences may compound those benefits. Social interactions may help students monitor and regulate their cognition better than self-explanations alone (e.g., Jarvela et al., 2015 ; Kirschner, Kreijns, Phielix, & Fransen, 2015 ; Kreijns, Kirschner, & Vermeulen, 2013 ; Phielix, Prins, & Kirschner, 2010 ; Phielix, Prins, Kirschner, Erkens, & Jaspers, 2011 ). Peers may be able to judge the quality of the explanation better than the explainer. In fact, recent research suggests that peer instruction facilitates learning even more than self-explanations (Versteeg, van Blankenstein, Putter, & Steendijk, 2019 ).

Not only does peer instruction generate new knowledge, but it may also improve students’ metacognition. Our data show that peer discussion prompted more thorough testing of the coherence of the answers. Specifically, students’ confidences were better aligned with accuracy following discussion than before. Improvements in metacognitive resolution indicate that discussion provides more thorough testing of answers and ideas than does answering questions on one’s own. Discussion facilitates the metacognitive processes of detecting errors and assessing the coherence of an answer.

Agreement among peers has important consequences for final behavior. For example, when peers agreed, students very rarely changed their answer (less than 3% of the time). Further, large increases in confidence occurred when students agreed (as compared to when they disagreed). Alternatively, disagreements likely engaged different discussion processes and prompted students to combine different answers. Whether students weighed their initial answer more than their partner’s initial answer remains debatable. When students disagreed with their partner, they were more likely to stick with their own answer than switch; they kept their own answer 66% of the time. Even when their partner was more confident, students only switched to their partner’s answer 50% of the time. The low rate of switching during disagreements suggests that students weighed their own answer more heavily than their partner’s answer. In fact, across prior research, deciders typically weigh their own thoughts more than the thoughts of an advisor (Harvey, Harries, & Fischer, 2000 ; Yaniv & Kleinberger, 2000 ).

Interestingly, peers agreed more frequently than expected by chance. When students were initially correct (64% of the time), 78% of peers agreed. When students were initially incorrect (36% of the time), peers agreed 43% of the time. Pairs of students, then, agree more than expected by a random distribution of answers throughout the classroom. These data suggest that students group themselves into pairs based upon likelihood of sharing the same answer. Further, these data suggest that student understanding is not randomly distributed throughout the physical space of the classroom. Across all classes, students were instructed to work with a neighbor to discuss their answer. Given that neighbors agreed more than predicted by chance, students seem to tend to sit near and pair with peers that share their same levels of understanding. Our results from peer instruction reveal that students physically locate themselves near students of similar abilities. Peer instruction could potentially benefit from randomly pairing students together (i.e. not with a physically close neighbor) to generate the most disagreements and generative activity during discussion.

Learning through peer instruction may involve deep processing as peers actively challenge each other, and this deep processing may effectively support long-term retention. Future research can examine the persistence of gains in accuracy from peer instruction. For example, whether errors that are corrected during peer instruction stay corrected on later retests of the material remains an open question. High and low-confidence errors that are corrected during peer instruction may result in different long-term retention of the correct answer; more specifically, the hypercorrection effect suggests that errors committed with high confidence are more likely to be corrected on subsequent tests than errors with low confidence (e.g., Butler, Fazio, & Marsh, 2011 ; Butterfield & Metcalfe, 2001 ; Metcalfe, 2017 ). Whether hypercorrection holds for corrections from classmates during peer instruction (rather than from an absolute authority) could be examined in the future.

The influence of partner interaction on accuracy may depend upon the domain and kind of question posed to learners. For simple factual or perceptual questions, partner interaction may not consistently benefit learning. More specifically, partner interaction may amplify and bolster wrong answers when factual or perceptual questions lead most students to answer incorrectly (Koriat, 2015 ). However, for more “intellective tasks,” interactions and arguments between partners can produce gains in knowledge (Trouche et al., 2014 ). For example, groups typically outperform individuals for reasoning tasks (Laughlin, 2011 ; Moshman & Geil, 1998 ), math problems (Laughlin & Ellis, 1986 ), and logic problems (Doise & Mugny, 1984; Perret-Clermont, 1980 ). Peer instruction questions that allow for student argumentation and reasoning, therefore, may have the best benefits in student learning.

The underlying benefits of peer instruction extend beyond the improvements in accuracy seen from pre-discussion to post-discussion. Peer instruction prompts students to retrieve information from long-term memory, and these practice tests improve long-term retention of information (Roediger III & Karpicke, 2006 ; Tullis, Fiechter, & Benjamin, 2018 ). Further, feedback provided by instructors following peer instruction may guide students to improve their performance and correct misconceptions, which should benefit student learning (Bangert-Drowns, Kulik, & Kulik, 1991 ; Thurlings, Vermeulen, Bastiaens, & Stijnen, 2013 ). Learners who engage in peer discussion can use their new knowledge to solve new, but similar problems on their own (Smith et al., 2009 ). Generating new knowledge and revealing gaps in knowledge through peer instruction, then, effectively supports students’ ability to solve novel problems. Peer instruction can be an effective tool to generate new knowledge through discussion between peers and improve student understanding and metacognition.

Acknowledgements

Not applicable.

Authors’ contributions

JGT collected some data, analyzed the data, and wrote the first draft of the paper. RLG collected some data, contributed significantly to the framing of the paper, and edited the paper. The authors read and approved the final manuscript.

Authors’ information

JGT: Assistant Professor in Educational Psychology at University of Arizona. RLG: Chancellor’s Professor in Psychology at Indiana University.

No funding supported this manuscript.

Availability of data and materials

Ethics approval and consent to participate.

The ethics approval was waived by the Indiana University Institutional Review Board (IRB) and the University of Arizona IRB, given that these data are collected as part of normal educational settings and processes.

Consent for publication

No individual data are presented in the manuscript.

Competing interests

The authors declare that they have no competing interests.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

