Abstract
Hispanic majority institutions have over 50% Hispanic student enrollment. We examined how social comparison may affect academic self-concept in this unique higher education landscape. We found that students who compare themselves to others who are worse off tend to experience a greater sense of academic self-concept in three domains than when they compare themselves to others who are better off. These findings have implications for improving academic success for students attending these emerging institutions.
In the presence of increasing diversity in higher education, understanding the factors that shape academic self-concept in students is paramount. Academic self-concept refers to a student’s perception of their own academic abilities and competence; it can affect important outcomes such as academic performance (Hermann et al., 2016) and career aspirations (Rudman & Phelan, 2010). One process that predicts academic self-concept is social comparison, in which students compare themselves to other students (Festinger, 1954; Jonkmann et al., 2012). However, most research that examines the association between academic self-concept and social comparison involves students at predominantly White institutions, leaving a gap in understanding in more diverse student populations (Bong & Skaalvik, 2003; Wang & Zhao, 2020).
The current study addresses this gap by examining these associations in students attending a Hispanic Majority Institution (HMI), where Hispanic students comprise more than 50% of the student population. HMIs present a novel social environment that shifts Hispanic students to majority status (Huynh, Wicks, et al., 2024), shifts White students to minority status, and shifts who the majority group is for students belonging to other minority groups on campus. These shifts in group status (i.e., one’s social identity; Tajfel & Turner, 1979, 2004) may lead to changes in how students compare themselves to others because the availability and proximity of targets of comparison are likely to be other Hispanic students as opposed to White students (Huynh et al., 2023). As such, these alterations in social comparison processes may also affect their academic self-concept. We assessed this supposition by contrasting social comparison types (i.e., four different approaches to social comparison based on comparison direction [upward or downward] and focus [similarities or differences]) on five kinds of academic self-concept (grade and effort, study habit/organizational self-perceptions, peer evaluation of academic ability, satisfaction with school, and self-evaluation with external standards). We did so while also controlling for students’ motivation for social comparison, general inclination to engage in social comparison (i.e., social comparison orientation), and social economic status. Below we provide a brief overview of these concepts and extend our discussion of this framework.
Academic Self-Concept
Academic self-concept is a multifaceted construct that encompasses a student’s beliefs about their intellectual competence, specific academic skills (e.g., study habits, test taking), overall potential for academic success, and satisfaction in their school environment (Reynolds, 1988). It also includes how students compare their own academic performance to relevant classmates (i.e., peer evaluation) and external benchmarks (i.e., self-evaluation to external standards (Bong & Skaalvik, 2003; Marsh, 2014; Reynolds, 1988; Shavelson et al., 1976).
Research suggests that a positive academic self-concept is not only associated with favorable outcomes but can also be influenced by a complex interplay of factors, including social and psychological dynamics, such as comparison processes (Trautwein & Möller, 2016). Students who believe in their academic abilities are more likely to engage in positive behaviors that promote success, such as setting goals and persevering through challenges (Omolade & Adeleye, 2021). In contrast, students with a negative academic self-concept may exhibit lower motivation and engagement (Prince & Nurius, 2014). This negativity can be further exacerbated by stress and social comparison with peers (Huh, 2015).
Among the diverse factors that affect academic self-concept (Trautwein & Möller, 2016), research shows that Hispanic students typically start college with lower self-reported academic self-concept than their peers (Nuñez, 2009). These differences are particularly evident in science, technology, engineering, and mathematics (STEM) disciplines (Mattern & Shaw, 2010). There are several reasons that might account for these academic self-concept discrepancies. For example, scholars have noted that Hispanic students may be vulnerable to lower academic self-concept because of their underrepresentation within the college/university context, the incongruency between expectations of their home cultures and those of the college/university setting, and their typically lower status within American society (Oseguera et al., 2009).
