Worrying about what others think: A social-comparison concern intervention in small learning groups

Abstract

Small-group learning has become commonplace in education at all levels. While it has been shown to have many benefits, previous research has demonstrated that it may not always work to the advantage of every student. One potential problem is that less-prepared students may feel anxious about participating, for fear of looking “dumb” in front of their peers. This study examines the impact of an intervention to reduce that sort of anxiety—termed social-comparison concern—in small learning groups at the university level. Over the course of an academic quarter (10 weeks), 144 students in 33 small learning groups participated in the study. Approximately one-third received an intervention designed to reduce social-comparison concern by modifying theories of intelligence and attributions for failure. One-third received a study-skills intervention, and the remaining third received no extra resources. The findings suggest that the intervention was successful and that instructors may want to infuse small-group work with discussion of the malleable nature of intelligence and of the reasons for academic success and failure.

Keywords

academic anxiety collaborative learning diversity higher education small-group learning social comparison STEM education

Peer and small-group learning

From preschool to college, peer-to-peer learning techniques are used more and more widely as teachers and faculty members seek to engage students more actively in learning (Cabrera et al., 2002; Dreyfus, 2002; Felder and Brent, 2007; Van der Linden et al., 2002). There is a good deal of evidence in support of collaborative learning generally, particularly in the sciences, with benefits to students including better assessment scores, better attitudes toward learning, and greater persistence (Cabrera et al., 2002; Dinan and Frydrychowski, 1995; Springer et al., 1999). However, as Dillenbourg et al. (1996) have written, “Collaboration in itself is neither efficient [nor] inefficient. Collaboration works under some conditions, and it is the aim of research to determine the conditions under which collaborative learning is efficient” (p. 194).

The practice of collaborative peer learning is grounded in constructivist learning theories, which in essence hold that to learn most effectively, students need to construct their own knowledge. In other words, the more students engage actively in their own learning, the more successfully they will learn. Vygotsky, a central theorist in the constructivist school, stressed in particular the importance of the social environment in learning, asserting that learning is essentially a social act: one cannot learn meaningfully without some sort of person-to-person interaction taking place (Vygotsky, 1978). Peer-learning techniques are thought to work well, in part, because students benefit from the cognitive conflict that arises from realizing that others’ perspectives differ from one’s own. When this conflict occurs, individuals seek to resolve it, and this process leads to deeper understanding (Piaget, 1977; also see Springer et al., 1999). Small-group discussion may further promote deep understanding by allowing students to think through ideas carefully, view problems from different perspectives, and rework answers (Myers and Lamm, 1976). Other research has suggested that in a peer-learning situation, it is not just the student receiving the explanation, but also the student offering it, who benefits (Webb et al., 2002). Viewed from the perspective of schema theory (a theory helping to explain the ways in which people cognitively organize information), the act of explaining material to a peer should be expected to encourage cognitive restructuring, or building meaningful mental relationships among ideas, which engenders future learning and long-term retention of knowledge (Ausubel, 1963).

One version of small-group learning has received special attention in the literature on post-secondary education: peer-led, small-group, problem-based learning (Born et al., 2002; Drane et al., 2005; Lyle and Robinson, 2003; Tien et al., 2002). In this model, small groups of students meet regularly with a peer, one who has additional expertise in the subject matter, to work collaboratively on conceptually oriented problems. Studies of a number of peer-led group learning programs have shown improved examination performance and end-of-term grades and increased course retention among participants (Bonsangue and Drew, 1995; Born et al., 2002; Drane et al., 2014; Freeman, 1995; Gosser et al., 1996; Tien et al., 2002). Other advantages cited in the literature include improved relationships among peers, greater self-esteem, and improved attitudes toward the discipline (Springer et al., 1999). The use of peer leaders in this model is thought to further promote learning in peer-to-peer contexts. Peer leaders are not only close to student participants in terms of their own learning, enabling them to serve as effective tutors (Vygotsky, 1978), but they also serve as models, similar students who have succeeded in the course, thus providing a vicarious experience of success for students (Bandura, 1997).

Although there is a great deal of support for the benefits of peer learning, the literature demonstrates that under certain circumstances, particular kinds of students may not reap these benefits. For instance, Gijlers and De Jong (2005) found that in groups with students at extreme ends of the preparation scale (i.e. very well prepared and very unprepared students), less-prepared students retreated from the activity. Students with medium levels of preparation have also been shown to participate less actively and receive less help than their peers in heterogeneous groups (Webb, 1980; Webb and Kenderski, 1984), suggesting that they may be essentially left out of the conversation that develops between students with high levels and those with low levels of preparation (Webb, 1995). Demographic differences also appear to have an impact. For example, Wilkinson and Fung (2002) found that majority-group students tended to dominate discussion and activity in mixed-ethnicity groups. A similar pattern has been found with regard to gender, with males more active participants than females (Hazelwood et al., 1992; Lindow et al., 1985; Lockheed et al, 1983; Lockheed and Harris, 1984; Underwood et al., 2000; Webb, 1984).

