Abstract
Prior literature suggests that self-transcendence (other-oriented) values may be a primary mechanism for moving beyond transitory feelings of empathy toward a compassionate orientation to addressing structural injustice. Active learning techniques in the classroom may then offer a fruitful platform for students to engage in the critical reflection connected to both values and compassion. This study uses a quasi-experimental pretest/posttest survey design to investigate the relationship among values, compassion, and active learning in a college-level Introduction to Sociology course. Student interviews conducted after semester completion are also used to explore the extent to which effects may persist beyond the conclusion of a course. Analyses suggest that exposure to active learning exercises as part of sociology course material results in a measurable shift toward self-transcendence and an increase in compassionate orientation, and values partially mediate this effect.
Keywords
The purpose of this study is to investigate the relationship between values, compassion, and the role of active learning in the higher education classroom. My research approach is grounded in the principles of the scholarship of teaching and learning (SoTL) as a framework of scholarly inquiry (McKinney 2018). I am guided by two broad research questions: (1) How are values related to active learning in a college-level Introduction to Sociology course? (2) To what extent are values connected to a compassionate orientation toward ending suffering and addressing injustice? My goal is to examine how a college-level sociology course structured for active learning may correlate with values focused on attending to the well-being of others compared with self-focused values (Batson 2011; Markus and Kitayama 1991) and a compassionate orientation as well as to assess the extent to which such effects may persist beyond the conclusion of a course.
The contributions of this study are threefold. First, it adds to the small body of empirical research that considers values and emotion specifically in the context of a college class (Connelly and Joseph-Salisbury 2019; B. Johnson 2005). Second, this research attempts to articulate a way to “bring social justice back to the center” (Feagin 2001:10) and move toward the direction of “macro-level mindfulness” (Lee 2015:275) within the sociology classroom. Much has been theorized about the transformative potential of education (Freire 2014; Gannon 2020), but there has been less guidance on the explicit “how” of accomplishing this. Third, the study contributes to research by faculty at a two-year institution. Because community college sociologists are afforded a marginalized status within the discipline as well as have higher teaching loads and fewer institutional resources for scholarly work, they are less likely to conduct or publish research and to be represented in SoTL literature (Paino et al. 2012; Rowell 2010). Yet community colleges are the likeliest institutions where a student first encounters “the public face of sociology” (Zipp 2012:304), and community college faculty are steeped in doing the work of social justice (Brown et al. 2016).
Emotion in the Classroom
If the sociological imagination is a foundation to the discipline, then the teaching of sociology is crucial in the effort to fulfill its empowering and enduring promise (Scanlan and Grauerholz 2009). Few students in introductory-level classes may go on to become social scientists, but if students are constituted as a public and teaching as a form of public sociology (Burawoy 2014), then we as faculty should seek to give them sociological tools that can be used in any field after a course has concluded.
Emotions are central to this “emancipatory project” (Hill Collins 1998) and the pedagogy of transformative teaching (hooks 1994). Emotions not only make us feel something, they make us feel like doing something (Jasper 2014). The sociological importance of attending to emotion is broadly evident in classic works (Turner and Stets 2006), and contemporary researchers note that emotional reactions shape our relationships with ourselves, with other individuals, and with social groups (Collins 1990; Thoits 1989). Persistent emotions, like righteous anger, may also be a prerequisite for collective attention to inequalities and social injustice (Jasper 2014; Schrock, Holden, and Reid 2004).
Attending to the emotional element is a necessary aspect of effective teaching and learning (Cavanagh 2016; Kaufman and Schipper 2018). For instance, empathy is rarely absent or far from sociology classrooms. Empathy produces a response to the moment, a “willingness to respond to the suffering of another as they’ve just now experienced it” (Bloom 2016:23). The quandary lies in getting that willingness to persist beyond the boundaries of an isolated emotional experience. Empathy by itself can feel exhausting or powerless and lead students to “check out” from uncomfortable material, from the course, and perhaps from the whole discipline (B. Johnson 2005). Similarly, the weight of being sociologically mindful without an outlet to attend to it may become a burden on the individual student (Yogan 2015).