• Aleven V, Koedinger KR. An effective metacognitive strategy: Learning by doing and explaining with a computer based cognitive tutor. Cognitive Science. 2002; 26 :147–179. doi: 10.1207/s15516709cog2602_1. [ CrossRef ] [ Google Scholar ]
• Atkinson RK, Renkl A, Merrill MM. Transitioning from studying examples to solving problems: Effects of self-explanation prompts and fading worked-out steps. Journal of Educational Psychology. 2003; 95 :774–783. doi: 10.1037/0022-0663.95.4.774. [ CrossRef ] [ Google Scholar ]
• Baayen, R. H. (2008). Analyzing linguistic data: A practical introduction to statistics . Cambridge: Cambridge University Press.
• Baayen RH, Davidson DJ, Bates DM. Mixed-effects modeling with crossed random effects for subjects and items. Journal of Memory and Language. 2008; 59 :390–412. doi: 10.1016/j.jml.2007.12.005. [ CrossRef ] [ Google Scholar ]
• Bangert-Drowns RL, Kulik JA, Kulik C-LC. Effects of frequent classroom testing. Journal of Educational Research. 1991; 85 :89–99. doi: 10.1080/00220671.1991.10702818. [ CrossRef ] [ Google Scholar ]
• Bargh JA, Schul Y. On the cognitive benefit of teaching. Journal of Educational Psychology. 1980; 72 :593–604. doi: 10.1037/0022-0663.72.5.593. [ CrossRef ] [ Google Scholar ]
• Bates D, Maechler M, Bolker B, Walker S. Fitting linear mixed-effects models using lme4. Journal of Statistical Software. 2015; 67 :1–48. doi: 10.18637/jss.v067.i01. [ CrossRef ] [ Google Scholar ]
• Bearison DJ, Magzamen S, Filardo EK. Sociocognitive conflict and cognitive growth in young children. Merrill-Palmer Quarterly. 1986; 32 (1):51–72. [ Google Scholar ]
• Beatty ID, Gerace WJ, Leonard WJ, Dufresne RJ. Designing effective questions for classroom response system teaching. American Journal of Physics. 2006; 74 (1):31e39. doi: 10.1119/1.2121753. [ CrossRef ] [ Google Scholar ]
• Beekes W. The “millionaire” method for encouraging participation. Active Learning in Higher Education. 2006; 7 :25–36. doi: 10.1177/1469787406061143. [ CrossRef ] [ Google Scholar ]
• Benware CA, Deci EL. Quality of learning with an active versus passive motivational set. American Educational Research Journal. 1984; 21 :755–765. doi: 10.3102/00028312021004755. [ CrossRef ] [ Google Scholar ]
• Bielaczyc K, Pirolli P, Brown AL. Training in self-explanation and self regulation strategies: Investigating the effects of knowledge acquisition activities on problem solving. Cognition and Instruction. 1995; 13 :221–251. doi: 10.1207/s1532690xci1302_3. [ CrossRef ] [ Google Scholar ]
• Bossert ST. Cooperative activities in the classroom. Review of Research in Education. 1988; 15 :225–252. [ Google Scholar ]
• Brooks BJ, Koretsky MD. The influence of group discussion on students’ responses and confidence during peer instruction. Journal of Chemistry Education. 2011; 88 :1477–1484. doi: 10.1021/ed101066x. [ CrossRef ] [ Google Scholar ]
• Brown AL, Palincsar AS. Guided, cooperative learning and individual knowledge acquisition. In: Resnick LB, editor. Knowing, learning, and instruction: essays in honor of Robert Glaser. Hillsdale: Erlbaum; 1989. pp. 393–451. [ Google Scholar ]
• Butchart S, Handfield T, Restall G. Using peer instruction to teach philosophy, logic and critical thinking. Teaching Philosophy. 2009; 32 :1–40. doi: 10.5840/teachphil20093212. [ CrossRef ] [ Google Scholar ]
• Butler AC, Fazio LK, Marsh EJ. The hypercorrection effect persists over a week, but high-confidence errors return. Psychonomic Bulletin & Review. 2011; 18 (6):1238–1244. doi: 10.3758/s13423-011-0173-y. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Butterfield B, Metcalfe J. Errors committed with high confidence are hypercorrected. Journal of Experimental Psychology: Learning, Memory, and Cognition. 2001; 27 (6):1491. [ PubMed ] [ Google Scholar ]
• Caldwell JE. Clickers in the large classroom: current research and best-practice tips. CBE-Life Sciences Education. 2007; 6 (1):9–20. doi: 10.1187/cbe.06-12-0205. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Chi M, VanLehn KA. Meta-cognitive strategy instruction in intelligent tutoring systems: How, when and why. Journal of Educational Technology and Society. 2010; 13 :25–39. [ Google Scholar ]
• Chi MTH, Bassock M. Learning from examples via self-explanations. In: Resnick LB, editor. Knowing, learning, and instruction: Essays in honor of Robert Glaser. Hillsdale: Erlbaum; 1989. pp. 251–282. [ Google Scholar ]
• Chi MTH, Bassock M, Lewis M, Reimann P, Glaser R. Self-explanations: How students study and use examples in learning to solve problems. Cognitive Science. 1989; 13 :145–182. doi: 10.1207/s15516709cog1302_1. [ CrossRef ] [ Google Scholar ]
• Chi MTH, de Leeuw N, Chiu MH, LaVancher C. Eliciting self-explanations improves understanding. Cognitive Science. 1994; 18 :439–477. [ Google Scholar ]
• Cortright RN, Collins HL, DiCarlo SE. Peer instruction enhanced meaningful learning: Ability to solve novel problems. Advances in Physiology Education. 2005; 29 :107–111. doi: 10.1152/advan.00060.2004. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Crouch CH, Mazur E. Peer instruction: Ten years of experience and results. American Journal of Physics. 2001; 69 :970–977. doi: 10.1119/1.1374249. [ CrossRef ] [ Google Scholar ]
• Cummings K, Roberts S. A study of peer instruction methods with school physics students. In: Henderson C, Sabella M, Hsu L, editors. Physics education research conference. College Park: American Institute of Physics; 2008. pp. 103–106. [ Google Scholar ]
• Deslauriers L, Schelew E, Wieman C. Improved learning in a large-enrollment physics class. Science. 2011; 332 :862–864. doi: 10.1126/science.1201783. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Duncan D. Clickers in the classroom: How to enhance science teaching using classroom response systems. San Francisco: Pearson/Addison-Wesley; 2005. [ Google Scholar ]
• Finley JR, Tullis JG, Benjamin AS. Metacognitive control of learning and remembering. In: Khine MS, Saleh IM, editors. New science of learning: Cognition, computers and collaborators in education. New York: Springer Science & Business Media, LLC; 2010. [ Google Scholar ]
• Giuliodori MJ, Lujan HL, DiCarlo SE. Peer instruction enhanced student performance on qualitative problem solving questions. Advances in Physiology Education. 2006; 30 :168–173. doi: 10.1152/advan.00013.2006. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Graesser AC, McNamara D, VanLehn K. Scaffolding deep comprehension strategies through AutoTutor and iSTART. Educational Psychologist. 2005; 40 :225–234. doi: 10.1207/s15326985ep4004_4. [ CrossRef ] [ Google Scholar ]
• Granovskiy B, Gold JM, Sumpter D, Goldstone RL. Integration of social information by human groups. Topics in Cognitive Science. 2015; 7 :469–493. doi: 10.1111/tops.12150. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Harvey N, Fischer I. Taking advice: Accepting help, improving judgment, and sharing responsibility. Organizational Behavior and Human Decision Processes. 1997; 70 :117–133. doi: 10.1006/obhd.1997.2697. [ CrossRef ] [ Google Scholar ]
• Harvey N, Harries C, Fischer I. Using advice and assessing its quality. Organizational Behavior and Human Decision Processes. 2000; 81 :252–273. doi: 10.1006/obhd.1999.2874. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Henderson C, Dancy MH. The impact of physics education research on the teaching of introductory quantitative physics in the United States. Physical Review Special Topics: Physics Education Research. 2009; 5 (2):020107. doi: 10.1103/PhysRevSTPER.5.020107. [ CrossRef ] [ Google Scholar ]
• Jaeger TF. Categorical data analysis: away from ANOVAs (transformation or not) and towards logit mixed models. Journal of Memory and Language. 2008; 59 :434–446. doi: 10.1016/j.jml.2007.11.007. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• James MC. The effect of grading incentive on student discourse in peer instruction. American Journal of Physics. 2006; 74 (8):689–691. doi: 10.1119/1.2198887. [ CrossRef ] [ Google Scholar ]
• Jarvela S, Kirschner P, Panadero E, Malmberg J, Phielix C, Jaspers J, Koivuniemi M, Jarvenoja H. Enhancing socially shared regulation in collaborative learning groups: Designing for CSCL regulation tools. Educational Technology Research and Development. 2015; 63 (1):125e142. doi: 10.1007/s11423-014-9358-1. [ CrossRef ] [ Google Scholar ]
• Jones ME, Antonenko PD, Greenwood CM. The impact of collaborative and individualized student response system strategies on learner motivation, metacognition, and knowledge transfer. Journal of Computer Assisted Learning. 2012; 28 (5):477–487. doi: 10.1111/j.1365-2729.2011.00470.x. [ CrossRef ] [ Google Scholar ]
• King A. Facilitating elaborative learning through guided student-generated questioning. Educational Psychologist. 1992; 27 :111–126. doi: 10.1207/s15326985ep2701_8. [ CrossRef ] [ Google Scholar ]
• Kirschner PA, Kreijns K, Phielix C, Fransen J. Awareness of cognitive and social behavior in a CSCL environment. Journal of Computer Assisted Learning. 2015; 31 (1):59–77. doi: 10.1111/jcal.12084. [ CrossRef ] [ Google Scholar ]
• Knight JK, Wise SB, Southard KM. Understanding clicker discussions: student reasoning and the impact of instructional cues. CBE-Life Sciences Education. 2013; 12 :645–654. doi: 10.1187/cbe.13-05-0090. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Koriat A. When two heads are better than one and when they can be worse: The amplification hypothesis. Journal of Experimental Psychology: General. 2015; 144 :934–950. doi: 10.1037/xge0000092. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Kreijns K, Kirschner PA, Vermeulen M. Social aspects of CSCL environments: A research framework. Educational Psychologist. 2013; 48 (4):229e242. doi: 10.1080/00461520.2012.750225. [ CrossRef ] [ Google Scholar ]
• Kuhn LM, Sniezek JA. Confidence and uncertainty in judgmental forecasting: Differential effects of scenario presentation. Journal of Behavioral Decision Making. 1996; 9 :231–247. doi: 10.1002/(SICI)1099-0771(199612)9:4<231::AID-BDM240>3.0.CO;2-L. [ CrossRef ] [ Google Scholar ]
• Lasry N, Mazur E, Watkins J. Peer instruction: From Harvard to the two-year college. American Journal of Physics. 2008; 76 (11):1066–1069. doi: 10.1119/1.2978182. [ CrossRef ] [ Google Scholar ]
• Laughlin PR. Group problem solving. Princeton: Princeton University Press; 2011. [ Google Scholar ]
• Laughlin PR, Bonner BL, Miner AG. Groups perform better than individuals on letters-to-numbers problems. Organisational Behaviour and Human Decision Processes. 2002; 88 :605–620. doi: 10.1016/S0749-5978(02)00003-1. [ CrossRef ] [ Google Scholar ]
• Laughlin PR, Ellis AL. Demonstrability and social combination processes on mathematical intellective tasks. Journal of Experimental Social Psychology. 1986; 22 :177–189. doi: 10.1016/0022-1031(86)90022-3. [ CrossRef ] [ Google Scholar ]
• Lucas A. Using peer instruction and i-clickers to enhance student participation in calculus. Primus. 2009; 19 (3):219–231. doi: 10.1080/10511970701643970. [ CrossRef ] [ Google Scholar ]
• Mazur E. Peer instruction: A user’s manual. Upper Saddle River: Prentice Hall; 1997. [ Google Scholar ]
• Mercier H, Sperber D. The enigma of reason. Cambridge: Harvard University Press; 2019. [ Google Scholar ]
• Metcalfe J. Learning from errors. Annual Review of Psychology. 2017; 68 :465–489. doi: 10.1146/annurev-psych-010416-044022. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Michaelsen LK, Watson WE, Black RH. Realistic test of individual versus group decision making. Journal of Applied Psychology. 1989; 64 :834–839. doi: 10.1037/0021-9010.74.5.834. [ CrossRef ] [ Google Scholar ]
• Miller RL, Santana-Vega E, Terrell MS. Can good questions and peer discussion improve calculus instruction? Primus. 2007; 16 (3):193–203. doi: 10.1080/10511970608984146. [ CrossRef ] [ Google Scholar ]
• Morgan JT, Wakefield C. Who benefits from peer conversation? Examining correlations of clicker question correctness and course performance. Journal of College Science Teaching. 2012; 41 (5):51–56. [ Google Scholar ]
• Moshman D, Geil M. Collaborative reasoning: Evidence for collective rationality. Thinking and Reasoning. 1998; 4 :231–248. doi: 10.1080/135467898394148. [ CrossRef ] [ Google Scholar ]
• Murayama K, Sakaki M, Yan VX, Smith GM. Type I error inflation in the traditional by-participant analysis to metamemory accuracy: A generalized mixed-effects model perspective. Journal of Experimental Psychology: Learning, Memory, and Cognition. 2014; 40 :1287–1306. [ PubMed ] [ Google Scholar ]
• Newbury P, Heiner C. Ready, set, react! getting the most out of peer instruction using clickers. 2012. [ Google Scholar ]
• Nielsen KL, Hansen-Nygård G, Stav JB. Investigating peer instruction: how the initial voting session affects students’ experiences of group discussion. ISRN Education. 2012; 2012 :article 290157. doi: 10.5402/2012/290157. [ CrossRef ] [ Google Scholar ]
• Noddings N. Small groups as a setting for research on mathematical problem solving. In: Silver EA, editor. Teaching and learning mathematical problem solving. Hillsdale: Erlbaum; 1985. pp. 345–360. [ Google Scholar ]
• Perret-Clermont AN. Social Interaction and Cognitive Development in Children. London: Academic Press; 1980. [ Google Scholar ]
• Perez KE, Strauss EA, Downey N, Galbraith A, Jeanne R, Cooper S, Madison W. Does displaying the class results affect student discussion during peer instruction? CBE Life Sciences Education. 2010; 9 :133–140. doi: 10.1187/cbe.09-11-0080. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Phielix C, Prins FJ, Kirschner PA. Awareness of group performance in a CSCL-environment: Effects of peer feedback and reflection. Computers in Human Behavior. 2010; 26 (2):151–161. doi: 10.1016/j.chb.2009.10.011. [ CrossRef ] [ Google Scholar ]
• Phielix C, Prins FJ, Kirschner PA, Erkens G, Jaspers J. Group awareness of social and cognitive performance in a CSCL environment: Effects of a peer feedback and reflection tool. Computers in Human Behavior. 2011; 27 (3):1087–1102. doi: 10.1016/j.chb.2010.06.024. [ CrossRef ] [ Google Scholar ]
• Pollock SJ, Chasteen SV, Dubson M, Perkins KK. The use of concept tests and peer instruction in upper-division physics. In: Sabella M, Singh C, Rebello S, editors. AIP conference proceedings. New York: AIP Press; 2010. p. 261. [ Google Scholar ]
• Porter L, Bailey-Lee C, Simon B. SIGCSE ‘13: Proceedings of the 44th ACM technical symposium on computer science education. New York: ACM Press; 2013. Halving fail rates using peer instruction: A study of four computer science courses; pp. 177–182. [ Google Scholar ]
• Price PC, Stone ER. Intuitive evaluation of likelihood judgment producers. Journal of Behavioral Decision Making. 2004; 17 :39–57. doi: 10.1002/bdm.460. [ CrossRef ] [ Google Scholar ]
• Priniski JH, Horne Z. Crowdsourcing effective educational interventions. In: Goel AK, Seifert C, Freska C, editors. Proceedings of the 41st annual conference of the cognitive science society. Austin: Cognitive Science Society; 2019. [ Google Scholar ]
• Rao SP, DiCarlo SE. Peer instruction improves performance on quizzes. Advances in Physiological Education. 2000; 24 :51–55. doi: 10.1152/advances.2000.24.1.51. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Renkl A, Stark R, Gruber H, Mandl H. Learning from worked-out examples: The effects of example variability and elicited self-explanations. Contemporary Educational Psychology. 1998; 23 :90–108. doi: 10.1006/ceps.1997.0959. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Rittle-Johnson B. Promoting transfer: Effects of self-explanation and direct instruction. Child Development. 2006; 77 :1–15. doi: 10.1111/j.1467-8624.2006.00852.x. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Roediger HL, III, Karpicke JD. Test-enhanced learning: Taking memory tests improves long-term retention. Psychological Science. 2006; 17 :249–255. doi: 10.1111/j.1467-9280.2006.01693.x. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Ryskin R, Benjamin AS, Tullis JG, Brown-Schmidt S. Perspective-taking in comprehension, production, and memory: An individual differences approach. Journal of Experimental Psychology: General. 2015; 144 :898–915. doi: 10.1037/xge0000093. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Sah S, Moore DA, MacCoun RJ. Cheap talk and credibility: The consequences of confidence and accuracy on advisor credibility and persuasiveness. Organizational Behavior and Human Decision Processes. 2013; 121 :246–255. doi: 10.1016/j.obhdp.2013.02.001. [ CrossRef ] [ Google Scholar ]
• Schwartz DL. The emergence of abstract representations in dyad problem solving. The Journal of the Learning Sciences. 1995; 4 :321–354. doi: 10.1207/s15327809jls0403_3. [ CrossRef ] [ Google Scholar ]
• Simon B, Kohanfars M, Lee J, Tamayo K, Cutts Q. Proceedings of the 41st SIGCSE technical symposium on computer science education. 2010. Experience report: peer instruction in introductory computing. [ Google Scholar ]
• Smith MK, Wood WB, Adams WK, Wieman C, Knight JK, Guild N, Su TT. Why peer discussion improves student performance on in-class concept questions. Science. 2009; 323 :122–124. doi: 10.1126/science.1165919. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Smith MK, Wood WB, Krauter K, Knight JK. Combining peer discussion with instructor explanation increases student learning from in-class concept questions. CBE-Life Sciences Education. 2011; 10 :55–63. doi: 10.1187/cbe.10-08-0101. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Sniezek JA, Buckley T. Cueing and cognitive conflict in judge–Advisor decision making. Organizational Behavior and Human Decision Processes. 1995; 62 :159–174. doi: 10.1006/obhd.1995.1040. [ CrossRef ] [ Google Scholar ]
• Sniezek JA, Henry RA. Accuracy and confidence in group judgment. Organizational Behavior and Human Decision Processes. 1989; 43 :1–28. doi: 10.1016/0749-5978(89)90055-1. [ CrossRef ] [ Google Scholar ]
• Son LK, Metcalfe J. Metacognitive and control strategies in study-time allocation. Journal of Experimental Psychology: Learning, Memory, and Cognition. 2000; 26 :204–221. [ PubMed ] [ Google Scholar ]
• Steiner ID. Group processes and productivity. New York: Academic Press; 1972. [ Google Scholar ]
• Thurlings M, Vermeulen M, Bastiaens T, Stijnen S. Understanding feedback: A learning theory perspective. Educational Research Review. 2013; 9 :1–15. doi: 10.1016/j.edurev.2012.11.004. [ CrossRef ] [ Google Scholar ]
• Tindale RS, Sheffey S. Shared information, cognitive load, and group memory. Group Processes & Intergroup Relations. 2002; 5 (1):5–18. doi: 10.1177/1368430202005001535. [ CrossRef ] [ Google Scholar ]
• Trouche E, Sander E, Mercier H. Arguments, more than confidence, explain the good performance of reasoning groups. Journal of Experimental Psychology: General. 2014; 143 :1958–1971. doi: 10.1037/a0037099. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Tullis JG. Predicting others’ knowledge: Knowledge estimation as cue-utilization. Memory & Cognition. 2018; 46 :1360–1375. doi: 10.3758/s13421-018-0842-4. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Tullis JG, Fiechter JL, Benjamin AS. The efficacy of learners’ testing choices. Journal of Experimental Psychology: Learning, Memory, and Cognition. 2018; 44 :540–552. [ PubMed ] [ Google Scholar ]
• Tullis JG, Fraundorf SH. Predicting others’ memory performance: The accuracy and bases of social metacognition. Journal of Memory and Language. 2017; 95 :124–137. doi: 10.1016/j.jml.2017.03.003. [ CrossRef ] [ Google Scholar ]
• Turpen, C., & Finkelstein, N. (2007). Understanding how physics faculty use peer instruction. In L. Hsu, C. Henderson, & L. McCullough (Eds.), Physics education research conference , (pp. 204–209). College Park: American Institute of Physics.
• Van Swol LM, Sniezek JA. Factors affecting the acceptance of expert advice. British Journal of Social Psychology. 2005; 44 :443–461. doi: 10.1348/014466604X17092. [ PubMed ] [ CrossRef ] [ Google Scholar ]
• VanLehn K, Jones RM, Chi MTH. A model of the self-explanation effect. Journal of the Learning Sciences. 1992; 2 (1):1–59. doi: 10.1207/s15327809jls0201_1. [ CrossRef ] [ Google Scholar ]
• Vedder P. Cooperative learning: A study on processes and effects of cooperation between primary school children. Westerhaven: Rijkuniversiteit Groningen; 1985. [ Google Scholar ]
• Versteeg M, van Blankenstein FM, Putter H, Steendijk P. Peer instruction improves comprehension and transfer of physiological concepts: A randomized comparison with self-explanation. Advances in Health Sciences Education. 2019; 24 :151–165. doi: 10.1007/s10459-018-9858-6. [ PMC free article ] [ PubMed ] [ CrossRef ] [ Google Scholar ]
• Vygotsky LS. The genesis of higher mental functioning. In: Wertsch JV, editor. The concept of activity in Soviet psychology. Armonk: Sharpe; 1981. pp. 144–188. [ Google Scholar ]
• Webb NM, Palincsar AS. Group processes in the classroom. In: Berliner DC, Calfee RC, editors. Handbook of educational psychology. New York: Macmillan Library Reference USA: London: Prentice Hall International; 1996. pp. 841–873. [ Google Scholar ]
• Wegner DM, Giuliano T, Hertel P. Cognitive interdependence in close relationships. In: Ickes WJ, editor. Compatible and incompatible relationships. New York: Springer-Verlag; 1985. pp. 253–276. [ Google Scholar ]
• Weldon MS, Bellinger KD. Collective memory: Collaborative and individual processes in remembering. Journal of Experimental Psychology: Learning, Memory, and Cognition. 1997; 23 :1160–1175. [ PubMed ] [ Google Scholar ]
• Wieman C, Perkins K, Gilbert S, Benay F, Kennedy S, Semsar K, et al. Clicker resource guide: An instructor’s guide to the effective use of personalresponse systems (clickers) in teaching. Vancouver: University of British Columbia; 2009. [ Google Scholar ]
• Wong RMF, Lawson MJ, Keeves J. The effects of self-explanation training on students’ problem solving in high school mathematics. Learning and Instruction. 2002; 12 :23. doi: 10.1016/S0959-4752(01)00027-5. [ CrossRef ] [ Google Scholar ]
• Yackel E, Cobb P, Wood T. Small-group interactions as a source of learning opportunities in second-grade mathematics. Journal for Research in Mathematics Education. 1991; 22 :390–408. doi: 10.2307/749187. [ CrossRef ] [ Google Scholar ]
• Yaniv I. The benefit of additional opinions. Current Directions in Psychological Science. 2004; 13 :75–78. doi: 10.1111/j.0963-7214.2004.00278.x. [ CrossRef ] [ Google Scholar ]
• Yaniv I. Receiving other people’s advice: Influence and benefit. Organizational Behavior and Human Decision Processes. 2004; 93 :1–13. doi: 10.1016/j.obhdp.2003.08.002. [ CrossRef ] [ Google Scholar ]
• Yaniv I, Choshen-Hillel S. Exploiting the wisdom of others to make better decisions: Suspending judgment reduces egocentrism and increases accuracy. Journal of Behavioral Decision Making. 2012; 25 :427–434. doi: 10.1002/bdm.740. [ CrossRef ] [ Google Scholar ]
• Yaniv I, Kleinberger E. Advice taking in decision making: Egocentric discounting and reputation formation. Organizational Behavior and Human Decision Processes. 2000; 83 :260–281. doi: 10.1006/obhd.2000.2909. [ PubMed ] [ CrossRef ] [ Google Scholar ]

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

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts
• Open access
• Published: 20 June 2024

Students’ perception of peer teaching in engineering education: a mixed–method case study

• Constantin Cătălin Dosoftei 1 &
• Lidia Alexa 1  

Humanities and Social Sciences Communications volume  11 , Article number:  793 ( 2024 ) Cite this article

578 Accesses

Metrics details

• Science, technology and society

Background : Engineering education is constantly evolving and adapting to meet the demand for diverse skills and competencies in graduates, in response to the changing global economy and technological advancements. This requires shifting from a traditional content-oriented and professor-focused approach towards a more interactive, student-centered approach in which students actively engage in all process stages. The study’s main objective was to examine the students’ perceptions of peer teaching and better understand the method’s perceived advantages and disadvantages. The research was conducted over two academic years (2021 and 2022) and involved 96 students. The research incorporated quantitative and qualitative data collected through online questionnaires completed by the students at the end of the semester. The results showed a cumulative positive response rate for all close-ended questions of over 60%. The correlation analysis revealed medium positive relationships among the variables, including self-confidence, academic performance, communication and active listening, teamwork, knowledge consolidation, student-teacher benefits, and teaching activity. The thematic analysis of the open-ended questions showed that 87% of the respondents perceived the peer-teaching experience as positive and valuable. The main advantages listed by students were better communication, practicality, increased attention and interaction, and overcoming student-teacher anxiety. The main disadvantage was the perceived lack of structure and experience in coordinating laboratory work. The study results indicate that peer-based instructional methods can lead to more effective dissemination of knowledge among students, as evidenced by the high percentage of respondents who reported improved comprehension through peer-to-peer explanations. At the same time, the efficacy of this approach is contingent upon the instructor’s preparation and support, which facilitates the learning process and enhances the classroom’s social dynamics.