Additionally, Hispanic college students attending Hispanic-Serving Institutions (HSIs; with Hispanic student enrollment of 25% or more) and emerging HSIs (with Hispanic student enrollment of 15%–24%) tend to begin their college career with lower academic self-concept than their Hispanic peers at non-HSIs (Cuellar, 2014). This trend is particularly true for students at HSIs, as their self-reported academic self-concept tends to be the lowest. Such patterns may be explained, in part, by the fact that students who have higher academic self-concept self-select into non-HSIs institutions (Cuellar, 2014). However, notably, this group of students makes the largest gains in their academic self-concept, such that after 4 years of college, their academic self-concept scores tend to be on level with their peers at non-HSIs (Cuellar, 2014). A number of factors can account for this growth in academic self-concept, such as individual (e.g., race/ethnic self-identification, familial interdependence; Cuellar, 2014) and institutional (e.g., Hispanic representation; Cole, 2007) characteristics and personal experiences (e.g., positive cross race interactions; supportive faculty interactions; Cole, 2007; Nuñez, 2009). Researchers note that a critical factor may be Title V funds devoted to developing academic programs and advising to support Hispanic students’ success in college (Santiago et al., 2004).
Although this study focuses on academic self-concept for all university students at an HMI, we acknowledge that the development and nurturing of academic self-concept for Hispanic students, in particular, are rooted in and influenced by significant demographic shifts (Irizarry, 2015), cultural identities (Irizarry, 2007), and systemic and institutional structures that begin well before university life. For example, “accountability-focused” systems built on emphasizing high-stakes standardized testing disproportionately harm Latino students, particularly those from immigrant and low socioeconomic backgrounds (Valenzuela, 2004). In particular, as Valenzuela (2010) has documented, U.S. schools systematically devalue and erode the cultural identities and social capital of Mexican immigrant students, while displaying superficial care to the Latino education experience. These issues may critically influence students’ academic self-concept and are important to recognize, even though they are beyond the scope of the current work.
Together, the body of research suggests that academic self-concept differs based on the racial/ethnic composition of the student body (Cole, 2007). However, the research has yet to examine the academic self-concept of students attending HMIs, and the specific subcategories of academic self-concept (Reynolds, 1988). Therefore, the current study addresses that gap by examining five different types of academic self-concept for undergraduate students attending an HMI and offers insight on how academic self-concept may unfold based on social comparison processes.
Social Comparison
Social comparison refers to the process by which people compare themselves with others in order to determine their own abilities, attitudes, and opinions (Festinger, 1954). Social comparison is contingent upon the direction and focus of the comparison. Therefore, four types of social comparison are rooted in two dimensions. First, comparisons can vary based upon direction; individuals can compare themselves to someone better off (i.e., upward) or worse off (i.e., downward). Second, comparisons can vary based upon the focus; individuals can focus on similarities (i.e., identification) or differences (i.e., contrast) between themselves and the comparison target. Consequently, individuals will engage in one or more of these four social comparison combinations (termed “social comparison types” in this paper: upward identification, upward contrast, downward identification, and downward contrast; Buunk et al., 2005; Dijkstra et al., 2008; Smith, 2000).
These four types of social comparisons may be driven by different motives, which refer to the driving force or internal state that leads an individual to compare themselves to others (Corcoran et al., 2011; Helgeson & Mickelson, 1995). Festinger (1954) originally proposed that people’s main motivation for social comparison was for self-knowledge by treating their peers as a reference point against which to evaluate their own abilities. However, over the years, researchers have discovered that people also engage in social comparison for many other reasons. For example, people may be motivated to compare themselves to others to help them think about their future and anticipate self-relevant outcomes (e.g., future self; Tigges, 2009). They may also be motivated to compare themselves to others to make themselves feel better through self-enhancement (Caliskan et al., 2024; Wills, 1981). These comparisons, therefore, can lead to both desirable and undesirable outcomes. Upward comparisons with high-achieving peers can inspire students to strive for self-improvement but may also lead to feelings of inadequacy or envy that decrease academic self-concept. Therefore, students might avoid these negative emotions by making downward comparisons to boost self-esteem, but this approach may foster academic complacency and reduce motivation.
In addition to having different motives for social comparison, people may also differ in their social comparison orientation (SCO), which refers to one’s general propensity to make comparisons with others (Gibbons & Buunk, 1999). Individuals high in SCO are more likely to make comparisons with others, especially when they belong to the minority group (Régner & Monteil, 2007; Turner, 1975). This increased propensity to compare with others can predict decreased hardiness and life satisfaction for undergraduate students (Civitci & Civitci, 2015).