Social-comparison concern

The notion of social-comparison concern stems from work in the 1950s by Festinger (1954), who theorized that individuals are driven to assess their own qualities, and that to do so, they make comparisons to similar others. The social-comparison literature describes two kinds of social comparison, upward (comparing oneself to perceived superior others) and downward (comparing oneself to perceived inferior others). In this study, we are concerned with upward comparisons, or comparisons which cause individuals to view themselves negatively in contrast to others.

Social comparison is often discussed in tandem with achievement goal theory, which is concerned with the different reasons behind learners’ motivations to work. Achievement goal theory holds that in learning situations, individuals’ behaviors are driven by distinct motivational orientations, and that the type of motivational orientation an individual adopts will influence the quality of that person’s learning (Ames, 1992; Dweck and Leggett, 1988). A long tradition of research related to achievement goal theory has demonstrated that people are motivated by either mastery goals (geared toward self-improvement) or performance goals (geared toward demonstrating one’s abilities to some audience) (Pintrich et al., 2003), with additional research showing that performance goals can be further divided into performance-approach (geared toward demonstrating ability) and performance-avoidance (geared toward avoiding demonstration of inability) goals (Elliot and Harackiewicz, 1996). Some, for example, Elliot and McGregor (2001), have proposed a further distinction between mastery-approach and mastery-avoidance goals, but that discussion is beyond the scope of this article.

Although social-comparison concern is a component of performance-goal orientation (Régner et al., 2007), in that a person with such an orientation is seeking to demonstrate their abilities, performance-goal orientation is a broader construct than is social-comparison concern. That is, performance-goal orientation is thought to encompass three more specific kinds of goals: outcome goals (e.g. get good grades), ability goals (e.g. confirm one’s intelligence), and normative goals, which involve social comparison (e.g. demonstrate that one is more skilled than others in the class) (Brophy, 2005; Grant and Dweck, 2003). In contrast, social-comparison concern involves strictly the attention an individual devotes to assessing their ability or performance relative to others.

In the learning realm, social-comparison concern can prompt or heighten students’ anxiety (Dijkstra et al., 2008). Several (e.g. Brophy, 2005; Mugny et al., 2001) posit that when individuals are faced with others whom they view as more competent at some task, the threat of feeling inferior hinders cognitive engagement, resulting in reduced ability to process information. Indeed, these authors found that when the competence threat disappears, individuals process information at deeper levels. Such effects may be stronger for people with strong performance (as opposed to mastery) goals (Dweck, 1986; Mugny et al., 2001), and performance-avoidance goals in particular (Elliot, 1999), in which the motivation behind performance is to avoid appearing incompetent. This process of increased cognitive load-reducing performance is similar to that believed to be at play in stereotype threat (Croizet et al., 2004; Schmader and Johns, 2003), a condition in which individuals facing the prospect of confirming some negative stereotype about their group perform more poorly than they would in the absence of such a threat (Steele, 1997).

In peer-learning situations, concern over one’s performance relative to others has been shown to have negative learning effects. Buchs et al. (2004) found that students were more concerned about their own competence levels when their partners were working on identical tasks, but less concerned when partners were working on complementary tasks (i.e. with less opportunity for comparison), and that this concern was detrimental to academic performance. Social-comparison concern in peer-learning settings has also been shown to be negatively related to student persistence and final course grade, as well as to self-efficacy (Micari and Drane, 2011). This effect may be particularly salient for less-prepared students and for students with lower levels of confidence. Kerr and Bruun (1983), for instance, found that in a group task for which there is a “right” approach or answer, less-prepared members can lose motivation, believing that their contributions would not be helpful, and Robinson-Staveley and Cooper (1990) found that individuals who had low expectations for success performed more poorly in a group setting than did individuals with high expectations, whereas there was no such difference among individuals working alone. Thus, the impact of social-comparison concern may account, at least in part, for the different levels of engagement among students noted in the literature (and described earlier in this article), for example, the lower levels of engagement among students with lower levels of preparation (Gijlers and De Jong, 2005).

Social-psychological interventions

Social-psychological interventions have successfully been used to improve outcomes for students considered at risk of low performance (Yeager and Walton, 2011). Rather than attempting to bolster students’ academic skills, these approaches aim to change the way students think about their own abilities, successes, and failures. Two related types of interventions with particular prominence in the literature are those designed to modify students’ theories of intelligence, encouraging a belief in intelligence as malleable (e.g. Blackwell et al., 2007), and those designed to train students to make external, unstable attributions for academic failure, shifting the reason for failure from intrinsic personal deficits to modifiable behaviors (e.g. Good et al., 2003; Wilson and Linville, 1982, 1985).

Interventions related to the nature of intelligence stem from the work of Dweck (1986), who posited that learners who hold an “entity” theory of intelligence, believing it to be something that one has only a fixed amount of, are less likely to persist through difficulty and succeed academically than those who hold a “incremental” theory of intelligence, believing it to be something that can grow with effort. A good deal of research (e.g. Blackwell et al., 2007; Mangels et al., 2006) has shown these patterns to hold in a range of situations. Interventions that teach students about the malleability of intelligence have been successful in changing student learning approach and academic performance, with young children (e.g. Kamins and Dweck, 1999), adolescents (e.g. Good et al., 2003), and college students (e.g. Aronson et al., 2002).