Similar content being viewed by others

Research on the influencing factors of promoting flipped classroom teaching based on the integrated UTAUT model and learning engagement theory

A meta-analysis to gauge the impact of pedagogies employed in mixed-ability high school biology classrooms

Nudge or not, university teachers have mixed feelings about online teaching

Introduction.

The engineering educational landscape has undergone significant shifts in recent years, with a growing emphasis on developing a broad range of skills and competencies in engineering students beyond technical expertise. The demand for engineers with diverse skills and competencies has risen in response to the increasing complexity of the global economy and technological advancement (Jamieson and Lohman, 2012 ).

This poses multiple challenges for more traditional and content-focused engineering education institutions, which predominately use lectures and demonstrations, teaching methods that no longer meet students’ 21st-century competencies and academic needs (Orji and Ogbuanya, 2018 ). To meet this demand, educational institutions have had to adapt curricula and teaching methods to better prepare students for success in the modern workforce.

Changes have been made in the instructional process in engineering schools worldwide, recognising the need for a more holistic approach to preparing engineers for the challenges and opportunities of the modern world. This has involved a shift from the content-oriented and instructor-focused approach (Lindblom-Ylänne et al. 2006 ) towards a more hands-on, active learning approach, such as cooperative learning and peer teaching, which effectively develops critical thinking, problem-solving, and communication skills in engineers, essential for success in the 21st century (Lima et al. 2017 ; Hartikainen et al. 2019 ; Tomkin et al. 2019 ; Tullis and Goldstone, 2020 ).

In this study, we investigate using a specific active learning technique, namely student peer teaching, in the context of an elective laboratory class on Hydropneumatics Drives offered within the bachelor’s degree program in Automation and Applied Informatics at the Faculty of Automatic Control and Computer Engineering. The course aims to give students a comprehensive understanding of pneumatic drives and their advantages over mechanical, hydraulic, or electrical equipment. The study was conducted over two consecutive academic years, 2021 and 2022, and focuses specifically on the laboratory component of the course.

The current paper is structured into five sections, each providing a comprehensive overview of the study’s objectives and methods. The first section examines the literature on peer learning and peer teaching in higher education. The second section presents the research setting and the specific peer teaching process and activities utilized in the Hydropneumatics Drives laboratory. The third section describes the methodology employed throughout the study, including the techniques and methods used to collect and analyse data. The fourth section presents the study’s findings, including a detailed discussion of the outcomes. Finally, the fifth section concludes the paper by highlighting the study’s limitations, providing recommendations for future research, and discussing the implications of the findings for Higher Education Institution (HEI) professors.

Peer learning and peer teaching

The word “peer” comes from the Latin word “par,” meaning equal and describes someone who is a member of the same social group, profession, or age range as oneself. Learning with and from one’s peers is a natural and common human activity, and this type of learning has been proven to be very beneficial for all parties involved (Meeuwisse et al. 2010 ; Soldner et al. 2012 ; Snyder et al. 2016 ; Gong et al. 2020 ).

Peer learning can be defined as “the use of teaching and learning strategies in which students learn with and from each other without the immediate intervention of a teacher” (Boud et al. 1999 ).

Peer learning is becoming increasingly popular in various disciplines and contexts because it offers many advantages for students as it allows them to learn by explaining their ideas to others and engaging in activities where they can learn from their peers. It creates a non-competitive empowering environment (Egbochuku and Obiunu, 2006 ) and helps them to develop skills such as organizing and planning learning activities, working effectively in teams, providing and receiving feedback, and evaluating their learning (Boud, 2001 ; Bene and Bergus, 2014 ; Williams and Reddy, 2016 ).

One critical benefit of peer learning is that it allows students to take on active roles in their education rather than being passive recipients of information. This can increase motivation and engagement, as students are more likely to be invested in the material when actively participating in the learning process (Glynn et al. 2006 ; Lucas, 2009 ; Rusli et al. 2020 ). Multiple previous studies have demonstrated that these programs not only enhance students’ self-assurance and better equip them for assessments/exams but also enhance academic achievements and encourage further academic pursuits (Altintas et al. 2016 ; Rohrbeck et al. 2003 ; Elshami et al. 2020 ; Williams and Reddy, 2016 ; Porter et al. 2013 ).

At the same time, peer teaching is a mutually beneficial process for both student learners and student teachers as it allows for revising and deepening knowledge (Boud, 2001 ; Capstick, 2004 ; Ramaswamy et al. 2001 ; Tullis and Goldstone, 2020 ; Boud, 2001 ). Student-teachers can improve their communication skills by explaining complex ideas to others, which is crucial for working in groups and with colleagues. The need to explain the material to others can increase both the willingness to acquire knowledge (Daud and Ali, 2014 ) and actual learning by allowing one to understand better, clarify, and internalize the information, identify misconceptions, and gain new perspectives (Webb, et al. 2009 ; Bene and Bergus, 2014 ; Erlich and Shaughnessy, 2014 ).

A widely recognized educational tool, the Learning Pyramid, suggests that the most effective way to learn and gain skills so necessary in engineering is by practicing or actively participating - with the most significant value of 90% retention, teaching the material to someone else - with 70% retention, and discussing the material with others - with 50% retention (Al-Badrawy, 2017 ; Gabor et al. 2022 ).

These methods have been the subject of significant research in recent years, with studies showing that they can effectively enhance student engagement, motivation, and achievement in various educational contexts (Felder and Silverman, 1988 ; Prince, 2004 ; Roseth et al. 2008 ; Secomb, 2008 ).

The studies focusing on engineering education revealed that active learning methods and, significantly, peer teaching effectively improved engineering students’ conceptual understanding and problem-solving skills (Felder and Silverman, 1988 ; Smith, et al. 2009 ), overall academic performance, and attitude toward learning (Prince, 2004 ; Freeman, et al. 2014 ; Tullis and Goldstone, 2020 ; Bene and Bergus, 2014 ; Hailikari et al. 2021 ).

The research setting

In the era of rapid technological advancement, engineering graduates are expected to have a strong innovative mindset and be equipped to tackle complex challenges posed by new technologies. The quality of education students receive during their studies, including acquiring essential skills and competencies, will play a significant role in meeting these demands. Revamping the way laboratory hours are conducted, using new and effective methods, provides students with an in-depth understanding of their chosen engineering field and boosts their confidence in their abilities. Therefore, laboratories are considered essential to engineering programs and are used as part of an active learning strategy (Rodgers, et al. 2020 ).

The experimental laboratory is critical to engineering education as it enables students to apply theoretical concepts to real-life scenarios. By allowing students to observe, measure, and analyse real-world phenomena, they gain a deeper understanding of engineering principles. Hands-on learning opportunities and exposure to the dynamic engineering field through the laboratory can significantly enhance student engagement and intrinsic motivation (Snětinová and Kácovský, 2019 ).

The Hydropneumatics Drives laboratory is used to test and study pneumatic drive systems. Pneumatic drives use pressurized gas, typically air, to power and control mechanical devices. These systems are used in various applications, including manufacturing, material handling, and automation. In a hydropneumatics drives laboratory, future engineers might test and analyse the performance of pneumatic actuators, valves, and other components and study the design and operation of pneumatic drive systems.

The Hydropneumatics Drives laboratory is a “hands-on” lab where the students learn through interactive activities or exercises that allow them to gain practical experience by performing a task or series of tasks, typically using didactic or industrial equipment and/or software for pneumatic circuit designing. With access to the latest equipment and technology, students can conduct experiments and research that would otherwise not be possible, providing a more authentic and valuable learning experience. In parallel with the new setting and equipment, in 2021, reciprocal peer teaching was introduced as a learning method. This approach can be effective in helping students to better understand concepts and retain information, as it allows them to actively engage with the material and learn from a peer who may have a different perspective or approach to the subject (Deslauriers et al. 2011 ).

Additionally, reciprocal peer teaching helps students develop essential soft skills, such as communication, critical thinking, problem-solving, teamwork, and collaboration, which are becoming more important in the new economic and industrial context (Tullis and Goldstone, 2020 ).

The laboratory is conducted with a group of students (usually four groups) formed out of 12–15, divided into three teams according to the student’s preferences. It is necessary to make an appointment for each student to take on the laboratory teacher role. Each student must take on this role at least once. After a complete rotation, when each student has taken on this role, for the remaining laboratory sessions, it is up to them to select the role, no longer being a requirement.

Considering an experimental laboratory’s complexity, the student-teachers must prepare for the working lab in advance. This training consists of two parts: first, they must read the laboratory description independently. In the second part, all student-teachers meet with the professor to highlight essential things from the next lab, starting with the learning goals and students’ expectations and ending with the results of the experiments and conclusions drawn from the results of the laboratory assignment. The entire process is presented in Fig. 1 .

The workflow in the experimental laboratory for Student Peer Teaching.

Through performing experiments and collecting data before the lab, student-teachers can apply and reinforce their understanding of scientific concepts and principles, which they will present and discuss with their colleagues in time of the laboratory. In the equipment portfolio, there are transparent or cut-away versions of teaching equipment which are imperative for understanding the principles of operation of a particular piece of equipment. These allow students to visualize the concepts they are learning about and can be used to demonstrate the principles of operation in a safe and controlled environment. It also allows peer teachers in the laboratory to focus on specific parts of the equipment, making the explanation more detailed and accurate for their colleagues to understand. Another tool for learning is the animation of working for each piece of equipment available from the equipment producer or the Internet.

However, the student-teachers can still use various other online resources to enhance their explanations and make them more detailed and precise, so that their colleagues can better understand.

During the labs, when the weight centre is shifted from the professor to the student-teachers, the professor can observe the entire learning process and act as a guide and facilitator, helping student-teachers present the procedures of the laboratory and providing guidance as needed. This is the basis of the pedagogy of engagement in which the professor assumes the role of designing and facilitating the learning experiences (Smith et al. 2005 ).

The study’s main objective was to examine the computer science student’s perception of peer teaching and better understand the method’s perceived advantages and disadvantages in a Hydropneumatics Drives laboratory context.

To evaluate the effectiveness of the chosen method of instruction, the research team aimed to provide answers to two research questions:

How do students perceive the peer-teaching experience?

What are the perceived advantages and disadvantages of the method from the student’s perspective?

To achieve this objective, the research team employed a pragmatic approach, incorporating both quantitative and qualitative data, to gain deeper insight into students’ views on the peer-teaching process.

Participants

The study participants were computer science students enrolled in the Hydropneumatics Drives laboratory course during two consecutive academic years: 2021 and 2022. There were 96 students in total, 42 students in the 2021 academic year and 54 students in the 2022 academic year. All students participated in the peer teaching process as both student teachers and learners, and therefore, they all had to complete the two questionnaires.

As seen from Tables 1 and 2 , 59 students completed the student-learner questionnaire, representing a 61% response rate, while 62 students completed the student–teacher questionnaire, representing a 65% response rate.

For context, 30.5% of the respondents completed the course in the 1st Semester of 2021, while 69.5% completed the course in the 1st Semester of 2022.

The student–teacher questionnaire is presented in Table 2 .

For the second questionnaire, 27.4% of the respondents completed the course in the 1st Semester of 2021, and 72.6% of the students completed the course in the 1st Semester of 2022.

Data collection

In the data collection stage, the students completed two questionnaires, one from the student-learner perspective and one from the student-teacher perspective. The students were asked to complete the online survey at the end of the semester, and the data was collected via Google Forms.

Both questionnaires had two parts, one that included close-ended questions using a 5-point liker scale (1 = Strongly disagree; 2 = Disagree; 3 = Neither agree nor disagree; 4 = Agree; 5 = Strongly agree) and open-ended questions regarding the advantages and disadvantages of the instructional process and recommendations.

The student-learners questionnaire included:

13 close-ended questions using a 5-point Likert scale for each evaluation criteria (see Table 3 ).

One open-ended question referred to the perceived advantages and disadvantages of the peer teaching method.

The student–teachers’ questionnaire included:

10 close-ended questions using a 5-point Likert scale for each evaluation criteria (see Table 4 ).

4 open-ended questions referred to the reasons they chose/did not choose to teach more than one seminar, the difficulties they faced, and the things they would do differently if given the opportunity.

Data analysis

The analysis was based on two main categories: quantitative and qualitative.

The quantitative analysis was executed using multiple tests in SPSS, while the qualitative one used manual coding in Excel on both student-teacher and student-learner questionnaires, with the same analysis steps being considered. For the quantitative analysis, 13 questions were designed to be studied on the student-learners scale and 10 for the student-teachers scale, respectively. As a first step, to validate the questionnaire items, a Reliability Analysis was run and the Cronbach’s Alpha values indicate there is a correspondence between the questions selected and they are relevant for the survey. The Alpha values being compared with the 0.8 threshold (0.891 > 0.8 for student-learners), (0.818 > 0.8 for student-teachers). Based on the validated items, a series of Descriptive Statistics determined an average cumulative positive impact on student-teachers of 74.66% based on the interval (62.9–95.1%) and the same average cumulative positive impact had a value of 80.12% (60.4% - 100%) for the student-learners scale. The last step in the quantitative section was to apply a correlation analysis to measure the strength of the relationship between the variables. Testing the Pearson Correlation Coefficient with a significance level chosen ( p -value < 0.05), a group of positive, strong relationships (r > 0.5) were described on both scales. For the student-learner questionnaire, 6 relationships are identified, with Pearson Correlation values between (r = 0.516 – r = 0.625) and 14 relationships for the student-teacher scale, having values between (r = 0.503 – r = 0.654).

The qualitative analysis reports four main themes grouped as two factors on each scale: the student-teacher scale describes Advantages and Disadvantages, and the student-learners define Difficulties and Improvements. The first step outlines going through the open-ended questions and manually coding the responses into keywords. Following this, each keyword, based on frequency, is grouped within its relevant theme.

Results and discussions

Insights from the quantitative analysis.

As presented in Table 3 , the cumulative positive response rate for all close-ended questions was over 60%.

According to the table, the positive response regarding the laboratory room was 100%, followed by the laboratory equipment used, with a cumulative positive value of 98.3%. The results show the importance of the laboratory setting and equipment for technical labs. The correlation analysis also supports this, as medium positive relationships ( > 0.5) between the following variables were identified: a strong relationship between laboratory equipment (Q10) and laboratory room (Q12) (r = 0.625) and a medium relationship between the number of students (Q11) and laboratory room (Q12) (r = 0.622). Other relevant statistical relationships were between the following variables: a medium relationship between student-teaching methods (Q2) and training (Q8) (r = 0.546), a medium relationship between preparation and knowledge (Q9), and an explanation by a colleague of the material (Q4) (r = 0.516), a medium relationship between expectations (Q6) and training (Q8) (r = 0.593) and medium relationship between expectations (Q6) and peer-led laboratories (Q13) (r = 0.572).

According to the data in Table 4 , a positive impact was observed for all questions regarding the respondents’ opinions, with a cumulative value of the first two response options (Agree and Strongly Agree) exceeding 60%.

According to the Pearson correlation coefficient (r), medium relationships exist among the items used in the analysis. Medium relationships were identified between self-confidence (Q1) and each of the following variables: academic performance (Q2) (r = 0.580), communication and active listening (Q3) (r = 0.580), student-teacher benefits (Q6) (r = 0.654), and teaching activity (Q7) (r = 0.513). Additionally, medium relationships were identified for academic performance with the following variables: communication and active listening (Q3) (r = 0.543), knowledge consolidation (Q4) (r = 0.522), teamwork (Q5) (r = 0.524), and student-teacher benefits (Q6) (r = 0.605). For communication and active listening skills, medium relationships were identified with the following variables: knowledge consolidation (Q4) (r = 0.576), teamwork (Q5) (r = 0.594), and teaching activity (Q7) (r = 0.503). For the knowledge consolidation variable, medium relationships were identified with teamwork (Q5) (r = 0.573) and student-learner benefits (Q10) (r = 0.554). A medium relationship was also identified between student-teacher benefits (Q6) and teaching activity (Q7) (r = 0.642).