Although social comparison types for college students are context dependent, a review of the literature demonstrates that students generally prefer to compare themselves to others who are higher performing and are similar to them in related and unrelated ways (i.e., upward identification; Dijkstra et al., 2008). There is limited research on Hispanic students’ general preferred social comparison type, although research on assimilation suggests that they might prefer downward identification (Blanton et al., 2000). Research on Hispanic students enrolled at HMIs (Huynh, Wicks, et al., 2024) broadly supports the general trend found by Dijkstra et al.’s (2008) review, with the exception of social comparison related to academic achievement. Hispanic students at HMIs tended to prefer downward identification and upward contrast in this domain. These preferences appear to serve as a self-protective function, as downward identification was more closely associated with higher self-reported academic confidence (Huynh, Wicks, et al., 2024).
In addition to examining the types of social comparison Hispanic students at HMIs engage in, recent work has demonstrated that students who are high in social comparison orientation tend to report higher feelings of rejection/exclusion and self-doubt, and that these associations were explained by increased perceived stress (Huynh, Thomas, et al., 2024). That is, a higher propensity to self-compare predicted increased stress, which predicted increased feelings of rejection and self-doubt. Although Huynh, Thomas, et al.’s (2024) study presented an explanatory framework to illustrate how the propensity to self-compare may lead to rejection/exclusion and self-doubt, they did not address the other academic self-concept subdomains as they relate to social comparison types. Moreover, Huynh, Wicks, et al.’s (2024) study examined different social comparison types, but did not examine the potential associations with academic self-concept related to specific academic skills/effort (e.g., study habits, test taking), satisfaction in their school environment, and comparison of their own academic performance to relevant external standards. Therefore, this study fills those gaps by assessing the associations between social comparison types and academic self-concept. Importantly, given the previous works, we controlled for the potential influence of SCO and social comparison motives in the current study. We also accounted for students’ self-reported socioeconomic status (SES) because of its established effect on social comparisons (Régner et al., 2002).
The Current Study
Given that social comparison depends on the proximity and availability of comparison targets (Festinger, 1954), Hispanic majority institutions represent an important context for studying social and psychological processes that could affect academic success for students who attend these types of higher education institutions. At HMIs, Hispanic students are in the majority. This context could affect social comparison processes for Hispanic students because when they look for targets of social comparison, they are likely to encounter other Hispanic students, who are members of their ingroup. Conversely, this context could affect White students because, as they look for targets of social comparison, their targets are likely to be Hispanic students, who are members of their outgroup. And for other minority groups, Hispanic students also represent their most likely comparison target. Hispanic students are members of their outgroup, but different from the usual majority group (e.g., White students at predominantly White institutions). These suppositions are in line with social comparison theory (Festinger, 1954) and social identity theory (Tajfel & Turner, 1979, 2004) but have yet to be empirically examined within the context of HMIs. Additionally, previous work has shown that both social comparison and self-concept can be affected by the demographic composition of the university (Cole, 2007; Huynh, Thomas, et al., 2024, Huynh, Wicks, et al., 2024), but research has yet to examine the full spectrum of academic self-concept subdomains.
To address these gaps, this study examined potential differences in five types of academic self-concept based on types of social comparisons for students attending an HMI. We did so while accounting for the potential influence of students’ motivation for making these comparisons, their general propensity to self-compare (SCO), and SES. Given this novel educational setting, we had no specific hypotheses about how the types of academic self-concept would vary by social comparison types. Our goals were to simply explore potential differences, report on them, and provide post-hoc explanations to contextualize these findings within the literature.
Method
We conducted a cross-sectional survey to gather data relevant to addressing the study’s research questions. The online survey was distributed to students attending a Hispanic Majority Institution in South Texas. Participants completed a series of assessments and questionnaires designed to capture the study constructs.