Interventions designed to reattribute failure to causes that can be changed are grounded in attribution theory (Weiner, 1986), which holds that people behave in large part based on the attributions they make for the outcomes of their actions. According to the theory, attributions people make fall along three dimensions: locus of causality (causes internal vs external to the person), stability (whether the cause can change in the future), and controllability (whether the cause can be controlled). Research shows that students who make either stable, internal attributions (e.g. “I did well because I am an intelligent person”) or internal, unstable, controllable attributions (e.g. “I did well because I worked really hard”) for their successes tend to have better outcomes than students who make other kinds of attributions (e.g. external: “I did well because my tutor helped me”). When it comes to failure, students who make unstable attributions (e.g. “I did poorly because I didn’t study enough”) tend to have better academic outcomes than those who make stable attributions (e.g. “I failed the test because I just don’t get algebra”) (Linnenbrink and Pintrich, 2002). Interventions which attempt to retrain students to make more adaptive attributions (e.g. internal attributions for success, unstable attributions for failure) have succeeded in improving motivation and academic outcomes (Haynes et al., 2008, 2009; Perry et al., 1993; Perry and Hall, 2009; Struthers and Perry, 1996; Wilson et al., 2002), as well as reducing academic anxiety and increasing feelings of hopefulness about academic performance (Hall et al., 2004; Martocchio, 1994; Ruthig et al., 2004).

The two dimensions addressed by these interventions—beliefs about intelligence and attributions for success/failure—are interrelated. Dweck (1999) and others have demonstrated that learners’ attributions for failure or success are dependent on their beliefs about the nature of intelligence. Those who believe that intelligence is fixed, for instance, will tend to attribute their academic failures to personal deficits, rather than some contextual factor. Thus, used together, the interventions should be expected to support each other: encouraging incremental (malleable) theories of intelligence should reinforce efforts to encourage unstable (changeable) and external (not personal-deficit) attributions for academic struggle.

A search of the literature reveals no attempts yet to use these kinds of interventions to reduce student social-comparison concern in small-group situations. This study is such an attempt, using a social–psychological intervention designed to reduce social-comparison concern in a small-group, peer-led learning environment, by helping students understand that they can improve their abilities (that intelligence is malleable) and that academic struggle does not necessarily indicate inability to succeed (attributing struggle to causes other than lack of ability).

Method

Setting

The study took place in a peer-led, small-group undergraduate learning program at a mid-sized, highly selective university in the Midwest region of the United States. This study took a quasi-experimental design approach to assess the differential impact of three approaches to facilitation on social-comparison concern.

The small-group learning program used in this study is associated with introductory-level science courses. However, the program is managed and run separately from the course, and students from more than one course section may be in the same small group. The focus of this research is to understand the levels of social-comparison concern within the small learning groups, rather than the social-comparison concern associated with specific courses.

Students participate in the program voluntarily, meeting weekly throughout the academic quarter (in this case 2010–2011), in groups of five to seven, along with a trained peer leader. Attendance is required, and students receive a transcript notation indicating their participation. In the 2-hour sessions, the groups work through problems related to a science course they are taking; these problems are fairly challenging and typically more conceptually oriented than what might appear on typical homework assignments. Peer leaders guide students through these problems, avoiding giving answers and encouraging students to work toward answers themselves and promoting group discussion of how to solve the problems. Students are assigned to workshop groups based on their own schedules.

Participants

In total, 195 undergraduate students in 33 learning groups from within the larger program took part in the study. Of them, 187 students took the Time 1 questionnaire; 150 took the Time 2 questionnaire, with 141 taking the questionnaire both times. (Sample size discrepancies between Times 1 and 2 are accounted for by students who had dropped the course or program, and several who were absent on the particular days when surveys were passed out.) Of the 141 students, 85 were women and 47 were men, with 11 not disclosing gender. In all, 102 were majority-group (white or Asian American) students, 19 minority group (either African American or Hispanic American), and 22 did not disclose ethnicity. (Note that in this study, both white and Asian American students are considered majority-group students because these groups are well represented in the sciences at the institution where the study took place. During the year in which the study took place, 2010, non-international white students made up approximately 43% and non-international Asian students approximately 40% of all students in the courses associated with the study.)

Procedure

Each student was part of a small learning group, and each learning group was assigned to one of three treatment groups. Treatment A group received information about malleability of intelligence and attributions for success/failure; Treatment B group received materials similar structurally to those in Treatment A group, but with different content; and Treatment C group received nothing beyond what would normally happen in a small study group. A total of 33 small groups participated, with 10 of the small groups (42 students) constituting the Treatment A groups, 8 small groups (38 students) Treatment B, and 15 small groups (61 students) Treatment C. The Treatment A groups and Treatment B groups comprised students working on introductory chemistry problems; the Treatment C group comprised students working on introductory biology problems. Although Treatment C involved using problems in biology, those problems had a very similar level of conceptual difficulty than those in chemistry, and the biology and chemistry groups had the same structure and format. Since the focus of this study is on group dynamics rather than on grade performance, we believe that having problems in two different disciplines is not a critical threat to validity in terms of assessing social-comparison concern within the peer-learning group.