Insights from the qualitative analysis

The responses from the open-ended questions were transcribed, divided into meaningful fragments, coded manually, and analysed using a thematic analysis, which represents the process of “identifying, analysing, and reporting patterns (themes) within data” (Braun and Clarke, 2006 ). The first step of the process consisted of a review of the initial transcribed versions done by the authors. The goal was to better understand the students’ perceptions regarding the overall value of the peer-teaching process and the method’s strengths and weaknesses. The open-ended question in the student–learner questionnaire referred to the perceived advantages and disadvantages of the peer teaching method, and several themes emerged predominantly from the 57 valid answers received. Detailed information on the number of themes and sample responses from the respondents is presented in Tables 5 to 8 .

When asked whether they think the method has proved valuable, 87% answered that the peer-teaching experience was positive and valuable mainly because they felt more comfortable interacting and asking questions. Two students considered that there was no value in the instructional method, and two gave neutral answers (both yes and no). The thematic analysis of the main advantages of the peer-teaching process listed by students is presented in Table 5 .

The main disadvantages of the method perceived by student-learners are presented in Table 6 .

In terms of disadvantages perceived by students, some listed the difference in expertise between student-teachers and professors, leading to students not trusting their peer teachers and, consequently, not learning as much from peers as they do from professors. This result is in line with other studies (Boud et al. 2001 ; Lelis, 2017 ; Sim, 2003 ). Several students mentioned this is a valuable instructional method, but it should be used occasionally. The results also highlight the relevance of several contextual factors, such as individualized teaching-learning style, confidence level, or motivation, that significantly impact the learning-teaching process (Ramm et al. 2015 ; Zarifnejad et al. 2018 ).

The open-ended questions in the student-teacher questionnaire asked about the reasons they chose/did not choose to teach more than one seminar, the difficulties they faced, and the things they would do differently if given the opportunity.

Out of the 62 students who completed the questionnaire, only 34% chose to teach a second time. Over 80% of the students who chose not to teach again did this due to busy academic schedules and inability to participate in the training sessions with the professor (10), impossibility due to activity format and team organization (10), lack of perceived incentives (2), lack of enjoyment of teaching activity (2), perceived lack of talent and lack of confidence (2). An important aspect to mention is the extra work and time student-teachers must put into participating in and delivering the class.

Out of the 34% who decided to teach more than one laboratory, most listed that they learn better when explaining the subject to a colleague because they feel a certain responsibility toward their peers.

However, the main advantages perceived by most peer teachers, regardless of whether they taught more than one laboratory, revolved around two aspects: gaining an in-depth understanding of the subject and developing better communication and presentation skills, both elements confirmed by previous studies on the matter (Tullis and Goldstone, 2020 ; Daud and Ali, 2014 ; Smith, et al. 2009 ).

In terms of the difficulties encountered in the teaching process, 20 students declared that they encountered no difficulties; for the other answers, the main categories identified are listed in Table 7 .

Although the instructional method has multiple benefits, the study revealed a series of drawbacks and challenges.

The first refers to the level of expertise and the need for consistent preparation to deliver quality content. Student responses reinforce the findings of prior research that emphasize the importance for peer-teachers to thoroughly understand the subject matter in order to teach effectively (Stigmar, 2016 ; Menezes and Premnath, 2016 ). The lack of confidence and perceived authority among their peers have also been listed in previous studies as challenges of the method (Irvine et al. 2018 ), as students are sometimes unsure of the tone to use to appear knowledgeable on the subject without seeming arrogant.

When asked what they would do differently if given the opportunity to teach again, nine out of 62 students said that they wouldn’t change anything, while the rest of the 53 listed aspects are included in the categories presented in Table 8 .

After looking across all the comments and comparing the perspectives from both roles, student and teacher, some interesting results arose on the perceived value of the peer teaching instructional method. First, from the student-learner perspective, the aspect of increased engagement and better communication mentioned by students participating in the study was listed by other studies as well (Boud, 2001 ; Lelis, 2017 ; Bulte et al. 2007 ; Lucas, 2009 ; Tullis and Goldstone, 2020 ). Another relevant aspect refers to the student-teacher benefits, namely, learning better by explaining the subject to a colleague. Through teaching, they gained an in-depth understanding of the subject and developed better communication and presentation skills.

The positive impact is also highlighted by the fact that students who participated in the laboratory sessions in the previous academic year showed an increased interest in pursuing bachelor thesis projects related to the pneumatic automation field over the past year.

Conclusions

This study aimed to examine engineering students’ perceptions of the advantages and disadvantages of peer teaching after implementing the method in a specific setting, namely a Hydropneumatics Drives laboratory. Although the analysis is limited to a specific context, the results are promising and support the available literature on peer teaching methods in engineering education.

The results show that students respond positively to the social elements of the peer teaching process, as many highlighted positive aspects related to „better communication” or „increased attention and interaction.” These outcomes are confirmed by previous studies on this matter (Hammond et al. 2010 ; Tullis and Goldstone, 2020 ) and highlight the importance of feeling comfortable asking questions and receiving answers in relevant and applicable terms. Furthermore, the fact that over 70% of respondents declared that they understood the laboratory better when a classmate explained it reinforces the results of previous research highlighting the impact of peer teaching on academic performance (Tullis and Goldstone, 2020 ; Rusli et al. 2020 ). However, the fact that there is a strong positive relationship between preparation and knowledge (Q9) and explanation by a colleague of the material (Q4), means that the success of the instructional method is highly dependent on the preparation of all stages and the professor’s ability to guide and provide support for student-teachers in the preparation and delivering process. An additional benefit of this method lies in the enhancement of empathy between students and professors, as the practicality of the teaching experience offers students a different viewpoint and promotes a deeper understanding of the pedagogical process.

An effective learning process is characterized by its ability to foster student independence, enhance confidence, and elevate motivation. Our results show that peer teaching can be a valuable method for training students to develop independence, enhance their confidence, and increase their enthusiasm to learn, as these are directly related to students assuming responsibility for their own learning. Overall, the study reveals that taking on the teacher role comes with both academic benefits (gaining an in-depth understanding of the subject) and personal benefits (developing better communication and presentation skills). This can further lead to another benefit for the students and the institution: opening the possibility to follow an academic career. This is important as the industry represents a more appealing option, especially for computer science graduates, and fewer decide to continue with a Ph.D. and remain as professors.

The study also has some limitations as it was conducted in a specific setting with a restricted number of computer science students who enrolled in the Hydropneumatics Drives laboratory. Students were all assigned both roles as teachers and as learners, and the demographic data was not included in the analysis. Future studies should be conducted on other courses with larger sample sizes and random assignments. Another useful direction for future studies is investigating the long-term effects of peer teaching on students’ academic performance and retention rates. This can provide valuable information regarding the long-term sustainability of this instructional method, as more research is needed to fully understand its potential impact and optimal implementation strategies.

Data availability

All data generated or analyzed during this study are included in this published article.

Al-Badrawy A-N (2017) Role of Engineering Design in Enhancing ABET Outcomes of Engineering Programs at Taif University. INTERNATIONAL JOURNAL APPLIED SCIENCE Technology. Vol. 6. 9-15., 39 1:9–14

Google Scholar  

Altintas T, Gunes A, Sayan H (2016) A peer-assisted learning experience in computer programming language learning and developing computer programming skills. Innov. Educ. Teach. Int. 53:329–337

Article   Google Scholar  

Bene K, Bergus G (2014) When learners become teachers: a review of peer teaching in medical student education. Fam. Medicion. 46(10):783–787

Boud, D (2001). Making the move to peer learning. In DC Boud, Peer Learning in Higher Education: Learning from and with each other (pp. 1-20.). London: Kogan Page (now Routledge)

Boud D, Cohen R, Sampson J (1999) Peer learning and assessment. Assess. High. Educ. 24:413–426

Boud D, Cohen R, Sampson J (2001) Peer learning in higher education: Learning from and with each other. Kogan Page, London

Braun V, Clarke V (2006) Using a thematic analysis in psychology. Qualit. Res. Psychol. 3(2):77–101

Bulte C, Betts A, Garner K, Durning S (2007) Student teaching: views of student near-peer teachers and learners. Med Teach. 29(6):583–590

Article   PubMed   Google Scholar  

Capstick, S (2004). Benefits and shortcomings of peer assisted learning (PAL) in Higher Education: An appraisal by students,. Peer Assisted Learning Conference Proceedings . 2004: Bournemouth University

Daud, S, & Ali, S (2014). Perceptions of learners about peer assisted learning and lectures. Int. J. Sci. Res, 1449–1455

Deslauriers L, Schelew E, Wieman C (2011) Improved learning in a large-enrollment physics class. Science 332:862–864

Article   ADS   CAS   PubMed   Google Scholar  

Egbochuku, E, & Obiunu, J (2006). The effects of reciprocal peer counselling in the enhancement of self-concept among adolescents. Education Project Innovation Inc., 126 (3)

Elshami W, Abuzaid M, Abdalla M (2020) Radiography students’ perceptions of Peer assisted learning. Radiography 26:e109–e113

Article   CAS   PubMed   Google Scholar  

Erlich DR, Shaughnessy AF (2014) Student–teacher education programme (STEP) by step: Transforming medical students into competent, confident teachers. Med. Teach. 36(4):322–332

Felder RM, Silverman LK (1988) Learning and teaching styles in engineering education. Eng. Educ. 78(7):674–681

Freeman S, Eddy SL, McDonough M, Smith MK, Okoroafor N, Jordt H, Wenderoth MP (2014) Active learning increases student performance in science, engineering, and mathematics. Proc. Natl Acad. Sci. 111(23):8410–8415

Article   ADS   CAS   PubMed   PubMed Central   Google Scholar  

Gabor G, Lucache DD, Dosoftei CC (2022) Learning PLC-based Automation by Using an Educational Elevator. 2022 Int. Conf. Exposit. Electr. Power Eng. (EPE) 36(4):683–687

Glynn, L, MacFarlane, A, Kelly, M, Cantillon, & Murphy, A (2006). Helping each other to learn-a process evaluation of peer assisted learning. BMC Med. Educat. 6 (1)

Gong HJ, Park H, Hagood TC (2020) Peer learning in STEM: a qualitative study of a student-oriented active learning intervention program. Interact. Learn. Environ. 31(4):1922–1932

Hailikari, T, Virtanen, V, Vesalainen, M, & Postareff, L (2021). Student perspectives on how different elements of constructive alignment support active learning. Active Learning in Higher Education

Hammond JA, Bithell CP, Jones L, Bidgood P (2010) A first year experience of student-directed peer-assisted learning. Act. Learn. High. Educ. 11(3):201–212

Hartikainen S, Rintala H, Pylväs L, Nokelainen P (2019) The Concept of Active Learning and the Measurement of Learning Outcomes: A Review of Research in Engineering Higher Education. Educ. Sci. 9(4):276

Irvine S, Williams B, McKenna L (2018) Near-peer teaching in undergraduate nurse education: An integrative review. Nurse Educ. Today 70:60–68

Jamieson L, Lohman J (2012) Innovation with impact: Creating a culture for scholarly and systematic innovation in engineering education. American Society for Engineering Education, Washington, DC

Lelis C (2017) Participation ahead: perceptions of masters degree students on reciprocal peer learning activities. J. Learn. Des. 10(2):14–24

ADS   MathSciNet   Google Scholar  

Lima R, Andersson P, Saalman E (2017) Active learning in engineering education: A (re)introduction. Eur. J. Eng. Educ. 42:1–4

Lindblom-Ylänne S, Trigwell K, Nevgi A, Ashwin P (2006) How approaches to teaching are affected by discipline and teaching context. Stud. High. Educ. 31:285–298

Lucas A (2009) Using peer instruction and i-clickers to enhance student participation in calculus. Primus 19(3):219–231

Meeuwisse M, Severiens SE, Born MP (2010) Learning environment, interaction, sense of belonging and study success in ethnically diverse student groups. Res. High. Educ. 51(6):528–545

Menezes S, Premnath D (2016) Near-peer education: A novel teaching program. Int. J. Med. Educ. 7:160–167

Article   PubMed   PubMed Central   Google Scholar  

Orji, TC, & Ogbuanya, TC (2018). Assessing the effectiveness of problem-based and lecture-based learning environments on students’ achievements in electronic works. Int. J. Electrical Engineering Educat., 1–20

Porter, L, Bailey-Lee, C, & Simon, B (2013). Halving fail rates using peer instruction: A study of four computer science courses. SIGCSE ‘13: Proceedings of the 44th ACM technical symposium on computer science education (pp. 177–182). New York: ACM

Prince M (2004) Does Active Learning Work? A Review of the Research. J. Eng. Educ. 93(3):223–231

Ramaswamy S, Harris I, Tschirner U (2001) Student Peer Teaching: An Innovative Approach to Instruction in Science and Engineering Education. J. Sci. Educ. Technol. 10(No. 2):165–171

Ramm D, Thomson A, Jackson A (2015) Learning clinical skills in the simulation suite: the lived experiences of student nurses involved in peer teaching and peer assessment. Nurse Educ. Today 35(6):823–827

Rodgers TL, Cheema N, Vasanth S, Jamshed A, Alfutimie A, Scully PJ (2020) Developing pre-Laboratory Videos for Enhancing Student Preparedness. Eur. J. Eng. Educ. 45(2):292–304

Rohrbeck CA, Ginsburg-Block MD, Fantuzzo JW, Miller TR (2003) Peer-assisted learning interventions with elementary school students: A meta-analytic review. J. Educ. Psychol. 95:240–257

Roseth CJ, Garfield JB, Ben-Zvi D (2008) Collaboration in learning and teaching statistics. J. Stat. Educ. 16:1

Rusli, M, Degeng, NS, Setyosari, P, & Sulton, M (2020). Peer teaching: Students teaching students to increase academic performance. Teaching Theology & Religion

Secomb J (2008) A systematic review of peer teaching and learning in clinical education. J. Clin. Nurs. 17(6):703–716

Sim, L (2003). Student perceptions of peer learning in the English unit ‘Romanticism and Revolution’,. In AB O’Sullivan (Ed.), Partners in Learning: Proceedings of the 12th Annual Teaching and Learning Forum . Perth, Australia

Smith KA, Sheppard SD, Johnson DW, Johnson RT (2005) Pedagogies of Engagement: Classroom-Based Practices. J. Eng. Educ. 94(1):87–101

Smith MK, Wood WB, Adams WK, Wieman C, Knight JK, Guild N, Su T (2009) Why peer discussion improves student performance on in-class concept questions. Science 323(5910):122–124

Snětinová, M, & Kácovský, P (2019). Hands-on experiments in the interactive physics laboratory: A study of students’ intrinsic motivation. J. Phys. Conference Series, 1287(1)

Snyder JJ, Sloane JD, Dunk RD, Wiles JR (2016) Peer-led team learning helps minority students succeed. PLoS Biol. 14(3):1–7

Soldner M, Rowan-Kenyon H, Inkelas KK, Garvey J, Robbins C (2012) Supporting students’ intentions to persist in STEM disciplines: The role of living-learning programs among other social-cognitive factors. J. High. Educ. 83(3):311–336

Stigmar M 2016Peer-to-peer teaching in higher education: A critical literature review. Mentoring Tutoring Partnership in Learning 24(20):124–136

Tomkin, J, Beilstein, S, Morphew, J, & Herman, G (2019). Evidence that communities of practice are associated with active learning in large STEM lectures. Int. J. STEM Educat. 6(1)

Tullis, JG, & Goldstone, RL (2020). Why does peer instruction benefit student learning? Cognit. Res. Principles Implications, 5(1)

Webb N, Franke M, De T, Chan A, Freund D, Shein P, Melkonian D (2009) ‘Explain to your partner’: teachers’ instructional practices and students’ dialogue in small groups. Camb. J. Educ. 39(1):49–70

Williams B, Reddy P (2016) Does peer-assisted learning improve academic performance? A scoping review. Nurse Educ. Today 42:23–29

Zarifnejad, G, Mirhaghi, A, & Rajabpoor, M (2018). Does peer education increase academic achievement in first year students? A mixed-method study. J. Peer Learning, 11

Download references

Author information

Authors and affiliations.