Participants
Undergraduate students received a recruitment email via their university address; 405 students opened the survey link, but only 261 completed at least 2/3 of the study. Of the complete responses, 191 self-identified as Hispanic/Latino, 45 as White/Caucasian, 6 as Black/African American, 3 as Asian/Pacific Islander, 2 as Middle Eastern, 1 as Native American, and 1 as Native Hawaiian. Additionally, 41 participants reported that they were both White/Caucasian and Hispanic/Latino, and 23 participants reported that they were two or more races/ethnicities. Fifty-three participants self-identified as male, 198 as female, 7 as non-binary/third gender, and 3 selected “prefer not to say.” The average age was 27.42 years (SD = 9.93; range = 18–65). All participants were recruited from a public regional university in South Central Texas, with an enrollment of approximately 7,300 students. Since its inception, the university has always been a Hispanic Majority Institution. The university has two residence halls (up to 740 students in total).
Measures
We described the assessments and questionnaires pertinent to this manuscript below. We would like to note that the variables included in this report are part of a larger data collection effort. All measures for that effort are posted on the Open Science Framework (OSF: DOI 10.17605/OSF.IO/6UD3F). See Table 1 for descriptive statistics.
Descriptive Statistics for Main Study Variables.
Academic Self-Concept
To measure academic self-concept, we used five subscales from the academic self-concept scale (ASCS; Reynolds, 1988): Grade and effort, study habit/organizational self-perceptions, peer evaluation of academic ability, satisfaction with school, and self-evaluation with external standards. There are a few instruments available to assess academic self-concept, such as the Self-Description Questionnaire (Marsh, 1990) and the Multidimensional Self-Concept Scale (Bracken, 1992). However, the Academic Self-Concept Scale (ASCS; Reynolds, 1988) is the most widely employed instrument in education and psychology research. Its extensive use and validation across diverse populations and its specific focus on the academic domain makes it the most aligned with objectives of this study.
Participants used a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree) to rate their (dis)agreement with the scale items. Eight items were included to measure grade and effort dimension (i.e., “For me, studying hard pays off.”; “I do well in my courses given the amount of time I dedicate to studying.”). Six items were included to measure study habits/ organizational self-perception (i.e., “I feel I do not study enough before a test.”; “I do not study as much as I should.”). Eight items were included for the peer evaluation subscale (i.e., “Others view me as intelligent.”; “Others consider me a good student.”). Four items were included to measure satisfaction with school (i.e., “I enjoy doing my homework.”; “I usually feel on top of my work by finals week.”). Three items were included to assess self-evaluation with external standard dimensions (i.e., “It is hard for me to keep up with my classwork.”; “I feel teachers’ standards are too high for me.”).
Social Comparison Type
We measured social comparison types in two different ways. First, we wanted to capture students’ social comparison process based on their personal intuition about their peers, as they would naturally do. Therefore, we asked them to select an option that “best describes how you see yourself in comparison to your classmates at [name of university redacted for review] generally?,” with no other information provided. In the second approach, we wanted to standardize students’ social comparison targets so that everyone could compare themselves to the same set of standards. Therefore, we provided them with objective information about their peers. They saw: “At [name of university redacted for review], the student body can be described by the following statistics: 71% are first-generation college students; 73% identify as Hispanic; 86% receive financial aid; and the average high school GPA is 3.4. Select the item below that best describes how you see yourself in comparison to the student body at [name of university redacted for review].”
For both questions, students selected from one of four options corresponding to the common types of social comparison discussed in the introduction: (1) “Generally speaking, they are better off than me, but I am similar to them in important ways. [upward identification]”; (2) “Generally speaking, they are worse off than me, but I am similar to them in important ways. [downward identification]”; (3) “Generally speaking, they are better off than me, but I am different from them in important ways. [upward contrast]”; (4) “Generally speaking, they are worse off than me, but I am different from them in important ways. [downward contrast].” The corresponding type of social comparison is presented here in brackets for clarity, but was not shown to participants. This measure has been used elsewhere to assess HMIs (Huynh, Wicks, et al., 2024).
