Empathy Towards Muslim Study
Emotional Mobilization and Attitude Formation Toward Muslims at the University of Michigan: A 2024 survey of 712 undergraduates reveals that empathy toward Muslims is often taught through campus narratives rather than lived experience.
Principal Investigator Kevin M. Waldman
psychFORM Research Lab
Abstract
Between January and May 2024, a cross-sectional survey of 712 University of Michigan undergraduates examined how emotion and peer learning shape perceptions of Muslim Americans in the United States. The 23-item, paired-balance instrument assessed learning source, emotional engagement, and perceived fairness toward Muslims.
72 percent of respondents reported that their understanding of Muslim experiences was learned primarily after entering college, chiefly through coursework and campus dialogue rather than family or pre-college media.
Among male students, 91 percent perceived cultural tension between Islam and Western society, compared with 61 percent of female students. Yet 84 percent of females - and only 11 percent of males - believed Muslims face widespread unfair treatment in the U.S. Emotional engagement during learning was the strongest predictor of perceived discrimination (β ≈ .45, p < .001), followed by gender (β ≈ .40, p < .001).
Open-ended responses highlighted emotionally vivid peer and classroom experiences - particularly personal stories shared by Muslim classmates or instructors - as central to belief formation.
These findings indicate that college represents a primary context for moral-emotional socialization: students internalize narratives of fairness and inclusion through affective as well as cognitive channels. The study extends moral-emotion and social-learning theories to higher-education contexts and outlines implications for balancing empathy-based pedagogy with critical reasoning. Most University of Michigan undergraduates reported learning about Muslim life and identity in the United States primarily through college instruction and campus dialogue, not through personal contact before entering university. Their empathy toward Muslim Americans, therefore, reflects emotion that was taught and modeled in academic settings rather than emotion derived from lived interpersonal experience.
Keywords: emotional learning, moral reasoning, empathy, higher education, Muslim Americans, institutional empathy, social learning theory, University of Michigan
Part 1. Introduction 1.1 The Role of Universities in Moral and Cultural Learning
Universities serve as key contexts in which emerging adults form durable moral and cultural beliefs. The developmental period between ages 18 and 24 is characterized by heightened openness to new ideas and exploration of multiple social identities (Arnett, 2000). Within that period, students learn not only academic content but also norms of empathy, fairness, and inclusion - often through emotionally charged classroom experiences. As higher education expands its commitment to diversity and intergroup understanding, university settings have become focal sites for moral-emotional learning.
The University of Michigan provides an especially fertile environment for studying these processes. It houses nationally recognized programs in intergroup relations, multicultural education, and diversity dialogue, many of which explicitly encourage students to examine moral and cultural difference through personal narrative. These pedagogical approaches aim to cultivate empathy for marginalized groups, yet little empirical work has quantified how such experiences translate into measurable beliefs. The present research addresses that gap by examining how emotional learning and peer interaction influence undergraduates’ perceptions of Muslim Americans in contemporary U.S. society.
1.2 Conceptual Framework: Emotion as a Moral Instructor
Moral-emotion theory posits that affective responses are primary in the moral-reasoning process (Haidt, 2001; Greene et al., 2004). Intuitive feelings - compassion, guilt, moral elevation - precede deliberate reasoning and strongly predict judgments about fairness. When educational settings intentionally evoke such emotions through personal testimony or stories of injustice, they can accelerate moral alignment but also introduce potential bias: emotion can overshadow cognitive evaluation (Bloom, 2016). Bandura’s (1977) social-learning theory complements this view by emphasizing that observation and reinforcement shape behavior and belief. In classrooms and peer discussions, the models are professors and classmates; the reinforcement is social approval for empathic or inclusive responses.
Combining these frameworks, the present study treats emotional learning as both a cognitive and social phenomenon. Students internalize moral narratives not only through what they hear, but also through the affective tone with which such information is presented and received. These emotional cues serve as heuristics for moral judgment - shortcuts that simplify complex sociocultural issues into intuitive categories of victimhood and fairness.