“Gheorghe Asachi” Technical University of Iasi, Iași, Romania

Constantin Cătălin Dosoftei & Lidia Alexa

You can also search for this author in PubMed   Google Scholar

Contributions

The authors contributed equally to the article writing process, from formulating the research plan to writing and editing the manuscript.

Corresponding author

Correspondence to Lidia Alexa .

Ethics declarations

Competing interests.

The authors declare that they have no competing interests.

Ethical approval

All procedures performed in our study were in accordance with the ethical standards of the Gheorghe Asachi Technical University of Iasi. The research was conducted in accordance with the Code of ethics and professional deontology of the Gheorghe Asachi Technical University of Iasi - TUIASI.COD.01, approved on 21.01.2016, edition 3, rev. 0.

Informed consent

Informed consent was obtained from all individual participants involved in the study. Participants were involved in an information session about the study and had the opportunity to ask questions before fill-up the questionnaire. Participants were informed that they could refuse to complete the questionnaire without penalty or consequences. The questionnaire ensured participants’ anonymity, as no identifying details were required.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Cite this article.

Dosoftei, C.C., Alexa, L. Students’ perception of peer teaching in engineering education: a mixed–method case study. Humanit Soc Sci Commun 11 , 793 (2024). https://doi.org/10.1057/s41599-024-03349-y

Download citation

Received : 10 April 2023

Accepted : 13 June 2024

Published : 20 June 2024

DOI : https://doi.org/10.1057/s41599-024-03349-y

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

The Power of Peer Learning

Fostering Students’ Learning Processes and Outcomes

• Open Access
• © 2023

You have full access to this open access Book

• Omid Noroozi   ORCID: https://orcid.org/0000-0002-0622-289X 0 ,
• Bram De Wever   ORCID: https://orcid.org/0000-0003-4352-4915 1

Education and Learning Sciences, Wageningen University & Research, Wageningen, The Netherlands

You can also search for this editor in PubMed   Google Scholar

Department of Educational Studies, Ghent University, GENT, Belgium

• Discusses cutting-edge pedagogical and technological developments to foster peer learning processes and outcomes
• Presents empirical findings on the relations between peer learning processes and outcomes
• Presents conceptual frameworks, pedagogical designs, and technology-enhanced tools for fostering peer learning
• Introduces new perspectives on peer learning, peer feedback, peer assessment, and peer interaction

Part of the book series: Social Interaction in Learning and Development (SILD)

58k Accesses

17 Citations

32 Altmetric

Buy print copy

Subscribe and save.

• Get 10 units per month
• Download Article/Chapter or eBook
• 1 Unit = 1 Article or 1 Chapter
• Cancel anytime
• Compact, lightweight edition
• Dispatched in 3 to 5 business days
• Free shipping worldwide - see info
• Durable hardcover edition

Tax calculation will be finalised at checkout

About this book

This open access book explores new developments in various aspects of peer learning processes and outcomes. It brings together research studies examining how peer feedback, peer assessment, and small group learning activities can be designed to maximize learning outcomes in higher, but also secondary, education.

Conceptual models and methodological frameworks are presented to guide teachers and educational designers for successful implementation of peer learning activities with the hope of maximizing the effectiveness of peer learning in real educational classrooms.

By providing empirical studies from different peer learning initiatives, both teachers and students in academic and professional contexts are informed about the state of the art developmentsof peer learning.

This book contributes to the understanding of peer learning challenges and solutions in all level of education and provide avenues for future research. It includes theoretical, methodological, and empirical chapters which makes it a useful tool for both teaching and research.

• Peer Learning in Higher Education
• Peer Feedback and Students’ Learning
• Peer Learning and Students’ Motivation
• Peer Learning Processes and Outcomes
• Peer Feedback and Peer Feedforward
• Peer Feedback and Online Learning
• Pedagogical Design of Peer Learning Processes
• Peer Learning and Students’ Individual Characteristics
• Peer Learning and Students’ Learning Perception
• Peer Learning and Students’ Emotional Responses
• Peer Learning and its Challenges

Table of contents (17 chapters)

Front matter, conceptual contributions on peer learning, the four pillars of peer assessment for collaborative teamwork in higher education.

• Bhavani Sridharan, Jade McKay, David Boud

Learning Analytics for Peer Assessment: A Scoping Review

• Kamila Misiejuk, Barbara Wasson

Support Student Integration of Multiple Peer Feedback on Research Writing in Thesis Circles

• Ya Ping Hsiao, Kamakshi Rajagopal

Methodological Contributions on Peer Learning

Peer assessment using criteria or comparative judgement a replication study on the learning effect of two peer assessment methods.

• Tine van Daal, Mike Snajder, Kris Nijs, Hanna Van Dyck

Using Stochastic Actor-Oriented Models to Explain Collaboration Intentionality as a Prerequisite for Peer Feedback and Learning in Networks

• Jasperina Brouwer, Carlos A. de Matos Fernandes

Comparing Expert and Peer Assessment of Pedagogical Design in Integrated STEAM Education

• Kyriaki Α. Vakkou, Tasos Hovardas, Nikoletta Xenofontos, Zacharias C. Zacharia

Technological Contributions on Peer Learning

Constructing computer-mediated feedback in virtual reality for improving peer learning: a synthesis of the literature in presentation research.

• Stan van Ginkel, Bo Sichterman

Web-Based Peer Assessment Platforms: What Educational Features Influence Learning, Feedback and Social Interaction?

• José Carlos G. Ocampo, Ernesto Panadero

Feed-Back About the Collaboration Process from a Group Awareness Tool. Potential Boundary Conditions for Effective Regulation

• Sebastian Strauß, Nikol Rummel

Viewbrics: A Technology-Enhanced Formative Assessment Method to Mirror and Master Complex Skills with Video-Enhanced Rubrics and Peer Feedback in Secondary Education

• Ellen Rusman, Rob Nadolski, Kevin Ackermans

Empirical Contributions on Peer Learning

Peerteach: teaching learners to do learner-centered teaching.

• Soren Rosier

A Thematic Analysis of Factors Influencing Student’s Peer-Feedback Orientation

• Julia Kasch, Peter van Rosmalen, Marco Kalz

Giving Feedback to Peers in an Online Inquiry-Learning Environment

• Natasha Dmoshinskaia, Hannie Gijlers

Peer Interaction Types for Social and Academic Integration and Institutional Attachment in First Year Undergraduates

• Emmeline Byl, Keith J. Topping, Katrien Struyven, Nadine Engels

How to Make Students Feel Safe and Confident? Designing an Online Training Targeting the Social Nature of Peer Feedback

• Morgane Senden, Dominique De Jaeger, Tijs Rotsaert, Fréderic Leroy, Liesje Coertjens

Editors and Affiliations

Education and learning sciences, wageningen university & research, wageningen, the netherlands.

Omid Noroozi

Bram De Wever

About the editors

Omid Noroozi (PhD, 2013) is associate professor at the Education and Learning Sciences group at Wageningen University and Research, the Netherlands. He has fostered an interest in understanding the relations among technology, pedagogy, and learning higher-order skills (e.g. critical thinking, reasoning, problem-solving, communication, collaboration, self-regulation, entrepreneurial thinking) with a specific focus on students’ argumentation competence development in higher education.

Bram De Wever (PhD, 2006) is associate professor at the Department of Educational Studies at Ghent University, Belgium and head of the research group TECOLAB at that department. His research is focusing on technology enhanced learning and instruction, peer assessment and feedback, computer-supported collaborative learning activities, inquiry learning and argumentative writing. Research settings include mostly secondary, higher, and adult education.

Bibliographic Information

Book Title : The Power of Peer Learning

Book Subtitle : Fostering Students’ Learning Processes and Outcomes

Editors : Omid Noroozi, Bram De Wever

Series Title : Social Interaction in Learning and Development

DOI : https://doi.org/10.1007/978-3-031-29411-2

Publisher : Springer Cham

eBook Packages : Education , Education (R0)

Copyright Information : The Editor(s) (if applicable) and The Author(s) 2023

Hardcover ISBN : 978-3-031-29410-5 Published: 21 June 2023

Softcover ISBN : 978-3-031-29413-6 Published: 21 June 2023

eBook ISBN : 978-3-031-29411-2 Published: 20 June 2023

Series ISSN : 2662-5512

Series E-ISSN : 2662-5520

Edition Number : 1

Number of Pages : XIV, 392

Number of Illustrations : 24 b/w illustrations, 44 illustrations in colour

Topics : Study and Learning Skills , Educational Psychology , Education, general

• Publish with us

Policies and ethics

• Find a journal
• Track your research

• Open access
• Published: 02 December 2022

School-based peer education interventions to improve health: a global systematic review of effectiveness

• Steven Dodd 1   na1 ,
• Emily Widnall 2   na1 ,
• Abigail Emma Russell 3 ,
• Esther Louise Curtin 4 ,
• Ruth Simmonds 5 ,
• Mark Limmer 1 &
• Judi Kidger 2  

BMC Public Health volume  22 , Article number:  2247 ( 2022 ) Cite this article

16k Accesses

36 Citations

18 Altmetric

Metrics details

Introduction

Peer education, whereby peers (‘peer educators’) teach their other peers (‘peer learners’) about aspects of health is an approach growing in popularity across school contexts, possibly due to adolescents preferring to seek help for health-related concerns from their peers rather than adults or professionals. Peer education interventions cover a wide range of health areas but their overall effectiveness remains unclear. This review aims to summarise the effectiveness of existing peer-led health interventions implemented in schools worldwide.

Five electronic databases were searched for eligible studies in October 2020. To be included, studies must have evaluated a school-based peer education intervention designed to address the health of students aged 11–18-years-old and include quantitative outcome data to examine effectiveness. The number of interventions were summarised and the impact on improved health knowledge and reductions in health problems or risk-taking behaviours were investigated for each health area separately, the Mixed Methods Appraisal Tool was used to assess quality.

A total of 2125 studies were identified after the initial search and 73 articles were included in the review. The majority of papers evaluated interventions focused on sex education/HIV prevention ( n  = 23), promoting healthy lifestyles ( n  = 17) and alcohol, smoking and substance use ( n  = 16). Papers mainly reported peer learner outcomes (67/73, 91.8%), with only six papers (8.2%) focussing solely on peer educator outcomes and five papers (6.8%) examining both peer learner and peer educator outcomes. Of the 67 papers reporting peer learner outcomes, 35/67 (52.2%) showed evidence of effectiveness, 8/67 (11.9%) showed mixed findings and 24/67 (35.8%) found limited or no evidence of effectiveness. Of the 11 papers reporting peer educator outcomes, 4/11 (36.4%) showed evidence of effectiveness, 2/11 (18.2%) showed mixed findings and 5/11 (45.5%) showed limited or no evidence of effectiveness. Study quality varied greatly with many studies rated as poor quality, mainly due to unrepresentative samples and incomplete data.

School-based peer education interventions are implemented worldwide and span a wide range of health areas. A number of interventions appear to demonstrate evidence for effectiveness, suggesting peer education may be a promising strategy for health improvement in schools. Improvement in health-related knowledge was most common with less evidence for positive health behaviour change. In order to quantitatively synthesise the evidence and make more confident conclusions, there is a need for more robust, high-quality evaluations of peer-led interventions using standardised health knowledge and behaviour measures.

Peer Review reports

Ensuring good health and wellbeing amongst school-aged children is a global public health priority and the contribution schools can make to this goal is increasingly recognised [ 1 ]. Worldwide, we have seen a rise in peer education interventions over recent decades [ 2 ]. For example, a survey in England revealed that 62% of primary and secondary schools had offered a peer-led intervention in 2009 [ 3 ]. Peer-led interventions within school settings are popular for many reasons, including the important role peers play within the lives of young people, a perception that this approach involves relatively few resources, and the more even balance of authority than in teacher-led lessons [ 4 ]. The use of peer educators for health improvement has also been linked with the importance of peer influence in adolescence [ 5 ]. This is a time of increased social development and peer attachments are central to young people’s development, particularly during adolescence [ 5 , 6 ]. Further, there is evidence that young people are more likely to seek help from informal sources of support such as friends in comparison to adults [ 7 ], and of older students being perceived as role models by their younger peers [ 8 ]. Benefits are also likely to exist for peer educators themselves, including opportunities to develop confidence and leadership skills, as well as many schools rewarding peer educators with a qualification or endorsement for their participation [ 9 ].

Existing peer education interventions cover a wide range of health areas, including mental health, physical health, sexual health, and a general promotion of healthy lifestyles including eating habits and smoking prevention [ 10 , 11 , 12 , 13 ]. There is also variation in the format or delivery of peer-led interventions including 1:1 peer mentoring, peer buddy initiatives, peer counselling, and peer education [ 14 , 15 , 16 , 17 ]. This review focuses specifically on peer education, which typically involves the selection and training of ‘peer educators’ or ‘leaders’, who subsequently relay health related information or skills to younger or similar aged students in their school, known as ‘peer learners’ or ‘recipients’.

Summary of related reviews

The current literature on peer education indicates a mixed evidence base regarding its effectiveness.

Ten previous reviews were found concerning health-related peer education among young people [ 10 , 12 , 18 , 19 , 20 , 21 , 22 , 23 , 24 ]. Of these, six concerned sexual health/HIV prevention, two concerned health promotion/education more broadly, one focused on substance abuse and one focused on mental health.

Kim and Free’s review concerning sexual health [ 21 ] found no overall effect of peer education on condom use, mixed findings on sexually transmitted infection (STI) prevention, and positive findings regarding improvements in knowledge, attitudes and intentions. Siddiqui et al. [ 20 ] reviewed peer education programmes for promoting the sexual and reproductive health of young people in India, revealing large variations in the way peer education is implemented as well as mixed effectiveness findings and limited effects of behaviour relative to knowledge. Maticka-Tyndale and Barnet [ 22 ] compiled a review into peer-led interventions to reduce HIV risk among youth using a narrative synthesis, and found that peer interventions led to positive change in knowledge and condom use, and had some success in changing community attitudes and norms, but no significant findings for effects on other sexual behaviours and STI rates. By comparison, Tolli’s review [ 12 ] regarding the effectiveness of peer education interventions for HIV prevention found no clear evidence of peer education effectiveness for HIV prevention, adolescent pregnancy prevention or sexual health promotion in young people of member countries of the European Union.

Mellanby et al. [ 23 ] reviewed the literature comparing peer-led and adult-led school health education and identified eleven studies. Seven of these studies found peer-led to be more effective for health behaviour change than adult-led and three of these studies found peer-led to me more effective for change in knowledge and attitudes. Harden et al. [ 24 ] identified 64 peer-delivered health interventions for young people aged 11 to 24 in any setting (i.e. not restricted to school settings), with only 12 evaluations judged to be methodologically sound. Of these 12, 7 studies (58%) showed a positive effect on at least one behavioural outcome. This review concluded an unclear evidence base for peer-delivered health promotion for young people.

MacArthur et al’s [ 19 ] investigation of peer-led interventions to prevent tobacco, alcohol and/or drug use among young people aged 11–21, comprised a meta-analysis, pooling 10 studies on tobacco use, and found lower prevalence of smoking among those receiving the peer-led interventions compared with controls. The authors also found that peer-led interventions were associated with benefit in relation to alcohol use, and three studies suggested an association with lower odds of cannabis use.