Social Comparison Orientation (SCO)
We measured SCO with the Scale of Social Comparison Orientation (Gibbons & Buunk, 1999). Participants used a 5-point Likert scale (1 = I disagree strongly, 5 = I agree strongly) to respond to statements such as: “I often compare myself with others with respect to what I have accomplished in life.”; “I always pay a lot of attention to how I do things compared with how others do things.” Higher scores are indicative of a higher propensity to self-compare.
Social Comparison Motives
Social comparison motivation was measured using the two subscales of Future Self and Self-enhancement of the Social Comparison Motives Scale (Tigges, 2009). We asked participants to think about the comparisons they make with other students at their university. We stated, “When you compare yourself to them, how often is this one of the reasons you are comparing? I compare myself. . .” Participants then used a 5-point Likert scale (1 = never, 5 = very often) to rate their agreement regarding the items. Six items were included on the sub-scale to measure future self (i.e., “To help me think about what I want.” and “To think about my future.”). Four items were included to measure self-enhancement (i.e., “To prove to myself I am very different from someone who is worse off than me.” and “To make myself feel better.”).
Demographics
Participants self-reported their age, sex, levels of education, major, and race/ethnicity. Socio-economic status was indicated by participants using the single-item version of the MacArthur Scale of Subjective Social Status-Adult Version (Singh-Manoux et al., 2005). Participants were shown a visual of a ladder with ten numbered rungs next to the statement, “Imagine a ladder that shows how your society is set up. At the top of the ladder are those who are best off, as in, have the most money, the highest amount of schooling, and jobs that bring the most respect. At the bottom are those who are worse off, as in, they have the least money, little to no education, no jobs or jobs that no one wants or respects. Now think of your family and please tell us where you think your family would be on this ladder by selecting the corresponding number below.” Response ranges from 1 to 10, with higher values indicating a higher subjective socio-economic status.
Procedures
Participants completed all procedures online using Qualtrics. We sent a recruitment email to all undergraduate students at [name of university, redacted for review]. The email explained the study and provided a Qualtrics link for those who volunteered to participate. After reading the information page and providing their consent, participants answered the questions about social comparison types first and then completed other measures, which were presented in random order to minimize order effects. Participants then provided demographic information and were presented with a debriefing statement. Participants were given the ability to skip questions to ensure the voluntary nature of the study. After completing the study, participants were provided with a separate web link through which they could give their email address to be entered into a drawing for one of eighteen $50 gift cards. All study materials and procedures were approved by the Institutional Review Board (IRB) at [redacted for review].
Data Analysis Plan
Our study’s primary predictor variable of social comparison type is categorical (upward identification, downward identification, upward contrast, downward contrast) and our outcome variables (five types of academic self-concept) are interval and related. Additionally, we wanted control for the potential influence of social comparison motivations, social comparison orientation, and SES. Given these parameters, we used multivariate analysis of covariance (MANCOVA) to test our hypotheses. MANCOVA allows us to decrease our chances of making a type I error compared to running several ANCOVA models from the outset. However, once we ran the main MANCOVA, we relied on individual ANCOVAs to determine where differences existed (if warranted by the MANCOVA). We ran two models, one with social comparison types based on participants’ personal intuition about their peers, and one with social comparison types based on standardized criteria about their peers as the independent variable. Dependent variables and covariates remained the same for both models. We recognize that we could have also used multiple regression by dummy coding our categorical independent variables, then running separate regression models for each dependent variable. However, we believe that the MANCOVA approach offers a more intuitive interpretation for our analyses focused on examining between group differences.
Prior to running our models, we assessed the assumptions of MANCOVA. Our independent variable is measured between participants, which supports the independence assumption. We used Cook’s distance as a proxy to examine potential univariate outliers; the largest value was .05, suggesting the absence of outliers. We then assessed the potential for multivariate outliers using Mahalanobis’ distance scores; none of the cases exceeded the critical threshold of χ2(4) = 18.47, indicating that there were no multivariate outliers. Moreover, the scatterplot matrix of the dependent variables and covariates suggests that the linearity assumption is met. Additionally, none of the correlations exceeded .90, suggesting that multicollinearity was not an issue. Also, none of the interaction terms between the independent variable and covariates were significant (ps > .28), suggesting that the assumption for homogeneity of regression slopes was met. However, the Shapiro-Wilk test was significant for all dependent variables and covariates, suggesting a violation of the normality assumption. Previous work shows that the family of MANOVA tests are robust to this assumption and outperforms the nonparametric alternative both in terms of power and minimizing type I errors (Finch, 2005). Additionally, Box’s Test of equality of covariance matrices was not significant at the conventional .001 level for this test (ps > .02), suggesting that this assumption is met. Lastly, we used Levene’s test to assess the homogeneity of variance assumption. All p values were higher than .08, suggesting that this assumption was not violated.