Treatment A

Treatment A was designed to reduce students’ feelings of social-comparison concern. Two strategies were used in an effort to accomplish this goal:

Helping students understand the malleable nature of intelligence.

Helping students understand that others have struggled academically and succeeded, and that struggle does not necessarily indicate lack of ability.

In the Treatment A condition, students

Received handouts during the second week of the program, welcoming them and providing brief information on the malleability of intelligence, as well as a reminder that many students feel academically discouraged but go on to do well (this reminder included quotes from students describing that experience). Students briefly discussed the handouts with their peer leaders.

Engaged in an icebreaker activity about facing academic challenges, in which each group member, including the peer leader, shared an academic challenge they had faced.

Watched a video, midway through the term, of former program students talking about having struggled and lacking confidence initially in the program, and later doing well. Students discussed these afterward in groups with peer leaders.

Received weekly email messages from peer leaders briefly reviewing the week’s material and encouraging a mastery-oriented learning orientation. These emails offered praise for hard work, acknowledged the difficulty of the worksheet problems, and encouraged persistence.

Treatment B

Treatment B was used to test whether changes would result simply as a result of increasing communication between students and leaders. Thus, Treatment B was structurally similar to Treatment A but had different content. Students in the Treatment B condition

Received handouts in the second week introducing the program and providing suggestions on effective study techniques.

Engaged in a general icebreaker activity led by the peer leader.

Received weekly email messages from peer leaders briefly reviewing the week’s material and offering study suggestions.

Treatment C

Students in the Treatment C condition did not receive any additional materials or information but met weekly and engaged in the normal activities of a small-group, peer-led study program.

Treatment fidelity

Peer leaders in Treatment groups A and B were required to check in regularly with program staff to report on activities; all reported that activities took place as intended. These peer leaders were also coached on how to effectively send email messages, and their emails were reviewed regularly by the researcher to ensure consistent messages. Peer leaders in Treatment group C reported that they had not engaged in any of the activities designed for the other treatment groups.

Measures and instrumentation

Students in all three groups took a pre- and post-term questionnaire to assess social-comparison concern. The questionnaire contained six items measuring social-comparison concern, as well as other items unrelated to this study. The pre-term questionnaire was taken in week 2 of the 10-week term and the post-term questionnaire in week 9.

Survey items measuring social-comparison concern were developed by the researchers based on previous interviews with students in the small-group program and field tested on a small group of undergraduate students to establish face validity. We used factor analysis (maximum likelihood with direct oblimin rotation, an oblique method allowing for some correlation between factors) to identify underlying factors among the survey items. Table 1 includes a description of the items, factor loadings, and reliability values for these measures.

Table 1.

Instrument factors and loadings.

Factor	Item	Loading
Social-comparison concern in the group (α = 0.860)	I sometimes leave the workshop feeling like I am the only one who doesn’t understand the material well.	0.718
	I sometimes leave the workshop feeling like I am not as smart as others.	0.628
	I sometimes feel intimidated to participate in the group.	0.788
	I worry about getting things wrong in front of my group.	0.685
	I feel different from other people in this group.	0.675
	If I made a mistake in front of others in the group, I would feel embarrassed.	0.711

Students took the survey before and after the treatment; the Time-1 measure was used as a covariate in the analysis. The independent variable in this study is the type of intervention used, with three possible types being Treatment A, Treatment B, or Treatment C.

Hypothesis

It was expected that for students in the Treatment A condition, social-comparison concern would be reduced more than in the other two groups. More specifically, our hypothesis was that end-of-term mean social-comparison concern score of students in Treatment A would be lower than that of students in the Treatment B condition and of students in the Treatment C condition, controlling for their start-of-term scores.

Analysis and results

We performed all analyses in SPSS 20. We created a score for social-comparison concern within SPSS for Time 1 (beginning-of-term questionnaire) and Time 2 (end-of-term questionnaire).

We tested for pre-existing differences on the covariate (social-comparison concern at Time 1) between the three treatment groups using analysis of variance (ANOVA). Results suggest the three groups were comparable in terms of pre-existing levels of social-comparison concern (F(2, 184) = 0.588, p = 0.556). We also tested the assumption of regression homogeneity for α = 0.05. The results of the F test of the interaction term of treatment type and covariate strongly support the homogeneity assumption (F(2, 138) = 0.062, p = 0.940). This result supports the use of the covariance-based adjustment of the social-comparison scores across different treatment groups.