A recent systematic review by King and Fazel of 11 school-based peer-led mental health interventions studies revealed mixed effectiveness [ 10 ]. Some studies showed significant improvements in peer educator self-esteem and social stress [ 25 ], but one study showed an increase in guilt in peer educators [ 26 ]. Two studies also found improvements in self-confidence [ 27 ], and quality of life in peer learners [ 28 ], but one study found an increase in learning stress and decrease in overall mental health scores [ 26 ]. The review concluded there is better evidence if benefits for peer educators compared to peer learners. The summary above of previous systematic assessments of the peer education approach reveals a limited evidence base for school-based peer education interventions. Only two reviews were included regarding school-based peer education, one of which occurred over 20 years ago [ 23 ], while the other [ 10 ] was more narrowly concerned with mental health outcomes.

Despite the widespread use of peer-led interventions, the evidence base across all health areas still remains limited and little is known regarding their overall effectiveness in terms of changing behaviours or increasing health-related knowledge and/or attitudes. Due to the limited evidence base of peer education interventions, this review is broad in scope and will cover global peer education interventions covering all health areas. Although some peer education interventions are targeted towards specific populations, this review focuses on universal interventions available to an entire cohort of students (for example whole class or whole year group). The review aims to summarise the effectiveness of existing peer-led health interventions in schools. This is a review of quantitative data; the qualitative peer education literature will be published in a separate review.

We followed the PICO (Population, Intervention, Comparator and Outcome) format to develop our research question. We completed the systematic review in accordance with the 2009 PRISMA statement [ 29 ] and registered it with PROSPERO (CRD42021229192).

Search strategy and selection criteria

Five electronic databases were searched for eligible studies: CINAHL, Embase, ERIC, MEDLINE and PsycINFO. The list of search terms (see Supplementary Materials ) were developed after scanning relevant literature for key terms. Searches took place during October 2020.

Once the search terms had been agreed amongst the study team, pilot searches were run to check that key texts were appearing. Search terms were subsequently refined and this process was repeated until all key texts appeared. Search strategies such as truncations were used to maximise results. No restrictions were placed on publication date, country or language.

Inclusion/exclusion criteria

To be included studies had to be concerned with school-based peer education interventions designed to address aspects of the health of pupils aged 11–18 years old. We are interested in this age group in particular as it is a period when peers take on a particularly important role in young people’s lives. Peer education interventions concerned with health are defined here as interventions in which school-aged children deliver the education of other pupils for the purposes of improving health outcomes or awareness/literacy relating to health, including knowledge, behaviours and attitudes. Interventions must have taken place within a school, during school hours and must be universal, i.e. not targeted towards a specific sub-group of students or students with a particular health condition.

Where comparators/controls existed, they had to include non-exposure to the interventions concerned, exposure to a differing version of the same intervention, or exposure to the intervention within a substantially differing context.

Papers were excluded from data synthesis if they satisfied any of the following criteria:

Peer education interventions only concerned academic outcomes (e.g., reading and writing achievement).

Interventions concerning anger management, behavioural problems, or social skills.

Interventions concerning traffic safety, health and safety, avoidance of injuries, or first aid.

Interventions concerning cultural, social or political awareness (e.g., media literacy).

Interventions in which health outcomes are secondary to other outcomes (e.g., interventions focused on reading that indirectly improve self-esteem).

One-to-one mentoring interventions.

Conference abstracts, research briefings, commentaries, editorials, study protocol papers and pre-prints.

Primary outcome(s)

Improvements in health, including health awareness and understanding as indicated by responses to questionnaires.

Reductions in health problems or risk-taking behaviours.

These outcomes may concern the peer educators and/or peer learners.

Data extraction, selection and coding

Two reviewers independently screened all papers according to the inclusion criteria above using the Rayyan online review platform. In cases where the reviewers were uncertain, or where the decision was disputed, the decision was discussed and agreed among the wider research team. Two reviewers (SD and EW) then divided the papers between them and independently extracted the data, discussing and queries that arose with each other and the wider team.

Data extraction included the following:

Bibliographic details – authors, year of publication, nation in which intervention was carried out

Aims of the study

Description of study design

Sample size and demographic characteristics.

Context into which the intervention is introduced (characteristics of the school involved, the area in which the school is located, characterisations of the student body, relevant policy considerations).

Description of intervention (including duration of intervention).

Outcome measures (measurement tools, time points of data collection).

Data concerning improvements in health.

Quality appraisal

We used the Mixed Methods Appraisal Tool (MMAT) to assess quality of reporting procedures. This tool consists of five specific quality rating items depending on study design (qualitative, quantitative randomized, quantitative non-randomized, quantitative descriptive and quantitative mixed methods). There are 5 quality questions specific to each study design, so all papers are rated between 0 to 5. The following ratings were used to summarise study quality; 0–1 indicating poor quality, 2–3 indicating average quality and 4–5 indicating high quality. Two reviewers (SD and EW) completed quality ratings on each paper and discussed any discrepancies between them.

Examples of randomized design quality questions included items such as: “ Is randomization appropriately performed ? And “ Are the groups comparable at baseline ?” Examples of non-randomized design quality questions included items such as: “ Are the participants representative of the target population?” and “Are there complete outcome data?”

Effectiveness summary

EW and SD completed data synthesis. Due to the volume of studies, and the large number and heterogeneity of outcome measures, in order to summarise effectiveness, we created the following scoring system to indicate effectiveness:

Significant effects are effects where there was an improvement in health-related outcomes either after the peer education intervention, or when compared to a control group, with a p value of <0.05. Due to the volume of studies and varied follow-up periods, we looked at effectiveness at first follow-up, which in the majority of papers was immediately post-intervention.

A total of 2125 articles were identified after the initial search and 73 articles were eligible for inclusion (see Fig. 1 for a flow diagram of the search). Study designs of the 73 articles were as follows: 23 were controlled trial designs (15 cluster or group randomised, 6 randomised controlled and 2 non-randomised). 15 used randomisation methods but were not controlled trials and the remaining 35 studies used uncontrolled non-randomised methods comparing intervention with a comparison group or using a pre-post survey.

Prisma flow diagram of included studies

Health and geographical areas

The 73 quantitative papers included in this review demonstrated a wide range of health areas. The majority of papers evaluated interventions aimed at sex education/HIV prevention ( n  = 23), promoting healthy lifestyles ( n  = 17) and reducing alcohol, smoking and substance use ( n  = 16). Fig. 2 illustrates number of papers per health area by peer learner or peer educator outcome focus and Table 2 illustrates a summary of proportion of health areas, overall effectiveness and quality ratings.

Number of papers by health area. NB See Supplementary Materials for full description of study designs and outcomes

Papers mainly focussed on peer learner outcomes (67/73, 91.8%), with only six papers (8.2%) focussing solely on peer educator outcomes and only five papers (6.8%) reporting on both peer learner and peer educator outcomes. The majority of papers that focussed on peer educator outcomes were those concerned with sex education (n = 4) and mental health (n = 3).

Papers typically reported knowledge, attitude and/or behavioural outcomes. Of the 73 papers, 42/73 (57.5%) reported knowledge outcomes, 43/73 (58.9%) reported attitude outcomes, 35/73 (47.9%) reported behavioural outcomes and 13/73 (17.8%) reported behavioural intentions.

As well as a broad range of health areas, the papers included in the review also spanned several different countries (Fig. 3 ).

Summary of number of papers by country

We have summarised the results first by student type and then by health area.

Results by student type

Summary of peer learner outcomes.

Of the 67 papers reporting peer learner health outcomes, 35/67 (52.2%) showed evidence of effectiveness (as per our thresholds shown in Table 1 ), 8/67 (11.9%) showed mixed findings and 24/67 (35.8%) found limited or no evidence of effectiveness.

Of the 35 papers that demonstrated effectiveness, 9/35 studies (25.7%) were rated as high quality. Therefore only 9/67 (13.4%) of the total papers showed evidence of effectiveness and were rated as high quality.

Twenty-one papers (31.3%) reported controlled trial designs (including 14 cluster or group randomised, and 5 randomised controlled and 2 non-randomised). Thirteen papers used randomisation methods but were not controlled trials and the remaining 33 papers used uncontrolled non-randomised methods comparing intervention with a comparison group or using a pre-post survey design.

Summary of peer educator outcomes

Of the 11 papers reporting on peer educator health outcomes, 4/11 (36.4%) showed evidence of effectiveness, 2/11 (18.1%) showed mixed findings and 5/11 (45.5%) showed limited or no evidence of effectiveness. Of the 4 papers showing evidence for effectiveness, 2 studies (50%) were rated as high quality.

Four papers had a randomised design comparing intervention vs. control or ‘peer educators vs. classmates’ one of which was a cluster randomised controlled trial. The remaining 7 papers used non-randomised intervention vs. control ( n  = 2) or pre-post survey designs ( n  = 5).

A full table of included studies, outcomes and effectiveness and quality ratings can be found in Supplementary Material 1 .

Results by health area

Sex education/hiv prevention.

Twenty-three studies concerned sex education/HIV prevention [ 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 , 52 ]. 9/23 studies had a randomised design with the 8 studies comparing peer-led to teacher-led or ‘lessons as usual’ and one study comparing peer-led with nurse-led. 14/23 involved non-randomised designs comparing intervention vs. control or a pre-post survey design. Studies covered a wide geographical range, among which there were 7 US studies, but also studies from Canada, UK, Africa, South Africa, Turkey and Greece.

Of the twenty-three papers, 21 reported peer learner outcomes, 4 papers reported peer educator outcomes, with 2 papers reporting on both peer educator and peer learner outcomes. The mean number of participants across the studies was 2033 (range: n  = 106–9000).

8/23 (34.8%) of studies showed evidence of effectiveness, and all studies demonstrating effectiveness consisted of knowledge and attitude outcomes rather than behavioural change.

Only 4/23 studies were rated high in quality (two of which showed evidence of effectiveness), whilst the majority of studies were rated medium quality (15/23) and 4/23 rated as low quality.

Healthy lifestyles (exercise, nutrition, oral health, health information)

Seventeen studies reported interventions addressing healthy lifestyles [ 53 , 54 , 55 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 , 64 , 65 , 66 , 67 , 68 , 69 ]. Of these papers, ten used a randomised controlled trial design primarily comparing peer-led vs. teacher-led or ‘lessons as usual’, but two oral health papers also used a dentist-led condition. Seven papers used non-randomised research designs comparing intervention vs. control or a pre-post survey design.

The most common focus was nutrition and exercise, but interventions also covered oral health, accessing health information online and interventions taking a more general approach to health improvement. Regarding geographical spread, 5/17 papers reported interventions carried out in the USA, with Australia, China, India and UK represented by two papers per country.

Sixteen of the seventeen papers reported peer learner outcomes, and only one reported peer educator outcomes. The mean number of participants per intervention was 1245 (range: n  = 76–4576).

7/17 papers in this health area were shown to be effective, 8/17 were found to be ineffective, and 2/17 showed mixed results. In other words, less than half (41.1%) showed evidence of effectiveness. Of the studies demonstrating effectiveness, the outcomes largely centred around knowledge and attitudes, but one study did demonstrate positive behaviour change [ 62 ].

Over half of the studies (9/17) were rated as high quality, 4/17 were rated medium quality and 4/17 low quality. Of the studies showing evidence for effectiveness, 4/7 (57.1%) were rated as high quality.

Alcohol, smoking, substance use

Sixteen papers were classified within the category of alcohol, smoking and substance use [ 70 , 71 , 72 , 73 , 74 , 75 , 76 , 77 , 78 , 79 , 80 , 81 , 82 , 83 , 84 , 85 ]. Ten of these papers had a randomised design (including 3 cluster randomised controlled trials) comparing peer-led (intervention) vs. teacher-led (control). Six papers were non-randomised and used either a pre-post survey design or intervention vs. control. The 16 papers varied in quality with six rated ‘high quality’, seven rated ‘medium quality’, and three rated ‘low quality’. Studies took place across more than 10 countries with one study being conducted internationally. The mean number of participants across all studies was 2165 (range: n = 105–10,730).

Fifteen papers evaluated the effect of the intervention on peer learner outcomes and only one paper evaluated the effect of the intervention on peer educator outcomes. 8/16 (50%) papers showed evidence of effectiveness. 2/16 (12.5%) papers showed mixed findings and 6/16 (37.5%) showed little to no evidence for effectiveness, including the peer educator outcome paper. Of the eight papers demonstrating evidence for effectiveness, only four (50%) were rated as high quality.

Of the studies demonstrating effectiveness, there was a combination of knowledge, attitude and behavioural outcomes, but more evidence for positive changes in knowledge and attitude.

Mental health and well-being

Six studies assessed mental health and well-being [ 27 , 86 , 87 , 88 , 89 , 90 ]. This category was inclusive of common mental health problems, self-harm and suicide prevention as well as broader topics such as self-esteem and social connectedness. Four of the six studies used non-randomised pre-post survey designs and two studies used randomised design, one of which was a cluster randomised controlled trial.

Of the six studies, 5/6 explored peer learner outcomes, 3/6 explored peer educator outcomes, 2 of which explored both peer learner and peer educator outcomes. The average sample size across the seven mental health studies was 1118 (range: n  = 50–4128).

Study quality was mixed, with two studies rated as high quality, three medium quality and one low quality. Outcome measures largely consisted of knowledge and attitude questionnaires, help-seeking behaviour and help-seeking confidence as well as condition-specific measures including body satisfaction and self-report of emotional and behavioural difficulties.

The majority of mental health studies (5/6) were rated as showing evidence for effectiveness and one study was rated ineffective. Of the studies demonstrating effectiveness, only one reported positive behaviour change (help-seeking behaviours) and this behaviour changed was observed in peer educators as opposed to peer learners [ 86 ].

Disease prevention

Four studies assessed outcomes relating to disease prevention [ 91 , 92 , 93 , 94 ] which included hepatitis, tuberculosis, cervical cancer and blood borne diseases. All four studies focused on peer learner outcomes and one study also included peer educator outcomes. Three of the four studies were non-randomised pre-post survey designs and one study was randomised. The average sample size across the four studies was 2116 (range: 1265–2930).

Three out of the four studies (75%) showed evidence for effectiveness and one study showed mixed results. No studies were rated as high quality, three were rated medium and one was rated low.

Outcomes were largely knowledge or intention based. Studies showing effectiveness mostly related to knowledge, intentions and attitudes and one study did find a positive change in behaviour [ 93 ].

Five included studies assessed asthma interventions [ 95 , 96 , 97 , 98 , 99 ]. 4/5 of these were randomised trials and one study used a non-randomised pre-post survey design. Average sample size across all studies was 427 (range: n  = 203–935). Three studies took place in Australia and two in the US. All papers evaluated the impact of the intervention on peer learner outcomes with none focussing on peer educator outcomes.

4/5 studies showed evidence for effectiveness with only one study showing no evidence for effectiveness. All studies were rated as medium quality. Measures ranged from asthma knowledge, quality of life, school absenteeism, asthma attacks at school and asthma tests. Effectiveness was largely observed for knowledge outcomes, there was less evidence for asthma attacks or symptoms.

Two studies conducted in Italy assessed bullying by evaluating the ‘NoTrap!’ anti-bullying intervention [ 100 , 101 ]. The first study rated as high quality, evaluated two independent trials and focussed on peer learner outcomes ( n  = 622; n  = 461). This study found significant reductions in victimization, bullying, cybervictimization and cyberbullying and was rated as high quality. The second study, rated as medium quality, focussed on peer educator outcomes ( n  = 524) and used a non-randomised, pre-post survey design but overall, only showed some evidence of effectiveness amongst males in terms of reduced victimization and increased prosocial behaviour and social support. No evidence was found for effectiveness among females.

Peer education interventions to improve student health cover a wide variety of topics and are used globally. This review aimed to summarise the results from peer education health interventions in secondary school students (aged 11–18-years-old), which were universal (rather than targeted interventions of sub-groups of students) and carried out at school.

Due to the heterogeneity of findings, range of health areas, types of studies and diversity of outcome measurements used, it was not possible to perform a meta-analysis or formal data synthesis to assess effectiveness. However, some broad conclusions can be made. A number of interventions appear to demonstrate evidence for effectiveness which indicates that peer education interventions can be an important school-based intervention for health improvement. Asthma interventions appeared to be particularly effective. In terms of outcome measures, the strongest evidence was for a positive change in knowledge and attitude measures, but there was less evidence overall for health behaviour outcomes which supports previous findings [ 20 , 22 ].