1.3 Empirical Background
Research on intergroup contact demonstrates that personal interaction across group lines can reduce prejudice under optimal conditions of equality and cooperation (Allport, 1954; Pettigrew & Tropp, 2008). Yet contact that is emotionally intense or framed around confession can also create ambivalence: individuals may experience compassion for the out-group alongside anxiety or self-consciousness (Nadler & Liviatan, 2006). Studies of intergroup dialogue programs on U.S. campuses show robust gains in empathic concern but mixed effects on cognitive complexity (Nagda et al., 2009; Rockenbach et al., 2015).
Despite extensive scholarship on diversity learning, few studies have isolated the sources from which students believe they learned about minority experiences. Most assume that attitudes observed in college stem from earlier socialization, family, or media exposure. The current investigation challenges that assumption by documenting that the majority of University of Michigan students attribute their knowledge of Muslim experiences specifically to their time in college.
1.4 Descriptive Findings as Context
Between January and May 2024, 712 undergraduates at the University of Michigan completed a paired-balance survey examining learning sources, emotional engagement, and perceptions of fairness toward Muslims in America. Results revealed striking patterns:
These descriptive figures point to two interconnected processes. First, universities function as major venues for exposure: most students encounter sustained discussion of Muslim identity only after matriculation. Second, gender differences suggest divergent emotional pathways. Female students report stronger empathic identification, while males register higher awareness of cultural difference. The combination implies that emotion and identity interact to shape moral reasoning.
1.5 Emotional Learning on Campus
Emotionally oriented pedagogy is common in diversity education. Intergroup dialogue courses often include structured storytelling, perspective-taking exercises, and reflective writing, all designed to elicit empathy (Nagda & Zúñiga, 2003). Such methods can powerfully alter perceptions, but their psychological mechanism remains under-specified. Within the Michigan sample, open-ended responses revealed that students frequently cited “personal stories shared by classmates or instructors” as the moments that most influenced their views. These experiences align with what Decety and Lamm (2006) describe as affective resonance - a process in which exposure to another’s emotion generates a shared visceral response that anchors later cognitive evaluations.
The emotional context of learning therefore matters as much as the informational content. When peers or instructors describe discrimination in ways that evoke sympathy, listeners are likely to encode those narratives as personal moral truths. While this process can enhance compassion, it also risks producing polarized evaluations in which one group is consistently framed as victim and another as privileged. Understanding this mechanism empirically is essential for designing balanced diversity curricula.
1.6 Gender Differences in Moral-Emotional Processing
Gendered socialization provides one plausible explanation for the divergent results. Numerous studies show that women report higher empathic concern and emotional contagion than men (Eisenberg & Spinrad, 2014). Women tend to use emotion as a moral cue, whereas men are more likely to rely on abstract principles or social norms (Gilligan, 1982). In the Michigan dataset, female participants’ higher endorsement of “unfair treatment” beliefs may reflect greater emotional engagement with stories of discrimination encountered on campus. Male participants’ greater perception of cultural tension may correspond to a cognitive appraisal of difference rather than an affective alignment. Both responses are products of social-learning environments but manifest through distinct psychological pathways.
1.7 Integrating Learning Source and Emotional Mediation
The intersection of where knowledge is acquired and how it is emotionally processed provides a novel angle on campus moral development. If 72 percent of students trace their beliefs about Muslim Americans to college exposure, then emotional learning at that stage becomes the primary driver of perception. The emotional mediator - measured here through self-reports of intensity and empathy - connects exposure to outcome. This conceptual model parallels mediated-path analyses in moral psychology, where emotion explains the relationship between moral framing and policy attitudes (Clifford, 2019).
By focusing on emotion as mediator rather than byproduct, the study moves beyond simple attitude measurement to explore causal mechanisms. It also situates higher education within the broader literature on cultural moralization: the process by which societies teach citizens to interpret identity categories through moral emotion rather than neutral description.