To test the hypothesis, analysis of covariance (ANCOVA) was performed for each of the Time 2 scores, using the Time 1 score as a covariate in the model. Data suggested that at least one of the three groups was marginally higher in social-comparison concern (F(2, 140) = 2.99, p = 0.054). We conducted follow-up contrasts to compare Treatment A to Treatment C and to Treatment B, in line with our hypothesis. Table 2 shows the results of the contrasts. The analysis revealed that the Treatment A group had a significantly lower value of social-comparison concern than the Treatment C group (p = 0.013). The effect size using Cohen’s d was 0.50, which as a general rule of thumb is considered a moderate effect (Cohen, 1988), and compares favorably to similar classroom-based interventions (e.g. Payton et al., 2008). No difference in average social comparison was found between Treatment A and Treatment B (p = 0.324), or between Treatment B and Treatment C (p = 0.229).

Table 2.

Difference in social-comparison concern scores (using adjusted means).

Comparison	Difference	p value
Treatment A versus Treatment C	0.486 (Treatment C–Treatment A)	0.013
Treatment A versus Treatment B	0.216 (Treatment B–Treatment A)	0.324
Treatment B versus Treatment C	0.267 (Treatment C–Treatment B)	0.229

Additional analyses

SAT/ACT level

Because previous research suggests that students with lower preparation levels may be especially susceptible to social-comparison concern, an additional analysis was carried out to compare the social-comparison concern scores of students with different levels of preparation. No single measure of prior academic achievement was available for all students, that is, grade point average (GPA) was available for some, SAT score for others, and ACT score for yet others. (SAT and ACT are both US standardized college admission tests; they are not acronyms but are the proper names of the tests.) Therefore, an SAT-to-ACT conversion score was used as a covariate for students lacking an ACT score, relying on the concordance table provided by ACT (http://www.act.org/aap/concordance/). While this is not a perfect measure, it was the best available for the data set. To compare students with higher and lower levels of preparation, we ran a median split in the ACT/SAT scores to produce high-prepared and low-prepared groups. We then ran two ANCOVAs on the Time 2 scores, again using the Time 1 score as a covariate, one for students in the high-prepared and one for students in the low-prepared group. For students in the high-prepared group (n = 60), there was no significant difference between Treatment A and Treatment C in average social-comparison concern (adjusted mean score, Treatment A = 2.825, adjusted mean score, Treatment C group = 3.150; F(1, 58) = 1.514, p = 0.224). For students in the low-prepared group (n = 44), there was a significant difference in social-comparison concern, with Treatment A score having a lower mean value than Treatment C score (adjusted mean score, Treatment A = 2.836, adjusted mean score, Treatment C = 3.523; F(1, 42) = 5.593, p = 0.023, Cohen’s d = 0.71) (see Figure 1).

Figure 1.

Social-comparison concern score by ACT/SAT level.

Summary of key findings

Students in Treatment A had significantly lower scores on the social-comparison concern measure than students in the Treatment C. This effect was greater for less-prepared students (as measured by ACT/SAT-conversion score). There were no differences in social-comparison concern between Treatment group A and Treatment group B or between Treatment group B and Treatment group C.

Discussion and conclusion

Students who work in small groups potentially benefit in numerous ways, but in some cases, certain students may not benefit and in fact may suffer from the anxiety triggered by having to perform in front of others. However, as this study demonstrates, this is not an inevitable situation. Faculty or program administrators can put relatively simple practices into place that reduce feelings of social-comparison anxiety for students who suffer from it, and which may very well benefit all students.

In this study, the primary intervention (Treatment A), sharing information on malleability of intelligence as well as the common experience of academic struggle, appeared to lead to lower levels of social-comparison concern. The larger effect for students with lower ACT/SAT-conversion scores is encouraging in light of research showing greater negative impact of social-comparison concern for less-prepared or less-confident students (e.g. Kerr and Bruun, 1983; Robinson-Staveley and Cooper, 1990). It is precisely these students who can benefit most from reduced anxiety in the small-group setting.

Interventions such as the one used in this study can be woven into the everyday practices of programs, so that they do not feel like “add-ons.” For instance, materials on malleability of intelligence and academic struggle can be handed out along with program welcome materials, and these ideas can be reinforced in sessions by teachers or facilitators. In peer-led programs, peer leaders can be trained on helping students reshape their understanding of intelligence and attribution style; there is no lack of accessible literature on both of these topics (e.g. Assessing Women in Engineering (AWE) Project, 2005; Rae-Dupree, 2008). Peer leaders and veteran students can also easily share personal stories in the small groups of their own academic challenges and how they overcame these.

One limitation of this study is its particular sample, which came from a subset of motivated students at a highly selective institution. That is, because students who register for the small-group program do so voluntarily, they are likely to be more academically motivated than the student body at large, and given that that larger student body is a highly well-prepared one, the results of this study may not generalize to more academically diverse student groups or institutions. Another limitation has to do with the composition of the three groups: because of logistical constraints, the Treatment C group students were working on biology material, while the other treatments group students were working on chemistry material. While we believe that this did not threaten study validity, since the structure and methods are identical across groups, future studies should use groups working on identical material, to avoid any possible confounding effect of the particular problems students are working on. A further limitation was that the intervention was run with peer leaders as the intermediaries so that the researchers trained the peer leaders, who communicated information to the students. While the researchers did receive regular feedback from peer leaders on all of these communications, and while in almost all cases the peer leaders reported that they felt the communications were successful, it is possible that some peer leaders did not feel comfortable talking about malleability of intelligence or academic challenges. If this was the case, the messages may not have been effectively transmitted. Future studies should consider either providing more extensive training to peer leaders or using a more highly trained person to communicate key information to students. More broadly, there is much room for research into the particular group practices and structures (e.g. in terms of size, composition, etc.) that may have an impact on students’ social-comparison concern. In particular, future qualitative research offering in-depth analysis of the communication behaviors and environmental characteristics that prompt students to feel social-comparison concern would contribute greatly to existing literature.