One potential means of ameliorating such burden is to differentiate empathy from compassion. Compassion is “the feeling that arises in witnessing another’s suffering and that motivates a subsequent desire to help” (Goetz, Keltner, and Simon-Thomas 2010:351). If empathy is “I know how you feel,” then compassion is “How can I help?” The underlying motivation to ease suffering makes it possible to engage anyone compassionately without personal familiarity or the potentially overwhelming burden of directly experiencing what someone else feels. A compassionate orientation to social life entails a desire not only to understand the reciprocal connections between personal troubles and public issues (Mills 1959) but also to alleviate social inequalities. Thus, compassion provides one path to prompting sociologically informed social action in a way that empathy alone cannot.
Values
Research suggests that values play a role in influencing a compassionate orientation. Unlike fixed personality traits or transitory attitudes, values operate as guiding mechanisms and core influences rather than the outcomes of some other situation or process (Hitlin and Piliavin 2004; Longest, Hitlin, and Vaisey 2013). Values define how we evaluate ourselves, and these standards are also connected to facets of social location. Individuals make decisions in accordance with the values shaped by their social class, and in doing so, they may also reproduce their class position (Bourdieu 1984; Kohn 1959). Race and gender similarly may influence values by shaping perceptions as well as differential role expectations and interactions (Elliott and Smith 2004; Ridgeway and Berger 1986). Although specific identities may become salient in different situations (Stryker and Serpe 1994), one’s values tend to remain stable across social situations and provide a bridge between group and role identities (Hitlin 2003). People are driven to act consistently with their internal values to be authentic and remain true to oneself (Erickson 1995).
Schwartz’s (2003, 2006) heterarchical system, in which 10 value categories are structured along two values dimensions, provides an ideal model for considering the connection between values and a compassionate orientation. The self-enhancement/self-transcendence dimension relates to the conflict between concern for the consequences of actions for the self (self oriented) and concern for the consequences of actions in the broader social context (other oriented). Hitlin (2003) found that the formation of a broad volunteer identity is predicted by self-transcendence and an orientation of care toward others. Self-transcendence values may also help circumvent the “empathy wall” of indifference or hostility to those outside our own social circles (Hochschild 2016). This suggests self-transcendence values are a primary factor in the development of a compassionate orientation. A person with greater self-transcendence may be more likely to behave in consistently compassionate ways across their social roles and encounters to benefit others, even if their current individual power is constrained by virtue of social location. Conversely, self-enhancement values may hinder the development of compassion in favor of motivating self-focused action.
Active Learning
Incorporating active learning in the classroom can be a useful teaching strategy for achieving the goals of shifting values and encouraging a compassionate orientation. Active learning refers to student-centered pedagogical techniques, like small-group exercises, simulations, or discussions (Atkinson and Lowney 2015; Baepler et al. 2016; Meyers and Jones 1993). Class exercises designed for active learning have been shown to be particularly useful in the teaching of theory and stratification patterns connected to gender, class, race, and other inequalities (Hartung 1991; Kleinman, Copp, and Sandstrom 2006).
Pedagogical approaches that attempt to create personal relevance while demonstrating an everyday sociological process can yield better concept retention and understanding (Murphy and Ribarsky 2013; Sobal et al. 1981). More importantly, these approaches help students develop the sociological imagination (Atkinson and Lowney 2015). Becoming aware of how others are affected and noticing whether others may be suffering necessitates becoming aware of one’s own role. Students who are prompted to actively reflect on how they personally engage in a process, as opposed to being presented only with data they might ideologically dismiss as things that other people do, may be more likely to engage in the critical thought connected to values and compassion. Active learning also provides a necessary platform for students to begin exploring how individual agency is connected to the process of constructing structural change via social action (B. Johnson 2005).
Summary
I posit that fostering the development of a compassionate orientation is an important and desirable outcome of a sociology course. This aligns with the foundational concepts outlined in the sociological literacy framework, particularly its first essential concept of developing “the sociological eye” (Pike et al. 2017:55). I also suggest that values are crucially connected to compassion. If taking an introductory sociology course shifts a student’s values toward self-transcendence, then there should be an accompanying increase in compassionate orientation. Given that values are guiding influences and aspects of identity, I would further expect a shift toward self-transcendence values and increased compassion to endure beyond the time frame of a single semester.