1.8 Methodological Innovation: The Paired-Balance Design
Standard attitude scales often force unidirectional responses - agreement with statements such as “Group X faces discrimination.” Such items risk conflating empathy with endorsement. The paired-balance method used in this study presents each concept in both positive and negative valence (e.g., “I sense cultural tension surrounding Islam’s place in Western society” / “Islamic and Western values can coexist without major conflict”). This design allows respondents to express simultaneous agreement with partially opposing statements, thereby capturing ambivalence. Analytically, it enables differentiation between affective and cognitive components of attitude, an important advancement for studying moral complexity in educational settings.
1.9 Research Questions and Hypotheses
Building on these foundations, the investigation addressed four main questions:
Corresponding hypotheses were as follows:
1.10 Significance and Contribution
This research contributes to three strands of scholarship. First, it empirically tests moral-emotion theory within an educational context, demonstrating how affective cues in classroom and peer settings translate into social judgments. Second, it extends social-learning theory by quantifying learning sources and identifying college as a dominant stage for moral acquisition. Third, it refines methodology for studying ambivalence through paired items, offering a replicable model for other domains of intergroup perception.
Beyond theoretical implications, the findings carry practical relevance for educators and administrators. If emotional framing drives moral perception, then university programs should incorporate reflective analysis that encourages students to contextualize feelings within evidence. Structured debriefings, mixed-method instruction, and explicit discussion of emotional influence could foster critical empathy - an understanding that combines compassion with evaluative rigor.
1.11 Overview of the Present Study
To explore these questions, the current study used a mixed quantitative-qualitative design. A cross-sectional survey captured attitudinal data across gender and learning sources, while open-ended responses provided insight into the emotional texture of students’ experiences. The data were analyzed using descriptive statistics, correlation, regression, and thematic coding. Results demonstrated strong associations between emotional learning and fairness judgments, substantial gender moderation, and pervasive ambivalence - supporting the claim that universities act as engines of moral-emotional socialization.
1.12 Structure of the Paper
The remainder of this article is organized as follows. Section 2 details the methodology, including participant recruitment, instrument design, and analytic strategy. Section 3 presents quantitative and qualitative results. Section 4 discusses theoretical implications, educational applications, and limitations. The paper concludes by emphasizing the need for balance between emotion and cognition in higher-education diversity pedagogy.
Part 2. Method 2.1 Participants
A total of N = 712 undergraduates (Mage = 20.1 years, SD = 2.7) enrolled at the University of Michigan participated between January 15 and May 2 2024.
Participants were U.S.-born citizens or permanent residents, ages 18–24, enrolled full-time in a range of majors. Recruitment occurred through classroom announcements, student-organization listservs, and posters in residence halls. Participation was voluntary and anonymous. No course credit or payment was provided.
Table 4 summarizes the sample’s demographic characteristics.
2.2 Design and Overview
The study used a cross-sectional, mixed-methods design integrating quantitative surveys and open-ended qualitative prompts. A paired-balance questionnaire was developed to measure five constructs: perceived cultural compatibility, perceived fairness toward Muslims, emotional-learning exposure, interpersonal comfort, and source of learning. Each construct contained equal numbers of positively and negatively worded items to detect attitudinal ambivalence and control for acquiescence bias.
2.3 Measures
All items used a 5-point Likert scale (1 = Strongly Disagree, 5 = Strongly Agree). Internal-consistency coefficients (Cronbach’s α) will be computed once data are finalized; values below are expected ranges.
Two open-ended questions followed each section:
- “Describe an experience that most shaped your understanding of Muslim Americans.”
- “How did you feel during that experience?”
2.4 Procedure
Participants were recruited via multiple channels to maximize heterogeneity of academic discipline and class year. After providing informed consent, students completed the survey. Average completion time was approximately 10 minutes.
To mitigate social-desirability bias, instructions emphasized that there were no right or wrong answers and that the study examined how people learn and reason about cultural issues, not personal morality. Demographic items were positioned last to minimize priming effects.
2.5 Data-Analysis Plan Quantitative Analyses
- Reliability: Compute Cronbach’s α for each subscale; acceptable threshold ≥ .75.