The advantages of small-group learning are so well documented and publicized, and the format so widely used, that it is probable that in some cases faculty or program administrators implement it without giving enough thought to how it might impact different students differently. For students who are worried about performing poorly in front of the group, or who feel that they are not as intelligent or talented as others in the group, small-group environments might actually do more harm than good. Incorporating simple practices such as those used in this study into small-group programs—or into small-group work within the larger classroom—may help those students, and most likely others as well.

Footnotes

Funding

The program in which the research took place was originally funded through the Andrew W. Mellon Foundation.

Author biographies

Marina Micari is an associate director at the Searle Center for Advancing Learning and Teaching at Northwestern University. Her research interests center on the undergraduate academic experience, including undergraduate–faculty interaction, underrepresented students in higher education, and undergraduate learning communities. Address: Northwestern University, 627 Dartmouth Place, Evanston, IL 60208, USA. [email: m-micari@northwestern.edu]

Pilar Pazos is an associate professor at the Engineering Management and Systems Engineering Department at Old Dominion University. Her research interests center on teams, collaborative learning, computer-supported collaboration, knowledge transfer, and engineering education. Address: Old Dominion University, 2101D Systems Academic and Research Building, Norfolk, VA 23529, USA. [email: MPazosLa@odu.edu]

References

Ames

(1992) Classrooms: Goals, structures, and student motivation. Journal of Educational Psychology 84(3): 261–71.

Aronson

Fried

Good

(2002) Reducing stereotype threat and boosting academic achievement of African-American students: The role of conceptions of intelligence. Journal of Experimental Social Psychology 38(2): 113–25.

Assessing Women in Engineering (AWE) Project (2005) Attribution theory. AWE research overviews. Available at: http://www.aweonline.org; http://www.engr.psu.edu/awe/misc/ARPs/AttributionWeb-_03_22_05.pdf (accessed 3 May 2013).

Ausubel

(1963) Cognitive structure and the facilitation of meaningful verbal learning. Journal of Teacher Education 14(2): 217–22.

Bandura

(1997) Self-Efficacy: The Exercise of Treatment C. New York: W.H. Freeman Publishers.

Blackwell

Trzesniewski

Dweck

(2007) Implicit theories of intelligence predict achievement across an adolescent transition: A longitudinal study and an intervention. Child Development 78(1): 246–63.

Bonsangue

Drew

(1995) Increasing minority students’ success in calculus. In: Gainen

Willemsen

(eds) Fostering Student Success in Quantitative Gateway Courses (New Directions for Teaching and Learning, no. 61). San Francisco, CA: Jossey-Bass, pp. 23–33.

Born

Revelle

Pinto

(2002) Improving biology performance with workshop groups. Journal of Science Education and Technology 11(4): 347–65.

Brophy

(2005) Goal theorists should move on from performance goals. Educational Psychologist 40: 167–176.

10.

Buchs

Butera

Mugny

(2004) Resource interdependence, student interactions and performance in cooperative learning. Educational Psychology 24: 291–314.

11.

Cabrera

Nora

Crissman

. (2002) Collaborative learning: Its impact on college students’ development and diversity. Journal of College Student Development 43(1): 20–34.

12.

Cohen

(1988) Statistical Power for the Behavioral Sciences, 2nd edn. Hillsdale, NJ: Lawrence Erlbaum Associates.

13.

Croizet

Després

Gauzins

. (2004) Stereotype threat undermines intellectual performance by triggering a disruptive mental load. Personality and Social Psychology Bulletin 30(6): 721–31.

14.

Dillenbourg

Baker

Blaye

. (1996) The evolution of research on collaborative learning. In: Spada

Reiman

(eds) Learning in Humans and Machine: Towards an Interdisciplinary Learning Science. Oxford: Elsevier, pp. 189–211.

15.

Dinan

Frydrychowski

(1995) A team-based learning method for organic chemistry. Journal of Chemistry Education 72(5): 429–31.

16.

Dijkstra

Kuyper

Van der Werf

. (2008) Social comparison in the classroom: A review. Review of Educational Research 78(4): 828–79.

17.

Drane

Smith

Light

. (2005) The gateway science workshop program: Enhancing student performance and retention in the sciences through peer facilitated discussion. Journal of Science Education and Technology (3): 227–352.

18.

Drane

Micari

Light

(2014) Students as teachers: Effectiveness of a peer-led STEM learning program over 10 years. Educational Research and Evaluation. DOI: 10.1080/13803611.2014.895388.

19.