I propose that active learning techniques in a course play a key role in fostering a compassionate orientation and that values, specifically the dimension of self-transcendence/self-enhancement, are the driving mechanism. The hypotheses emerging from this argument are as follows:
Methods
This research was approved by the Institutional Review Board for the Protection of Human Subjects at the author’s institution.
Sample
The sample for this study was drawn from undergraduate students 18 years or older at a large northeastern two-year institution. The college serves a sizable proportion of low-income, minority, and nontraditional students, which is typical for a community college and comparable to other two-year institutions in the state (Ma and Baum 2016). The general population at this institution is 59 percent female and 29 percent racial/ethnic minority, with an average age of 25 years. Approximately half are first-generation college students, and nearly two-thirds of all students receive income-based Pell grants.
Participants in this sample were registered students who self-enrolled in one of my six face-to-face Introduction to Sociology class sections during the 16-week fall 2019 semester. Incomplete data from participants who did not complete both the beginning and end of semester surveys due to either course withdrawal or absence were omitted from the analysis, resulting in a final sample of 171 participants. Of the final sample, 39 participated in follow-up interviews.
Research Design
The first phase of the research was a quasi-experimental two-group pretest/posttest design for evaluating the effectiveness of an intervention. Two course sections were chosen to serve as the control group by randomly selecting course ID numbers, and the remaining four course sections served as the experimental group. For the pretest, a voluntary self-administered survey with a values inventory was given to students in experimental and control group sections during the second week of the class. During the semester, students in the experimental group sections engaged in three specific active-learning classroom exercises and completed an associated written reflective assignment in class. During the last week of the course, a second voluntary self-administered survey with values inventory constituted the posttest given to students in all course sections.
To eliminate any potential discomfort or risk to participants, neither the pretest nor posttest survey was connected to any type of graded assessment, nor were the surveys connected to any offer of extra credit for participation. A colleague administered the surveys with informed consent, and I was not present in the class for either pretest or posttest. The colleague assigned a numerical identifier to each name and retained all participant data until the end of the semester. I did not have access to the collected surveys or consent forms until after final course grades were posted.
The second phase of the research involved data collection from a subset of follow-up interviews conducted after the conclusion of the course. Fall 2019 students who had completed both the pretest and posttest surveys were invited to participate in an audio-recorded interview with me during the latter half of the spring 2020 semester.
Measures
The structure of the class that participants took in this study did not vary by course section, and there were no comparison sections utilizing alternative substantive material or a different instructor. The course objectives were the same across all six course sections, and all included the same weighted grade components of weekly quizzes, exams, participation, and in-class assignments/exercises. What varied between the experimental and control groups was the intervention variable of active learning.
Active learning
Students in the experimental group engaged in three specific active-learning classroom exercises with a written reflective assignment completed at the end of the class period. The control group sections were provided traditional in-class lectures and individual homework assessments to complete, but they did not include the three active-learning exercises and writing assignments designated for the experimental group sections in this study. The variable active learning was dummy coded as 1 = experimental group, 0 = control group.
These activities were designed to highlight the macro-, meso-, and microlevels and effects of inequality as well as to consider individual and structural ways of addressing them. 1 Activity 1, Class Inequality, is a game I created by drawing on published activities and suggestions for using stratified and intersectional variants of Monopoly (Coghlan and Huggins 2004; Messina-Yauchzy 2007; Smith 2017). I conduct this activity after students have been assigned to read the textbook chapter on U.S. stratification plus three short essays reflecting on life at different socioeconomic positions (New York Times 2005; Tirado 2013; Wood 2011). Activity 2, Gender Inequality, is an exercise on gendered stereotyping in occupations I created to serve as a lead-in to covering the binary gender system (Adkins 2018). I conduct the activity after students have been assigned to read the textbook chapters on sex, gender, and sexuality. Activity 3, Race Inequality, is a “fake quiz” class exercise I created by drawing on published activities for teaching about structured inequality and institutional discrimination (Eells 1987; Obach 2000). I conduct the activity after students have been assigned to read the race/ethnicity textbook chapter and two book excerpts (Bonilla-Silva 2018; A. Johnson 2018).
Values
The pretest and posttest surveys incorporated the Portrait Values Questionnaire (PVQ), developed to measure 10 basic category values and their associated value dimensions (Schwartz 2003). Participants were asked to read short portraits of hypothetical people, then indicate how similar each described person is to themselves on a scale ranging from 1 = not at all like me to 6 = very much like me. Each of the category values comprising the self-transcendence/self-enhancement dimension is represented by a group of PVQ inventory items as shown in Table 1.