Results
Below, we reported outcomes from the MANOVA first, then individual ANCOVAs for each significant dependent variable. We reported the results from the first model involving participants’ social comparison based on personal intuition about their peers, and then the results from the second model involving participants’ social comparison based on the presented standardized criteria. For each model, we reported each type of academic self-concept (ASC) whenever there were differences between the four types of social comparison, and specified where those differences lay. Additionally, we noted any significant covariates in the model. Lastly, we summarized these findings by highlighting significant differences across the two models. See Figure 1 for a summary of the findings.
Differences in various self-concepts based on types of social comparison.
The MANCOVA shows a significant effect for academic self-concept when social comparison strategy was based on participants’ personal intuition about their peers, F(15, 644) = 2.16, p = .007, Wilk’s Lambda = .87,
Using Tukey’s post-hoc tests to unpack the significant effects, we found that for academic self-concept relating to grades and effort, downward identification (M = 4.33, SD = 0.69) had higher scores than upward contrast (M = 3.86, SD = 0.76), t = 2.70, p = .038, 95% CI for mean difference [0.02, 0.72]. Motivation for social comparison based on a desire to improve for the future was a significant covariate, F(1, 243) = 5.19, p = .02,
Academic self-concept relating to peer evaluations based on academic standards was higher for downward identification (M = 4.07, SD = 0.61) than for upward identification (M = 3.66, SD = 0.70), t = 2.89, p = .02, 95% CI for mean difference [0.04, 0.70] and upward contrast (M = 3.60, SD = 0.82), t = 3.24, p = .007, 95% CI for mean difference [0.09, 0.79]. Moreover, downward contrast (M = 4.16, SD = 0.71) was also higher than upward identification (M = 3.66, SD = 0.70), t = 3.10, p = .01, 95% CI for mean difference [0.09, 0.94], and upward contrast (M = 3.60, SD = 0.82), t = 3.42, p = .004, 95% CI for mean difference [0.14, 1.00]. In other words, the downward comparison type had higher scores than the upward comparison type, irrespective of whether it was contrast or identification, and there was no difference between the two upward comparison types. SES was a significant covariate, F(1, 243) = 4.23, p = .04,
Academic self-concept relating to self-evaluation with external standards was higher for downward identification (M = 3.72, SD = 0.81) than for upward contrast (M = 3.25, SD = 0.96), t = 2.72, p = .036, 95% CI for mean difference [0.02, 0.82]. Social comparison orientation was a significant covariate, F(1, 244) = 5.29, p = .02,
The second MANCOVA model with social comparison strategy based on standardized criteria was also significant, F(15, 649) = 2.09, p = .009, Wilk’s Lambda = .88,
Again, using post-hoc tests to unpack the significant outcomes, we found that academic self-concept pertaining to grades and effort was higher for downward identification (M = 4.32, SD = 0.56) than for upward identification (M = 3.93, SD = 0.78), t = 2.94, p = .019, 95% CI for mean difference [0.04, 0.65] and for upward contrast (M = 3.88, SD = 0.75), t = 2.94, p = .019, 95% CI for mean difference [0.05, 0.76]. There were no significant covariates in this model, all ps > .065.