Although many studies did demonstrate positive results, findings overall were very mixed and several studies were of poor quality. In addition to the shortcomings picked up on by our quality appraisal, many papers lacked methodological detail and clarity regarding the intervention procedure, particularly in regard to how peer educators were selected and trained, which seems to be an important factor in those studies that found positive results and was also emphasised in a previous review [ 10 ]. Further, there were widespread problems of data reporting including noting ‘significant’ results without providing any measure of effect size or between-study variability. Other problems included selective reporting of results, such as selective emphasis on anomalous positive results, or only revealing measures of statistical significance in the case of positive effects. Interestingly, there did not appear to be a relationship between study quality and findings, given that several studies rated as effective were rated both high and low quality with a similar picture for studies showing mixed effectiveness and ineffectiveness.

In terms of frequency of health areas covered, our findings are similar to a recent ‘review of reviews’ of peer education for health and wellbeing which found that the majority of reviews focused on sexual health and HIV/AIDS interventions [ 13 ]. This previous review focused on both children and adults, however, in line with our findings, it found mixed effectiveness and considerable diversity in methods, findings and rigour of evaluation. It was particularly noted that details of peer educator training were rarely provided in HIV/AIDS interventions which supports our findings. Notably, however, the quality of studies was actually highest for peer education programs in HIV/AIDS, which differed to our review which found few studies rated as high quality. This discrepancy may be due to the different measures used to assess quality. Like our study, this review concluded that each health area showed some promising results, but also pointed to a need for higher levels of quality and rigour in future evaluations.

Despite the rising prevalence in mental health difficulties, there were relatively few studies focused on mental health outcomes, particularly more general preventative approaches to mental health and well-being, with many of the included studies focusing on suicide prevention, self-harm or specific disorders. However, many of mental health studies included in this review showed evidence for effectiveness, suggesting peer education approaches for mental health should be further studied and evaluated.

Another key finding of our review is that papers tended to focus more on peer learner outcomes and therefore impacts of peer-led interventions on peer educators themselves appear to be under-explored. This has been reported by previous reviews [ 10 ] and highlights the importance of examining and comparing both peer educators’ and learners’ outcomes within studies. In this context, we found more evidence of peer learners benefitting from the interventions, with 55.2% of studies showing a positive effect, versus only 36.4% for peer educators. This contrasted with a previous review of mental health interventions that concluded peer educators seemed to yield more benefits from participating in the interventions, possibly due to the attention they are given during training and throughout the programmes [ 10 ].

Although common measures existed across studies, including health knowledge, health intentions, and health behaviours, many studies used novel or unvalidated measurements, indicating a need for more standardised health literacy measures and a need for future validation work in this area. This supports two systematic reviews carried out in 2015, firstly a review of health literacy measures which found a lack of comprehensive instruments to measure health literacy and suggested the need for the development of new instruments [ 102 ], and secondly a review of mental health literacy measures which found a number of unvalidated measures and lack of measures that measured all components of mental health literacy concurrently [ 103 ].

Although there are a number of existing reviews summarising the extent to which peer education may improve young peoples health, the literature is still lacking on why peer education is effective within the quantitative literature. It remains unclear which mechanisms involved in peer education lead to its effectiveness (or ineffectiveness). Although many peer education studies are grounded in theory such as Diffusion of Innovation Theory [ 104 ] and Bandura’s Social Cognitive/Social Learning Theory [ 105 , 106 ], the literature is lacking a more nuanced analysis of the mechanisms through which peer education improve young people’s health. This is therefore a key area for future research.

A recent review of peer education and peer counselling for health and well-being highlights how peer education interventions are inherently difficult to quality control and evaluate [ 13 ], partly due to what makes peer education attractive; peer education defies the conventions of traditional formal education and allows young people to learn by more unstructured means, in more ‘real world’ ways, benefiting from meaningful examples and conversations with their peers. Although there are an increasing number of well-designed peer education studies [ 13 ], new evaluation methods may be needed given the complexity and multi-component nature of peer-education approaches (i.e., training, more informal teaching approaches and informal diffusion of knowledge).

Limitations

Despite our review being comprehensive, we acknowledge certain limitations. ‘Peer education’ is a complex and widely contested term and therefore how studies described their approach varied substantially. This may have meant some relevant studies were not picked up from our initial search. A previous review [ 10 ] also noted this potential limitation, with unclear and heterogeneous methods precluding meta-analysis. Therefore, a consensus on how to define ‘peer education’ and using standardised measures to assess effectiveness would facilitate more definitive synthesis of the evidence. Another potential limitation of our approach is that we only searched scientific databases, and therefore could have missed important evidence in the grey literature as we retrieved a relatively small number of initial records ( n  = 2125). Despite this, given the wide variety of study type, age range, health area and country reviewed, this suggests our search strategy was fairly robust, and yielded results that were representative of the breadth in the current literature base.

This review focussed on universal peer education interventions delivered within the secondary school setting during school hours. Further research could explore the effectiveness of varying forms of peer education including 1:1 mentoring, more targeted (not universal) interventions, as well as peer education interventions in other settings including youth clubs or community and local organisations.

Due to the breadth of this review, we did not conduct a detailed comparison between knowledge, attitude and behavioural outcomes, however the studies demonstrating effectiveness tended to show positive change on knowledge and attitude outcomes, but less evidence was seen for positive behavioural change. This is in line with previous reviews which have suggested that peer education better improves health knowledge but often does not lead to behavioural gains [ 13 , 107 ]. To this vein, it remains unclear the differential impact on behavioural intention and actual performance of behaviour, and therefore we urge future researchers to measure outcomes relating to knowledge and attitude, intentions, and actual behaviour in order to synthesise the evidence in a more standardised way. Although the literature is heterogeneous, there is available data to conduct distinct analysis on different outcome measures (knowledge, attitude and behaviour) to create a more nuanced understanding of each health area.

Given the large number of studies and variation in outcome measures (behaviour, knowledge, attitude), this review focussed on findings at first follow-up (usually immediately after intervention) and therefore the effectiveness findings are not likely to represent longer-term effects of peer education interventions, which would require further research. In addition, due to the low number of optimally designed randomised-controlled trials identified, our review could not meaningfully compare results between randomised and non-randomised studies. However, as more high quality trials continue to be published in this growing area of research, a future review could be conducted that looks into the effect of randomisation on young people’s outcomes. Our results also focused on p-values rather than effect sizes due to the large variability in how and what studies measures, future researchers should aim to agree on more standardises ways of measuring outcomes to enable better synthesis.

To conclude, school-based peer education interventions occur worldwide and span a number of health areas. A number of interventions appear to demonstrate evidence for effectiveness, suggesting peer education may be a promising strategy for health improvement in schools. However overall evidence for effectiveness and study quality are mixed. Improvement in health-related knowledge was most common with less evidence for positive health behaviour change. In order to synthesise the evidence and make more confident conclusions, it is imperative that more robust, high-quality evaluations of peer-led interventions are conducted and that studies follow reporting guidelines to describe their methods and results in sufficient detail so that meta-analyses can be conducted. In addition, further research is needed to develop understanding of the intervention mechanisms that lead to health improvement in peer education approaches as well as more focussed work on standardising and validating health literacy and behaviour measurement tools.

Pre-registration

This review was pre-registered on PROSPERO: CRD42021229192. One deviation was made from the original protocol which was the use of a different quality appraisal tool. Initially we had planned to use the Canadian Effective Public Health Project Practice (EPHPP) Quality Assessment Tool for Quantitative Studies and the Critical Appraisals Skills Programme (CASP) checklist for qualitative studies. The authors instead used a combined mixed methods tool (the Mixed Methods Appraisal Tool; MMAT) for both quantitative and qualitative studies. This was due to the large volume and variation of studies which meant there were benefits to using a single brief quality check tool across all included studies, allowing us to standardise scores across study types. The qualitative studies will be discussed in a separate realist review on key mechanisms of peer education interventions.

Availability of data and materials

All data generated or analysed during this study are included in this published article and its supplementary information files.

Fazel M, Hoagwood K, Stephan S, Ford T. Mental health interventions in schools 1: Mental health interventions in schools in high-income countries. Lancet Psychiatry. 2014;1(5):377–87.

Article   PubMed   PubMed Central   Google Scholar  

McKeganey SP, Neil. The rise and rise of peer education approaches. Drugs. 2000;7(3):293–310.

Google Scholar  

Houlston C, Smith PK, Jessel J. Investigating the extent and use of peer support initiatives in English schools. Educ Psychol. 2009;29(3):325–44.

Article   Google Scholar  

Winterton CI, Dunk RD, Wiles JR. Peer-led team learning for introductory biology: relationships between peer-leader relatability, perceived role model status, and the potential influences of these variables on student learning gains. Discipli Interdisciplin Sci Educ Res. 2020;2(1):1–9.

Blakemore SJ, Robbins TW. Decision-making in the adolescent brain. Nat Neurosci. 2012;15(9):1184–91.

Article   CAS   PubMed   Google Scholar  

Lam CB, McHale SM, Crouter AC. Time with peers from middle childhood to late adolescence: developmental course and adjustment correlates. Child Dev. 2014;85(4):1677–93.

NHS Digital. Mental Health of Children and Young People in England, 2020: Wave 1 follow up to the 2017 Survey. England: Health and Social Care Information Centre; 2020. Available online: https://digital.nhs.uk/data-and-information/publications/statistical/mental-health-of-children-and-young-people-in-england/2020-wave-1-follow-up# . Accessed 01 Aug 2022.

Johnson EC, Robbins BA, Loui M. What do students experience as peer leaders of learning teams? What Do Students Experience as Peer Leaders of Learning Teams? 2015.

Morgan D, Robbins J, Tripp J. Celebrating the Achievements of Sex and Relationship Peer Educators: The Development of an Assessment Process. Sex Educ. 2004;4(2):167–83.

King T, Fazel M. Examining the mental health outcomes of school-based peer-led interventions on young people: A scoping review of range and a systematic review of effectiveness. PLoS One. 2021;16(4):e0249553.

Article   CAS   PubMed   PubMed Central   Google Scholar  

Boyle J, Mattern CO, Lassiter JW, Ritzler JA. Peer 2 peer: Efficacy of a course-based peer education intervention to increase physical activity among college students. J Am Coll Heal. 2011;59(6):519–29.

Tolli MV. Effectiveness of peer education interventions for HIV prevention, adolescent pregnancy prevention and sexual health promotion for young people: a systematic review of European studies. Health Educ Res. 2012;27(5):904–13.

Topping KJ. Peer Education and Peer Counselling for Health and Well-Being: A Review of Reviews. Int J Environ Res Public Health. 2022;19(10):6064.

Dennison S. Peer mentoring: Untapped potential. J Nurs Educ. 2010;49(6):340–2.

Article   PubMed   Google Scholar  

Thalluri J, O'Flaherty JA, Shepherd PL. Classmate peer-coaching:" A Study Buddy Support scheme". J Peer Learn. 2014;7(1):92–104.

Boulton MJ. School peer counselling for bullying services as a source of social support: a study with secondary school pupils. Bri J Guidance Counsel. 2005;33(4):485–94.

Abdi F, Simbar M. The peer education approach in adolescents-narrative review article. Iran J Public Health. 2013;42(11):1200.

PubMed   PubMed Central   Google Scholar  

Mahat G, Scoloveno MA. Effectiveness of adolescent peer education programs on reducing HIV/STI risk: an integrated review. Res Theory Nurs Pract. 2018;32(2):168–98.

MacArthur GJ, Harrison S, Caldwell DM, Hickman M, Campbell R. Peer-led interventions to prevent tobacco, alcohol and/or drug use among young people aged 11–21 years: a systematic review and meta-analysis. Addiction. 2016;111(3):391–407.

Siddiqui M, Kataria I, Watson K, Chandra-Mouli V. A systematic review of the evidence on peer education programmes for promoting the sexual and reproductive health of young people in India. Sex Reprod Health Matters. 2020;28(1):1741494.

Kim CR, Free C. Recent evaluations of the peer-led approach in adolescent sexual health education: A systematic review. Perspect Sex Reprod Health. 2008;40(3):144–51.

Maticka-Tyndale E, Barnett JP. Peer-led interventions to reduce HIV risk of youth: a review. Eval Program Plann. 2010;33(2):98–112.

Mellanby AR, Rees JB, Tripp JH. Peer-led and adult-led school health education: a critical review of available comparative research. Health Educ Res. 2000;15(5):533–45.

Harden A, Oakley A, Oliver S. Peer-delivered health promotion for young people: a systematic review of different study designs. Health Educ J. 2001;60(4):339–53.

Yogev A, Ronen R. Cross-age tutoring: Effects on tutors’ attributes. J Educ Res. 1982;75(5):261–8.

Song Y, Loewenstein G, Shi Y. Heterogeneous effects of peer tutoring: Evidence from rural Chinese middle schools. Res Econ. 2018;72(1):33–48.

Ellis LA, Marsh HW, Craven RG. Addressing the challenges faced by early adolescents: a mixed-method evaluation of the benefits of peer support. Am J Community Psychol. 2009;44(1–2):54–75.

Shah S, McCallum GB, Wilson C, Saunders J, Chang AB. Feasibility of a peer-led asthma and smoking prevention program (ASPP) in australian schools with high indigenous youth. Respirology. 2017;22:35.

Moher D, Shamseer L, Clarke M, Ghersi D, Liberati A, Petticrew M, et al. Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement. Syst Rev. 2015;4(1):1.

Parwej S, Kumar R, Walia I, Aggarwal AK. Reproductive health education intervention trial. Indian J Pediatr. 2005;72(4):287–91.

Rotz D, Goesling B, Manlove J, Welti K, Trenholm C. Impacts of a School-Wide, Peer-Led Approach to Sexuality Education: A Matched Comparison Group Design. J Sch Health. 2018;88(8):549–59.

Mellanby AR, Newcombe RG, Rees J, Tripp JH. A comparative study of peer-led and adult-led school sex education. Health Educ Res. 2001;16(4):481–92.

Timol F, Vawda MY, Bhana A, Moolman B, Makoae M, Swartz S. Addressing adolescents' risk and protective factors related to risky behaviours: Findings from a school-based peer-education evaluation in the Western Cape. SAHARA J. 2016;13(1):197–207.

Aten MJ, Siegel DM, Enaharo M, Auinger P. Keeping middle school students abstinent: outcomes of a primary prevention intervention. J Adolesc Health. 2002;31(1):70–8.

Caron F, Godin G, Otis J, Lambert LD. Evaluation of a theoretically based AIDS/STD peer education program on postponing sexual intercourse and on condom use among adolescents attending high school. Health Educ Res. 2004;19(2):185–97.

Ebreo A, Feist-Price S, Siewe Y, Zimmerman RS. Effects of peer education on the peer educators in a school-based HIV prevention program: where should peer education research go from here?...including commentary by Main DS. Health Educ Behav. 2002;29(4):411–24.

Mason-Jones AJ, Flisher AJ, Mathews C. Who are the peer educators? HIV prevention in South African schools. Health Educ Res. 2011;26(3):563–71.

Menna T, Ali A, Worku A. Effects of peer education intervention on HIV/AIDS related sexual behaviors of secondary school students in Addis Ababa, Ethiopia: a quasi-experimental study. Reprod Health. 2015;12(1):84.

Siegel DM, Aten MJ, Roghmann KJ, Enaharo M. Early effects of a school-based human immunodeficiency virus infection and sexual risk prevention intervention. Arch Pediatr Adolesc Med. 1998;152(10):961–70.

Siegel DM, Aten MJ, Enaharo M. Long-term effects of a middle school- and high school-based human immunodeficiency virus sexual risk prevention intervention. Arch Pediatr Adolesc Med. 2001;155(10):1117–26.