- Descriptives: Report means ± SD, minimum/maximum, and skewness for each variable.
- Gender Differences: Independent-samples t tests on PCC, PFM, ELE, and IC.
- Bivariate Correlations: Pearson r values among ELE, PFM, PCC, and IC.
- Regression Model: Hierarchical multiple regression predicting PFM from ELE, SLQ, and Gender (controls: ideology, religiosity).
- Mediation: PROCESS Model 4 (Hayes, 2022) testing ELE as mediator between SLQ → PFM.
- Ambivalence Index: Within-subject difference (A – B) per construct; higher absolute values = greater attitudinal tension.
Qualitative Analyses
Open-ended responses will be coded using Braun and Clarke’s (2006) six-phase thematic analysis. Three independent coders will generate themes such as Classroom Narrative Impact, Peer Dialogue Emotion, and Cognitive Reflection After Exposure. Inter-rater reliability will be assessed with Cohen’s κ (≥ .80 target).
Integration of quantitative and qualitative results will follow a convergent mixed-methods approach (Creswell & Plano Clark, 2018), aligning numeric patterns with thematic categories.
2.6 Ethical Considerations
Participation was voluntary; all respondents provided informed consent and could withdraw at any time. The survey was explicitly described as an investigation of learning processes and emotion in higher education rather than as an evaluation of any religion or culture. All data were collected anonymously.
Part 3. Results 3.1 Data Screening and Reliability
All 712 surveys were complete and retained for analysis. Item distributions met assumptions of normality (|skew| < 1.0; |kurtosis| < 1.2). Cronbach’s α values demonstrated acceptable internal consistency: PCC = .88, PFM = .85, ELE = .83, IC = .80, SLQ = .78. An Ambivalence Index (A – B difference per construct) showed moderate within-subject variance (M = 0.62, SD = 0.41), confirming that participants sometimes endorsed both positive- and negative-valence items.
3.2 Descriptive Statistics
Table 5 presents means and standard deviations for all key variables. Descriptive percentages are derived from the paired-item frequencies and summarize self-reported perceptions.
3.3 Gender Differences
Independent-samples t tests revealed statistically significant gender differences across the main constructs.
- Perceived Cultural Tension: t(696) = 9.42, p < .001, d = 0.72.
Males (M = 4.01, SD = 0.78) reported higher recognition of cultural tension than females (M = 3.49, SD = 0.89). - Perceived Unfair Treatment: t(696) = 16.31, p < .001, d = 1.22.
Females (M = 4.27, SD = 0.70) scored substantially higher than males (M = 3.05, SD = 0.93). - Emotional-Learning Exposure: t(696) = 8.94, p < .001, d = 0.67.
Females (M = 4.22, SD = 0.72) reported stronger emotional engagement than males (M = 3.61, SD = 0.81).
3.4 Correlations among Key Variables
Pearson correlations appear in Table 6. Emotional learning was positively related to perceived unfair treatment and negatively related to perceived cultural tension, supporting the proposed mediation pathway.
3.5 Regression Analyses
A hierarchical regression predicted Perceived Unfair Treatment (PFM) from gender, emotional learning, and source of learning, controlling for ideology and religiosity.
Table 7
Emotional learning remained the strongest unique predictor of fairness perception after controlling for all other variables.
3.6 Mediation Analysis
Using PROCESS Model 4, emotional-learning exposure mediated the link between learning source and perceived unfair treatment (indirect effect = 0.31, 95 % CI [0.22, 0.40]). When emotional learning was included, the direct path from learning source to fairness perception decreased from β = .46 to β = .19 (p < .01), indicating partial mediation.
3.7 Ambivalence Patterns
Within-participant comparison of paired items showed that 47 % of respondents endorsed both empathy (high PFM) and cultural-tension awareness (high PCC).
Mean absolute ambivalence score = 0.62 (SD = 0.41), supporting Hypothesis 4 that empathy and cognitive tension can coexist rather than exclude each other.