Academic self-concept pertaining to peer evaluations with academic standards was higher for downward identification (M = 4.03, SD = 0.67) than for upward identification (M = 3.60, SD = 0.79), t = 3.41, p = .004, 95% CI for mean difference [0.10, 0.40]. SES significantly predicted this type of academic self-concept, F(1, 245) = 4.57, p = .03,
Academic self-concept pertaining to self-evaluations with external standards was higher for downward contrast (M = 3.79, SD = 0.98) than for upward identification (M = 3.22, SD = 0.87), t = 2.73, p = .03, 95% CI for mean difference [0.20, 0.51]. Downward identification (M = 3.66, SD = 0.80) was also higher than upward identification (M = 3.22, SD = 0.87), t = 3.17, p = .009, 95% CI for mean difference [−0.08, 0.77]. Social comparison orientation was the only significant covariate, F(1, 244) = 5.95 p = .02,
Summary of Results
When it comes to academic self-perceptions relating to grades and effort, downward identification resulted in higher scores than upward contrast in both models, and higher scores than upward identification in the second model. In reference to academic self-perceptions relating to academic standards, both downward comparison types resulted in higher scores than both upward comparison types in the first model; and downward identification resulted in higher scores than upward identification in the second model. Lastly, for self-evaluation with external standards, downward identification was higher than upward contrast in the first model; downward contrast and downward identification were both higher than upward identification in the second model. Overall, downward comparison types resulted in higher scores across the three academic self-perceptions than upward comparison types. Social comparison types did not result in different scores for academic self-perceptions relating to study habits and satisfaction with school. All of these relationships accounted for the potential effects of social comparison motives, social comparison orientation, and SES.
Discussion
The analyses revealed a clear tendency for students to report more positive academic self-concept when they engage in downward social comparison (e.g., comparing themselves to peers they perceive as less academically successful). This was evident across several dimensions of academic self-concept, including how students viewed their grades and effort, adherence to academic standards, and evaluation of their abilities compared to external benchmarks. We did not find differences in academic self-concept related to study habits and satisfaction with school.
The general affirmative trend of relying on downward comparisons supports research about the self-enhancement goals of social comparison (Caliskan et al., 2024; Wills, 1981). When students compare themselves to those they perceive as less successful, it can boost their self-esteem and create a sense of relative advantage. This can be particularly salient for students who may be struggling academically or experiencing self-doubt. By comparing themselves downward, they can create a more positive narrative about their own academic standing. These findings are particularly interesting to consider in the context of past research on self-concept for Hispanic students and Hispanic Serving Institutions, which indicate that students’ academic self-concept tended to be lower when Hispanic student enrollment is higher (i.e., highest in HSIs compared to emerging HSIs and non-HSIs; Cuellar, 2014). The persistent use of downward comparison strategies may reflect students’ response to address this deficit in self-concept. That is, perhaps HSIs and HMIs, by their composition, lead to a higher need to validate one’s performance and enhance one’s low academic self-concept by making more frequent downward comparisons.
However, it is important to note that this effect was not observed for aspects of academic self-concept related to study habits and satisfaction with school. That is, academic self-concept based on study habits and satisfaction with school did not differ based on the type of social comparison to which students typically engage. This suggests that although social comparison may influence how students perceive their academic abilities, it may not necessarily translate to changes in their actual behaviors or overall satisfaction with their academic experience. From the perspective of a college student, this makes intuitive sense. It is common to feel a sense of relief or reassurance when one sees that others are struggling more than oneself. This sentiment can serve as a temporary ego boost that helps people feel better about their own academic performance. However, it does not necessarily change their study habits or how they feel about school in general. As such, institutions that want to implement programs to foster academic self-concept related to study habits and satisfaction with school may want to explore factors other than social comparison. For example, they may want to explore factors such as intrinsic motivation, self-efficacy, and value beliefs (Trautwein & Möller, 2016).
Furthermore, our findings indicate that engagement in downward comparison strategies is related to higher self-evaluations for some aspects of academic self-concept. This finding is relatively alarming as there are noted downsides to downward comparisons. Although it could enhance one’s sense of accomplishment and competency temporarily, it could potentially make students complacent about how they set and pursue academic goals (Dijkstra et al., 2008). As such, it might be critical for researchers to identify more productive strategies to enhance students’ academic self-concept without the potential downsides that accompany downward comparison strategies. A more balanced approach to social comparison, utilizing both upward and downward comparisons strategically, is likely to be more beneficial for students in the long run.