Portrait Values Questionnaire Value Categories and Dimensions with Associated Item Inventory.
In accordance with Schwartz’s (2006) recommendation to mitigate concerns with social desirability biases in the PVQ, I ipsatized scores by subtracting each participant’s mean for the sum of all response items from the mean of each individual response item. Each of the values category variables were created by averaging these centered scores on related inventory items. Then the compiled means for benevolence and universalism were summed to create the self-transcendence variable, and the compiled means for achievement and power were summed to create the self-enhancement variable.
Compassionate orientation
The pretest and posttest surveys incorporated two items drawn from the Compassionate Love Scale Humanity–Strangers Form (Sprecher and Fehr 2005): (1) “If she encounters a stranger who needs help, she would do almost anything she could to help them,” and (2) “He would rather engage in actions that help others, even if they are strangers, than engage in actions that would help himself.” For consistency with the rest of the survey, these two items were rewritten to be presented in the same hypothetical third-person style of the PVQ and given the same response options ranging from 1 = not at all like me to 6 = very much like me. The scores on these two items were averaged.
Control variables
Education was measured by number of college credit hours completed, with 0 to 60 credit hours evenly split into four categories. Two items asked about each parent’s education level to determine whether the participant was a first-generation college student. Low-income status was measured by an item asking if the participant is a Pell Grant recipient. Both gender and race were measured in a series of categories allowing participants to self-identify, and then the variables were dummy coded as white = 1 and male = 1. Age was measured by asking for birth year. For family roles, marital status was measured in five categories, then collapsed and dummy coded as married = 1. A dichotomous item asked the participant to indicate if they have children, and this was similarly coded as parent = 1.
Course impact
Interviews were semistructured with four broad questions: (1) “In general, how did taking this course impact (or not impact) you? What did you get out of the course or take away from it?” (2) “Can you describe specific material, topics, or course activities that made a lasting impression, either a positive or negative impression?” (3) “Since taking the course, are there any situations or experiences you can tell me about where you behaved in a certain way or made a particular choice based on something you learned or encountered in the course?” (4) “What other general impressions, feedback, or comments on the overall course do you have to share?”
Results
Table 2 presents the descriptive statistics for the aggregate sample (n = 171) and then separately for the control group (two course sections, n = 58) and experimental group (four course sections, n = 113). Participants ranged in age from 18 to 50 years old, with a mean age of 24, and had completed an average of 16 to 30 college credit hours. Nearly half of the participants were low income, and almost two-thirds were the first in their families to attend college. Men constituted 23.3 percent of the sample, and 80.1 percent of the participants were white.
Means and Standard Deviations of Control Variables by Sample Group.
Mean significantly different from full sample, p < .05.
None of the means from the experimental group and the control group differ significantly from each other, and none of the means from either group differ significantly from those of the aggregate. The full sample overrepresents women and first-generation students compared with the typical averages for this specific institution, but all other sample characteristics are within the expected range for this and other community colleges (Ma and Baum 2016).
Correlations, Means, and t Test
Table 3 presents the correlations for all study variables. Self-transcendence, self-enhancement, and compassionate orientation each produced a time 1 (t1) and a time 2 (t2) variable. The t1 and t2 variables are significantly correlated, which is expected since there is no reason to anticipate dramatic shifts over the period of a single semester, and the strongest predictor of values or compassion at t2 should be their initial level at t1. As polar ends of a continuum where prioritizing one side of the dimension involves deprioritizing the opposing side, self-transcendence and self-enhancement at t1 and t2 are almost perfectly negatively correlated.
Correlation Matrix for All Study Variables.
Note: t1 = time 1; t2 = time 2.
p < .05. **p < .01. ***p < .001 (two tailed).
As should be expected based on social psychology and values literature (Burke, Stets, and Cerven 2007; Elliott and Smith 2004; Prince-Gibson and Schwartz 1998), being male is significantly correlated with lower self-transcendence and greater self-enhancement. Being a parent is also correlated with lower self-enhancement, which reiterates that those who are caregiving are most likely to hold values emphasizing concern for others (Schwartz and Rubel-Lifschitz 2009). This also connects to the correlation between male and self-enhancement, given that women are those most likely to be responsible for taking care of children.