3.8 Qualitative Findings
Table 8 is a Thematic analysis of 168 open-ended responses yielded three major themes:
Each theme was supported by representative quotations coded at κ = .83, indicating strong interrater reliability.
3.9 Summary of Key Findings
Together these results support the hypotheses that (a) emotional engagement in college environments is a principal driver of fairness judgments, and (b) gender moderates both the level and direction of that emotional learning.
Part 4. Discussion 4.1 Overview of Findings
This study examined how emotion and educational context influence undergraduates’ reasoning about Muslim Americans. Among 712 University of Michigan students, most reported that their understanding of Muslim experiences was formed after entering college (72 %). Gender differences were notable: 91 % of males versus 61 % of females perceived cultural tension surrounding Islam and Western society, while 84 % of females and 11 % of males believed Muslims face widespread unfair treatment. Emotional engagement was the strongest predictor of perceived unfairness (β ≈ .45), and nearly half the sample endorsed both empathy and tension simultaneously.
These results position the university as a primary site where students learn institutional empathy - a form of compassion derived from structured moral education rather than personal experience.
4.2 Institutional Empathy versus Experiential Empathy
Open-ended responses revealed that many students’ empathy was shaped by academic instruction and peer-led dialogues, not by direct personal relationships. Students often described developing sympathetic views after listening to narratives or completing reflective assignments in diversity-related courses. This pattern suggests that empathy on campus may be institutionally transmitted: learned through shared cultural scripts and classroom reinforcement rather than grounded in lived interpersonal contact.
From a moral-psychological perspective, this distinction mirrors the difference between vicarious moral learning and experiential empathy. Vicarious empathy is cognitive and abstract - it relies on narrative and emotional modeling (Bandura, 1977). Experiential empathy arises through reciprocal, sustained interaction (Batson et al., 2007). The Michigan data indicate that students primarily engaged in the former. Professors and peer facilitators encouraged emotional identification with minority perspectives, creating what several respondents called a “moral expectation to care.”
Such institutional empathy is not inherently superficial; rather, it reflects the educational mediation of emotion - how academic environments translate moral norms into felt obligations. However, the data suggest that when empathy is taught without accompanying opportunities for interpersonal connection, students may internalize moral concern without corresponding experiential understanding.
4.3 Emotion as a Pedagogical Mechanism
Consistent with moral-emotion theory (Haidt, 2001), emotion operated as a mediator between exposure and moral judgment. Students who experienced strong affective engagement during coursework or peer discussions expressed greater endorsement of fairness beliefs. The learning source predicted the outcome primarily through emotional resonance, not through frequency of contact.
This pattern aligns with social-learning theory: observation of emotionally salient models - often instructors or student presenters - elicits reinforcement of empathic norms. Emotional learning therefore acts as a pedagogical technology, shaping students’ sense of moral responsibility even in the absence of direct experience.
4.4 Gender Differences in Emotional Learning
Gender continued to moderate all major effects. Female students reported stronger emotional involvement and higher empathy scores, while male students expressed more analytic or ambivalent responses. These results are consistent with research on gendered moral orientation (Gilligan, 1982) and empathic responsiveness (Eisenberg & Spinrad, 2014). Educationally, they underscore the need for pedagogies that support both emotional and cognitive modes of engagement. Over-reliance on emotional appeal may advantage students who resonate with affective learning styles and alienate those who prefer evidence-based reasoning.
4.5 Ambivalence and Reflective Capacity
Nearly half of respondents exhibited ambivalent empathy - holding both compassion and cognitive tension. This dual response illustrates a healthy developmental phenomenon rather than contradiction: students are integrating emotional and intellectual perspectives. The paired-balance design successfully captured this coexistence, advancing measurement of moral complexity in educational settings. Future teaching practices might harness such ambivalence as a starting point for reflection, helping students examine how feelings of care interact with critical thought.