Practical Implications
The findings of our study underscore the significant impact of social comparison on academic self-concept, particularly among students at HMIs. The tendency for downward comparison to boost self-concept highlights its potential as a tool for educators in higher education settings. Faculty can leverage this understanding by designing classroom activities and assignments that facilitate controlled social comparison experiences. For example, group projects that allow students to observe and learn from peers with varying skill levels can create opportunities for both upward and downward comparisons, fostering a more nuanced understanding of their own abilities.
Academic advisors and counselors can utilize these findings to inform their approach to student support. By recognizing the potential positive impact of downward comparison, they can guide students towards healthy and productive comparison types. This may involve helping students identify appropriate reference groups, encouraging them to focus on their own progress and growth, and providing resources for coping with the challenges of upward comparison.
Moreover, these findings highlight the importance of creating a supportive and inclusive campus climate that values diversity and celebrates individual achievements (e.g., Hurtado et al., 2015). By fostering a sense of belonging and promoting a growth mindset, institutions can mitigate the negative effects of social comparison and empower students to reach their full potential (Kolyda, 2023). Overall, by understanding the complex dynamics of social comparison and its impact on academic self-concept, higher education institutions can develop more effective strategies to support student success and well-being.
Limitations and Future Directions
Our study examined different dimensions of academic self-concepts, which provides a more nuanced view of this construct. However, this approach makes it more difficult to draw conclusions about students’ academic self-concept more generally. Future studies can address this gap for students attending HMIs by using both a multidimensional measure and a unidimensional measure of academic self-concept simultaneously. Additionally, the study’s findings should be interpreted judiciously, given that the data are cross-sectional, which limits the ability to make causal connections.
Importantly, we did not compare these findings across Hispanic and non-Hispanic students, given the small number of non-Hispanic students in our sample. It would be interesting to parse out the effects of social comparison based on different groups (e.g., Hispanic students, White students, and other minority groups). This approach would allow a full examination of how the unique context of HMIs could affect social comparison and academic self-concept. Future studies could address this gap by incorporating specified recruitment strategies and broadening their recruitment to include more HMIs in their sample.
Relatedly, our findings are based on the current study sample comprised of mostly Hispanic students at a Hispanic Majority Institution, which limits our ability to extend these findings to students at HSIs and non-HSIs. We speculate that these findings may be different in contexts where readily available targets of social comparison are not other minority students, such as those at predominantly White institutions (PWIs). However, past research has demonstrated that minority students may prefer other minority students as their peer group at PWIs (Blanton et al., 2000), so the larger composition of the university may not critically affect their social comparison tendencies as their peer group is their most readily available comparison targets. Moreover, we presume that the actual diversity composition of the student body (i.e., higher enrollment number of Hispanic students) may alter social comparison processes and self-concept. However, this assumption fails to incorporate the social nature of social comparison. For example, students might not see themselves as part of the majority because that evaluation requires a subjective self-categorization process as part of the formation of one’s social identity (Tajfel & Turner, 1979). Therefore, it is critical that future research incorporate students’ perceptions about their majority/minority status at various non-HSIs, HSIs, and HMIs.
Despite these limitations, our study still makes an important contribution to the literature, particularly given the scarcity of works focused on the experience of students attending Hispanic majority institutions. Our study provides a foundation from which future research can explore these important topics.
Conclusion
Our study examined differences in academic self-concept based on social comparison types for students attending a Hispanic majority institution. We found evidence to demonstrate that students generally had higher views about their grades, academic standards, and abilities when they engaged in downward social comparisons (both contrast and identification). These trends highlight how students from these institutions are potentially engaging in social comparison to address gaps in academic self-concept. This knowledge is critical for improving student success, particularly because downward social comparison has been noted to have negative drawbacks, such as reduced motivation for academic achievement. Future research can build on these findings to help address the social and psychological dynamics of Hispanic majority institutions to bolster success for traditionally underserved students.
Footnotes
Data Availability Statement
The data that support the findings of this study are available upon reasonable request from the corresponding author.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was supported by funding from the National Science Foundation Grant: 2222219.