Dreyfus

(2002) How are we doing? Steady growth in implementing peer-led team learning. Progressions: Peer-Led Team Learning 3(3–4): 1–5.

20.

Dweck

(1986) Motivational processes affecting learning. American Psychologist 41(10): 1040–8.

21.

Dweck

(1999) Self-Theories: Their Role in Motivation, Personality, and Development. Philadelphia, PA: Psychology Press.

22.

Dweck

Leggett

(1988) A social–cognitive approach to motivation and personality. Psychological Review 95(2): 256–73.

23.

Elliot

(1999) Approach and avoidance motivation and achievement goals. Educational Psychologist 34(3): 169–89.

24.

Elliot

Harackiewicz

(1996) Approach and avoidance achievement goals and intrinsic motivation: A mediational analysis. Journal of Personality and Social Psychology 70(3): 461–75.

25.

Elliot

McGregor

(2001) A 2 × 2 achievement goal framework. Journal of Personality and Social Psychology 80(3): 501–19.

26.

Felder

Brent

(2007) Chapter 4: Cooperative learning. In: Mabrouk

(ed.) Active Learning: Models from the Analytical Sciences (ACS Symposium Series, 970). Washington, DC: American Chemical Society, pp. 34–53.

27.

Festinger

(1954) A theory of social comparison processes. Human Relations 7(2): 117–40.

28.

Freeman

(1995) MathExcel: A special opportunity in calculus. Unpublished report, Department of Mathematics, University of Kentucky, Lexington, KY.

29.

Gijlers

De Jong

(2005) The relation between prior knowledge and students’ collaborative discovery learning processes. Journal of Research in Science Teaching 42(3): 264–82.

30.

Good

Aronson

Inzlicht

(2003) Improving adolescents’ standardized test performance: An intervention to reduce the effects of stereotype threat. Journal of Applied Developmental Psychology 24(6): 645–62.

31.

Gosser

Roth

Gafney

. (1996) Workshop chemistry: Overcoming the barriers to student success. The Chemical Educator [online]1(1): 2–16.

32.

Gosser

Roth

(1998) The workshop chemistry project: Peer-led team learning. Journal of Chemical Education 75(2): 185–7.

33.

Grant

Dweck

(2003) Clarifying achievement goals and their impact. Journal of Personality and Social Psychology 85: 541–53.

34.

Hall

Hladkyj

Perry

. (2004) The role of attributional retraining and elaborative learning in college students’ academic development. Journal of Social Psychology 144(6): 591–612.

35.

Haynes

Daniels

Stupnisky

. (2008) The effect of attributional retraining on mastery and performance motivation among first-year college students. Basic and Applied Social Psychology 30(3): 198–207.

36.

Haynes

Daniels

Stupnisky

. (2009) A review of attributional retraining treatments: Fostering engagement and persistence in vulnerable college students. In: Smart

(ed.) Higher Education: Handbook of Theory and Research, vol. 24. Dordrecht: Springer, pp. 227–72.

37.

Hazelwood

Roth

Hasbach

. (1992) Gender and discourse: The unfolding “living text” of a science lesson (Tech. Rep. No. 60). East Lansing: Center for the Learning and Teaching of Elementary Subjects, Michigan State University.

38.

Kamins

Dweck

(1999) Person vs. process praise and criticism: Implications for contingent self-worth and coping. Developmental Psychology 35: 835–47.

39.

Kerr

Bruun

(1983) Dispensability of member effort and group motivation losses: Freerider effects. Journal of Personality and Social Psychology 44(1): 78–94.

40.

Lindow

Wilkinson

Peterson

(1985) Antecedents and consequences of verbal disagreements during small-group learning. Journal of Educational Psychology 77(6): 658–67.

41.

Linnenbrink

Pintrich

(2002) Motivation as an enabler for academic success. School Psychology Review 31(3): 313–27.

42.

Lockheed

Harris

(1984) Cross-sex collaborative learning in elementary classrooms. American Educational Research Journal 21: 275–94.

43.

Lockheed

Harris

Nemcef

(1983) Sex and social influence: Does sex function as a status characteristic in mixed-sex groups of children? Journal of Educational Psychology 75(6): 877–88.

44.

Lyle

Robinson

(2003) A statistical evaluation: Peer-led team learning in an organic chemistry course. Journal of Chemical Education 80(2): 132–4.

45.

Mangels

Butterfield

Lamb

. (2006) Why do beliefs about intelligence influence learning success? A social cognitive neuroscience model. Social Cognitive and Affective Neuroscience 1(2): 75–86.

46.

Martocchio

(1994) Effects of conceptions of ability on anxiety, self-efficacy, and learning in training. Journal of Applied Psychology 79(6): 819–25.

47.

Micari

Drane

(2011) Intimidation in small learning groups: The roles of social-comparison concern, comfort, and individual characteristics in student academic outcomes. Active Learning in Higher Education 12(3): 175–87.

48.

Mugny

Butera

Falomir

(2001) Social influence and threat in social comparison between self and source’s competence: Relational factors affecting the transmission of knowledge. In: Butera

Mugny

(eds) Social Influence in Social Reality. Seattle, WA; Göttingen: Hogrefe & Huber, pp. 225–46.