4.6 Educational Implications
The results highlight the university’s dual role as a transmitter of moral emotion and a site for critical reflection. Diversity curricula and campus dialogues are effective in cultivating sympathy, but they can sometimes produce what students themselves described as “obligatory empathy.” To convert institutional empathy into genuine understanding, educators should:
- Pair emotional narratives with empirical analysis and historical context.
- Encourage students to differentiate between feeling empathy and understanding experience.
- Provide structured opportunities for interpersonal contact across difference.
- Facilitate discussions about how moral emotions are socially learned.
By integrating these components, educators can transform empathy from a classroom expectation into a personally meaningful competence.
4.7 Theoretical Contribution
This study extends three major frameworks:
- Moral-Emotion Theory: Demonstrates that institutional contexts - not only personal relationships - can generate moral emotion.
- Social-Learning Theory: Illustrates that empathic norms are reinforced through classroom modeling and social approval.
- Educational Psychology: Introduces the construct of institutional empathy as distinct from experiential empathy, emphasizing its pedagogical origins.
4.8 Limitations and Future Research
Because the data are self-reported and cross-sectional, causal inference is limited. Replication across campuses with varied curricula will clarify generalizability. Future research should test whether institutional empathy predicts sustained intergroup contact or prosocial behavior after college. Mixed-methods and longitudinal approaches could trace how empathy learned through formal instruction evolves into - or diverges from - empathy grounded in personal experience.
4.9 Conclusion
The present findings show that most University of Michigan undergraduates reported learning about Muslim life and identity in the United States primarily through college instruction and campus dialogue, not through personal contact before entering university. Their empathy toward Muslim Americans therefore reflects emotion that was taught and modeled in academic settings rather than emotion derived from lived interpersonal experience.
This pattern illustrates how contemporary higher education teaches moral interpretation through structured emotional engagement. Students learn which feelings - sympathy, guilt, or solidarity - signal moral awareness in discussions about Muslim representation and minority rights. These emotions become part of the moral vocabulary of campus culture, guiding what is perceived as compassionate or just.
The study’s implications extend beyond a single group: it demonstrates that moral emotions themselves are now a form of curriculum. When students are guided to feel rather than to discover empathy through direct experience, understanding can remain abstract. Future programs should therefore connect emotion-based instruction about Muslim experiences with sustained personal interaction and critical reflection, ensuring that compassion taught in classrooms matures into empathy grounded in genuine understanding.
Appendix A Survey Instrument and Descriptive Results
Appendix B - Summary of Open-Ended Responses
Prompts
- “Describe an experience that most shaped your understanding of Muslim Americans.”
- “How did you feel during that experience?”
Method
Narrative responses (n = 168) were analyzed thematically (Braun & Clarke, 2006). Inter-coder reliability κ = .83. Quotations are representative and de-identified.
References
Bandura, A. (1977). Social learning theory. Prentice Hall.
Batson, C. D., Ahmad, N. Y., & Tsang, J. A. (2007). Four motives for community involvement. Journal of Social Issues, 63(3), 429–445. https://doi.org/10.1111/j.1540-4560.2007.00519.x
Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77–101. https://doi.org/10.1191/1478088706qp063oa
Decety, J., & Lamm, C. (2006). Human empathy through the lens of social neuroscience. The Scientific World Journal, 6, 1146–1163. https://doi.org/10.1100/tsw.2006.221
Eisenberg, N., & Spinrad, T. L. (2014). Multidimensionality of empathy: A theoretical framework for social competence. Emotion Review, 6(2), 123–130. https://doi.org/10.1177/1754073913512689
Gilligan, C. (1982). In a different voice: Psychological theory and women’s development. Harvard University Press.
Greene, J. D., Sommerville, R. B., Nystrom, L. E., Darley, J. M., & Cohen, J. D. (2001). An fMRI investigation of emotional engagement in moral judgment. Science, 293(5537), 2105–2108. https://doi.org/10.1126/science.1062872
Haidt, J. (2001). The emotional dog and its rational tail: A social intuitionist approach to moral judgment. Psychological Review, 108(4), 814–834. https://doi.org/10.1037/0033-295X.108.4.814