Stephenson JM, Strange V, Forrest S, Oakley A, Copas A, Allen E, et al. Pupil-led sex education in England (RIPPLE study): cluster-randomised intervention trial. Lancet. 2004;364(9431):338–46.

Stephenson J, Strange V, Allen E, Copas A, Johnson A, Bonell C, et al. The long-term effects of a peer-led sex education programme (RIPPLE): a cluster randomised trial in schools in England. PLoS Med. 2008;5(11):224 discussion e.

Strange V, Forrest S, Oakley A. What influences peer-led sex education in the classroom? A view from the peer educators. Health Educ Res. 2002;17(3):339–49.

Borgia P, Marinacci C, Schifano P, Perucci CA. Is peer education the best approach for HIV prevention in schools? Findings from a randomized controlled trial. J Adolesc Health. 2005;36(6):508–16.

Fisher JD, Fisher WA, Bryan AD, Misovich SJ. Information-motivation-behavioral skills model-based HIV risk behavior change intervention for inner-city high school youth. Health Psychol. 2002;21(2):177–86.

Huang H, Ye X, Cai Y, Shen L, Xu G, Shi R, et al. Study on peer-led school-based HIV/AIDS prevention among youths in a medium-sized city in China. Int J STD AIDS. 2008;19(5):342–6.

Mahat G, Scoloveno MA. HIV peer education: Relationships between adolescents' HIV/AIDS knowledge and self-efficacy. J HIV AIDS Soc Serv. 2010;9(4):371–84.

Merakou K, Kourea-Kremastinou J. Peer education in HIV prevention: an evaluation in schools. Eur J Pub Health. 2006;16(2):128–32.

Michielsen K, Beauclair R, Delva W, Roelens K, Van Rossem R, Temmerman M. Effectiveness of a peer-led HIV prevention intervention in secondary schools in Rwanda: results from a non-randomized controlled trial. BMC Public Health. 2012;12(1):729.

Ozcebe H, Akin L, Aslan D. A peer education example on HIV/AIDS at a high school in Ankara. Turk J Pediatr. 2004;46(1):54–9.

PubMed   Google Scholar  

Visser MJ. HIV/AIDS prevention through peer education and support in secondary schools in South Africa. Sahara J. 2007;4(3):678–94.

Jennings JM, Howard S, Perotte CL. Effects of a school-based sexuality education program on peer educators: the Teen PEP model. Health Educ Res. 2014;29(2):319–29.

Cohen RY, Felix MR, Brownell KD. The role of parents and older peers in school-based cardiovascular prevention programs: implications for program development. Health Educ Q. 1989;16(2):245–53.

Lytle LA, Murray DM, Perry CL, Story M, Birnbaum AS, Kubik MY, et al. School-based approaches to affect adolescents' diets: results from the TEENS study. Health Educ Behav. 2004;31(2):270–87.

Shankar P, Sievers D, Sharma R. Evaluating the Impact of a School-Based Youth-Led Health Education Program for Adolescent Females in Mumbai, India. Ann Glob Health. 2020;86(1):57.

Forneris T, Fries E, Meyer A, Buzzard M, Uguy S, Ramakrishnan R, et al. Results of a rural school-based peer-led intervention for youth: goals for health. J Sch Health. 2010;80(2):57–65.

Foley BC, Shrewsbury VA, Hardy LL, Flood VM, Byth K, Shah S. Evaluation of a peer education program on student leaders' energy balance-related behaviors. BMC Public Health. 2017;17(1):695.

Cui Z, Shah S, Yan L, Pan Y, Gao A, Shi X, et al. Effect of a school-based peer education intervention on physical activity and sedentary behaviour in Chinese adolescents: a pilot study. BMJ Open. 2012;2(3):e000721.

Ishak SIZS, Siew CY, Mohd Shariff Z, Mun CY, Moh TN. Effectiveness of a school-based, peer-led intervention program on the adolescents' body composition, eating behaviors and health-related quality of life. Ann Nutr Metab. 2019;75(3):359.

Tamiru D, Argaw A, Gerbaba M, Ayana G, Nigussie A, Jisha H, et al. Enhancing Personal Hygiene Behavior and Competency of Elementary School Adolescents through Peer-Led Approach and School-Friendly: A Quasi-Experimental Study. Ethiop J Health Sci. 2017;27(3):245–54.

Bogart LM, Elliott MN, Cowgill BO, Klein DJ, Hawes-Dawson J, Uyeda K, et al. Two-Year BMI Outcomes From a School-Based Intervention for Nutrition and Exercise: A Randomized Trial. Pediatrics. 2016;137(5):e20152493.

Shrewsbury VA, Venchiarutti RL, Hardy LL, Foley BC, Bonnefin A, Byth K, et al. Impact and cost of the peer-led Students As LifeStyle Activists programme in high schools. Health Educ J. 2020;79(1):3–20.

Bell SL, Audrey S, Cooper AR, Noble S, Campbell R. Lessons from a peer-led obesity prevention programme in English schools. Health Promot Int. 2017;32(2):250–9.

Bogart LM, Cowgill BO, Elliott MN, Klein DJ, Hawes-Dawson J, Uyeda K, et al. A randomized controlled trial of students for nutrition and eXercise: a community-based participatory research study. J Adolesc Health. 2014;55(3):415–22.

Ajuwon GA, Ajuwon AJ. Teaching high school students to use online consumer health resources on mobile phones: outcome of a pilot project in Oyo State, Nigeria. J Med Lib Assoc. 2019;107(2):194–202.

Haleem A, Siddiqui MI, Khan AA. School-based strategies for oral health education of adolescents--a cluster randomized controlled trial. BMC Oral Health. 2012;12:54.

Vangipuram S, Jha A, Raju R, Bashyam M. Effectiveness of peer group and conventional method (Dentist) of oral health education programme among 12-15 year old school children - A randomized controlled trial. J Clin Diagn Res. 2016;10(5):ZC125–ZC9.

Sebire SJ, Jago R, Banfield K, Edwards MJ, Campbell R, Kipping R, et al. Results of a feasibility cluster randomised controlled trial of a peer-led school-based intervention to increase the physical activity of adolescent girls (PLAN-A). Int J Behav Nutr Phys Act. 2018;15(1):50.

Ping HU, Lingli HAN, Manoj S, Huan Z, Yong Z, Hui LI, et al. Evaluation of Cognitive and Behavioral Effects of Peer Education Model-Based Intervention to Sun Safe in Children. Iran J Public Health. 2014;43(3):300–9.

Perry CL, Grant M, Ernberg G, Florenzano RU, Langdon MC, Myeni AD, et al. WHO Collaborative Study on Alcohol Education and Young People: outcomes of a four-country pilot study. Int J Addict. 1989;24(12):1145–71.

Weichold K, Silbereisen RK. Peers and teachers as facilitators of the life skills program IPSY - Results from a pilot study. Sucht. 2012;58(4):247–58.

Lachausse RG. The effectiveness of a multimedia program to prevent fetal alcohol syndrome. Health Promot Pract. 2008;9(3):289–93.

Erhard R. Peer-led and adult-led programs--student perceptions. J Drug Educ. 1999;29(4):295–308.

Audrey S, Holliday J, Campbell R. It's good to talk: adolescent perspectives of an informal, peer-led intervention to reduce smoking. Soc Sci Med. 2006;63(2):320–34.

Bloor M, Frankland J, Langdon NP, Robinson M, Allerston S, Catherine A, et al. A controlled evaluation of an intensive, peer-led, schools-based, anti-smoking programme. Health Educ J. 1999;58(1):17–25.

Campbell R, Starkey F, Holliday J, Audrey S, Bloor M, Parry-Langdon N, et al. An informal school-based peer-led intervention for smoking prevention in adolescence (ASSIST): a cluster randomised trial. Lancet. 2008;371(9624):1595–602.

Lotrean LM, Dijk F, Mesters I, Ionut C, De Vries H. Evaluation of a peer-led smoking prevention programme for Romanian adolescents. Health Educ Res. 2010;25(5):803–14.

Mall ASK, Bhagyalaxmi A. An Informal School-based, Peer-led Intervention for Prevention of Tobacco Consumption in Adolescence: A Cluster Randomized Trial in Rural Gandhinagar. Indian J Community Med. 2017;42(3):143–6.

Mohammadi M, Ghaleiha A, Rahnama R. Effectiveness of a peer-led behavioral intervention program on tobacco use-related knowledge, attitude, normative beliefs, and intention to smoke among adolescents at Iranian Public High Schools. Int J Prev Med. 2019;10(1):260245.

Murray DM, Richards PS, Luepker RV, Johnson CA. The prevention of cigarette smoking in children: two- and three-year follow-up comparisons of four prevention strategies. J Behav Med. 1987;10(6):595–611.

Perry CL, et al. Peer Teaching and Smoking Prevention among Junior High Students. Adolescence. 1980;15(58):277–82.

CAS   PubMed   Google Scholar  

Botvin GJ, Baker E, Filazzola AD, Botvin EM. A cognitive-behavioral approach to substance abuse prevention: one-year follow-up. Addict Behav. 1990;15(1):47–63.

Demirezen D, Karaca A, Konuk Sener D, Ankarali H. Agents of change: the role of the peer education program in preventing adolescent substance abuse. J Child Adolesc Subst Abuse. 2019;28(5):376–87.

Severson HH, Glasgow R, Wirt R, Brozovsky P, Zoref L, Black C, et al. Preventing the use of smokeless tobacco and cigarettes by teens: results of a classroom intervention. Health Educ Res. 1991;6(1):109–20.

Aslan D, Sahin A. Adolescent peers and anti-smoking activities. Promot Educ. 2007;14(1):36–40.

Wyman PA, Brown CH, LoMurray M, Schmeelk-Cone K, Petrova M, Yu Q, et al. An outcome evaluation of the Sources of Strength suicide prevention program delivered by adolescent peer leaders in high schools. Am J Public Health. 2010;100(9):1653–61.

Ciao AC, Latner JD, Brown KE, Ebneter DS, Becker CB. Effectiveness of a peer-delivered dissonance-based program in reducing eating disorder risk factors in high school girls. Int J Eat Disord. 2015;48(6):779–84.

Eisenstein C, Zamperoni V, Humphrey N, Deighton J, Wolpert M, Rosan C, et al. Evaluating the peer education project in secondary schools. J Public Ment Health. 2019;18(2):58–65.

Kaveh MH, Hesampour M, Ghahremani L, Tabatabaee HR. The effects of a peer-led training program on female students' self-esteem in public secondary schools in Shiraz. J Adv Med Educ Prof. 2014;2(2):63–70.

Parikh SV, Taubman DS, Antoun C, Cranford J, Foster CE, Grambeau M, et al. The Michigan Peer-to-Peer Depression Awareness Program: School-Based Prevention to Address Depression Among Teens. Psychiatr Serv. 2018;69(4):487–91.

Isik M, Set T, Khan AS, Avsar UZ, Cansever Z, Acemoglu H. Prevalence of Blood Brotherhood among High School Students in Erzurum and the Effect of Peer-led Education on this Practice. Eurasian J Med. 2013;45(2):83–7.

Sadoh AE, Okonkwo C, Nwaneri DU, Ogboghodo BC, Eregie C, Oviawe O, et al. Effect of peer education on knowledge of human papilloma virus and cervical cancer among female adolescent students in Benin city, Nigeria. Ann Global Health. 2018;84(1):121–8.

Acemoglu H, Palanci Y, Set T, Vancelik S, Isik M, Polat H. An intervention study for viral hepatitis: Peer-led health education among high school students. Saudi Med J. 2011;32(2):183–7.

Liu Q, Liu L, Vu H, Liu X, Tang S, Wang H. Comparison between peer-led and teacher-led education in tuberculosis prevention in rural middle schools in Chongqing, China. Asia Pac J Public Health. 2015;27(2):NP2101–NP11.

Al-sheyab N, Gallagher R, Crisp J, Shah S. Peer-led education for adolescents with asthma in Jordan: a cluster-randomized controlled trial. Pediatrics. 2012;129(1):e106–12.

Al-sheyab NA, Alomari MA, Shah S, Gallagher R. "Class smoke-free" pledge impacts on nicotine dependence in male adolescents: A cluster randomized controlled trial. J Subst Abus. 2016;21(6):566–74.

Gibson PG, Shah S, Mamoon HA. Peer-led asthma education for adolescents: impact evaluation. J Adolesc Health. 1998;22(1):66–72.

McCallum GB, Chang AB, Wilson CA, Petsky HL, Saunders J, Pizzutto SJ, et al. Feasibility of a Peer-Led Asthma and Smoking Prevention Project in Australian Schools with High Indigenous Youth. Front Pediatr. 2017;5:33.

Shah S, Peat JK, Mazurski EJ, Wang H, Sindhusake D, Bruce C, et al. Effect of peer led programme for asthma education in adolescents: cluster randomised controlled trial. BMJ. 2001;322(7286):583–5.

Palladino BE, Nocentini A, Menesini E. Evidence-based intervention against bullying and cyberbullying: Evaluation of the NoTrap! program in two independent trials. Aggress Behav. 2016;42(2):194–206.

Zambuto V, Palladino BE, Nocentini A, Menesini E. Why do some students want to be actively involved as peer educators, while others do not? Findings from NoTrap! Anti-bullying and anti-cyberbullying program. Eur J Dev Psychol. 2019;16(4):373–86.

Tavousi M, Ebadi M, Fattahi E, Jahangiry L, Hashemi A, Hashemiparast M, et al. Health literacy measures: A systematic review of the literature. 2015.

Wei Y, McGrath PJ, Hayden J, Kutcher S. Mental health literacy measures evaluating knowledge, attitudes and help-seeking: a scoping review. BMC Psychiatry. 2015;15(1):291.

Kaminski J. Diffusion of innovation theory. Can J Nurs Inform. 2011;6(2):1–6.

Bandura A, Walters RH. Social learning theory: Englewood cliffs Prentice Hall; 1977.

Bandura A. The evolution of social cognitive theory. In: Smith KG, Hitt MA, editors. Great Minds in Management. Oxford: Oxford University Press; 2019. p. 9–35.

Milburn K. A critical review of peer education with young people with special reference to sexual health. Health Educ Res. 1995;10(4):407–20.

Download references

Acknowledgements

Not applicable.

This research study is funded by the National Institute for Health and Care Research (NIHR) School for Public Health Research (project number SPHR PHPES025). The views and opinions expressed in the paper are those of the authors and do not necessarily reflect those of the NIHR. The funding body played no role in the design, analysis, interpretation or writing of the manuscript.

Author information

Steven Dodd and Emily Widnall are joint first authors.

Authors and Affiliations

Faculty of Health and Medicine, Lancaster University, Lancaster, UK

Steven Dodd & Mark Limmer

Population Health Sciences, University of Bristol, Bristol, UK

Emily Widnall & Judi Kidger

College of Medicine and Health, University of Exeter, Exeter, UK

Abigail Emma Russell

Faculty of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine, London, UK

Esther Louise Curtin

Mental Health Foundation, London, UK

Ruth Simmonds

You can also search for this author in PubMed   Google Scholar

Contributions

All authors contributed to the design of the systematic review. SD led on designing the search strategy with input from all co-authors. SD carried out the initial searches across four databases. SD and EW led on retrieving papers and screening abstracts and full papers. EW and SD led on data extraction with support from AR. SD and EW drafted the initial manuscript. All co-authors reviewed the manuscript and approved the final version.

Corresponding author

Correspondence to Emily Widnall .

Ethics declarations

Ethics approval and consent to participate, consent for publication, competing interests.

None to declare.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1., rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Dodd, S., Widnall, E., Russell, A.E. et al. School-based peer education interventions to improve health: a global systematic review of effectiveness. BMC Public Health 22 , 2247 (2022). https://doi.org/10.1186/s12889-022-14688-3

Download citation

Received : 09 August 2022

Accepted : 21 November 2022

Published : 02 December 2022

DOI : https://doi.org/10.1186/s12889-022-14688-3

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Peer education
• School-based interventions
• Systematic review
• School health
• Adolescents

BMC Public Health

ISSN: 1471-2458

• General enquiries: [email protected]