49.

Myers

Lamm

(1976) The group polarization phenomenon. Psychological Bulletin 83: 602–27.

50.

Payton

Weissberg

Durlak

. (2008) Positive Impact of Social and Emotional Learning for Kindergarten to Eighth-Grade Students: Findings from Three Scientific Reviews. Chicago, IL: Collaborative for Academic, Social, and Emotional Learning (CASEL).

51.

Perry

Hall

(2009) Attributional retraining in academic settings. In: Anderman

Anderson

(eds) The Psychology of Classroom Learning: An Encyclopedia. Montvale, NJ: Thomson Gale, pp. 73–6.

52.

Perry

Hechter

Menec

. (1993) Enhancing achievement motivation and performance in college students: An attributional retraining perspective. Research in Higher Education 34(6): 687–720.

53.

Piaget

(1977) The Development of Thought: Equilibration of Cognitive Structures. New York: Viking Press.

54.

Pintrich

Conley

AMM

Kempler

(2003) Current issues in achievement goal theory and research. International Journal of Educational Research 39(4–5): 319–37.

55.

Rae-Dupree

(2008) If you’re open to growth, you tend to grow. The New York Times, 6 July. Available at: http://www.nytimes.com/2008/07/06/business/06unbox.html?_r=0

56.

Régner

Escribe

Dupeyrat

(2007) Evidence of social comparison in mastery goals in natural academic settings. Journal of Educational Psychology 99(3): 575–83.

57.

Robinson-Staveley

Cooper

(1990) Mere presence, gender, and reactions to computers: Studying human-computer interaction in the social context. Journal of Experimental Social Psychology 26(2): 168–83.

58.

Ruthig

Perry

Hall

. (2004) Optimism and attributional retraining: Longitudinal effects on academic achievement, test anxiety, and voluntary course withdrawal. Journal of Applied Social Psychology 34(4): 709–30.

59.

Schmader

Johns

(2003) Converging evidence that stereotype threat reduces working memory capacity. Journal of Personality and Social Psychology 85(3): 440–52.

60.

Springer

Stanne

Donovan

(1999) Effects of small-group learning on undergraduates in science, mathematics, engineering, and technology: A meta-analysis. Review of Educational Research 69(1): 21–51.

61.

Steele

(1997) A threat in the air: How stereotypes shape intellectual identity and performance. American Psychologist 52(6): 613–29.

62.

Struthers

Perry

(1996) Attributional style, attributional retraining, and inoculation against motivational deficits. Social Psychology of Education 1: 171–187.

63.

Tien

Roth

Kampmeier

(2002) Implementation of a peer-led learning instructional approach in an undergraduate organic chemistry course. Journal of Research in Science Teaching 39(7): 606–32.

64.

Underwood

Wood

(2000) When does gender matter? Interactions during computer based problem solving. Learning and Instruction 10(5): 447–62.

65.

Van der Linden

Erkens

Schmidt

. (2002) Chapter 3: Collaborative learning. In: Simons

Linden

Du’ffy

(eds) New Learning. New York: Kluwer Academic Publishers, pp. 37–54.

66.

Vygotsky

(1978) Mind in Society: The Development of Higher Psychological Processes. Cambridge, MA: Harvard University Press.

67.

Webb

(1980) An analysis of group interaction and mathematical errors in heterogeneous ability groups. British Journal of Educational Psychology 50(3): 266–76.

68.

Webb

(1984) Sex differences in interaction and achievement in cooperative small groups. Journal of Educational Psychology 76(1): 33–4.

69.

Webb

(1995) Group collaboration in assessment: Multiple objectives, processes, and outcomes. Educational Evaluation and Policy Analysis 17(2): 239–61.

70.

Webb

Kenderski

(1984) Student interaction and learning in small group and whole class settings. In: Peterson

(eds) The Social Context of Instruction: Group Organization and Group Processes. New York: Academic Press, pp. 153–70.

71.

Webb

Farivar

Mastergeorge

(2002) Productive helping in collaborative groups. Theory into Practice 41(1): 13–20.

72.

Weiner

(1986) An Attributional Theory of Motivation and Emotion. New York: Springer Verlag.

73.

Wilkinson

IAG

Fung

IYY

(2002) Small-group composition and peer effects. International Journal of Educational Research 37(5): 425–47.

74.

Wilson

Linville

(1982) Improving the academic performance of college freshmen: Attribution therapy revisited. Journal of Personality and Social Psychology 42(2): 367–76.

75.

Wilson

Linville

(1985) Improving the performance of college freshmen with attributional techniques. Journal of Personality and Social Psychology 49(1): 287–93.

76.

Wilson

Damiani

Shelton

(2002) Improving the academic performance of college students with brief attributional treatments. In: Aronson

(ed.) Improving Academic Achievement: Impact of Psychological Factors on Education. San Diego, CA: Academic Press, pp. 89–108.

77.

Yeager

Walton

(2011) Social-psychological interventions in education: They’re not magic. Review of Educational Research 81: 267–301.