The positive correlation between low income and self-transcendence at t1 aligns with previous research suggesting that those with lower socioeconomic status are more likely to be charitable and less likely to prioritize self-interest over the well-being of others (Piff et al. 2010). The similar correlation between self-transcendence at t1 and being a first-generation student aligns with research demonstrating that these students are more likely to be dependent upon family capital (Gofen 2009). Yet both these correlations with self-transcendence at t1 are reduced to nonsignificance at t2. This could suggest that the improved self-efficacy among first-generation and low-income students that is gained from attending college courses (Ives and Castillo-Montoya 2020; Ramos-Sánchez and Nichols 2007) is related to an increase in prioritizing academic self-interest and achievements.
Compassionate orientation is not significantly correlated with any of the control variables directly even though previous studies demonstrate that there is reason to expect that race, age, income, and gender are related to compassion (Sprecher and Fehr 2005; Stellar et al. 2012). As with values (Longest et al. 2013), it may be the case that none of these factors are a significant direct influence in isolation but that their various combinations have a stronger and demonstrable influence on compassion. This may further indicate an intervening or mediating influence between social status and level of compassion.
Additionally, I conducted paired-sample t tests with all outcome variables as a preliminary step to examine whether the paired pretest and posttest means differed significantly. These analyses (not shown) indicate that means were significantly different between t1 and t2 for all variables of interest. This suggests that a measurable change or shift in values and compassionate orientation did occur between the beginning and the end of the course. As anticipated, self-transcendence and compassionate orientation increased, while self-enhancement decreased.
Regression Models
To test the study hypotheses, I ran a series of ordinary least squares linear regression models for direct and indirect effects. 2 Because using posttest minus pretest “change scores” in analyses may be affected by regression to the mean (Allison 1990), I used the method in which posttest t2 variables are regressed on pretest t1 variables. Multicollinearity testing indicated that none of the study variables had a variance inflation factor greater than 1.8.
Table 4 shows the result of regression analyses for direct effects on the values dimensions. Active learning is significantly related to self-enhancement (b = −.407, p < .01) and to self-transcendence (b = .372, p < .01) in the expected direction, supporting Hypotheses 1a and 1b. Gender is also significant: Being male is positively associated with self-enhancement but negatively associated with self-transcendence values. It is unsurprising as a reflection of socialization patterns and patriarchal social context that men are more likely to score higher on the self-enhancement dimension, which incorporates the specific category values of power and achievement (Schwartz and Rubel 2005).
Regression of Values on Active Learning and Control Variables.
Note: t1 = time 1; t2 = time 2.
p < .05. **p < .01. ***p < .001 (two tailed).
Table 5 shows the results of regression analyses for direct effects on compassionate orientation (Model 1), mediation via self-enhancement values (Model 2), and mediation via self-transcendence values (Model 3). Model 1 indicates that active learning is significantly positively related to compassionate orientation (b = −.415, p < .01), supporting Hypothesis 1c.
Regressions of Compassionate Orientation on Active Learning, Values, and Control Variables.
Note: t1 = time 1; t2 = time 2.
p < .05. **p < .01. ***p < .001 (two tailed).
The mediation models incorporate self-enhancement and self-transcendence values at t2, which were previously shown as significantly related to active learning, as a predictor variable for compassionate orientation. Self-enhancement is negatively associated (b = −.245, p < .001) and reduced the significance level of active learning on compassion compared with the direct-effects model. Similarly, self-transcendence is positively associated (b = .260, p < .001) and reduced the significance level of active learning compared with the direct-effects model.
Additional testing of Models 2 and 3 using a percentile bootstrap estimation approach (Hayes 2017, 2020) indicated a small but significant indirect effect of self-enhancement (b = .1028, SE = .0506, 95 percent confidence interval [CI] = [.0215, .2175]) and likewise a small but significant indirect effect of self-transcendence (b = .1007, SE = .0499, 95 percent CI [.0217, .2106]). Gender is again significant in both the direct-effects and mediation models with respect to self-transcendence: Being male is negatively associated with lower compassionate orientation. The mediation models also explain more variance in compassionate orientation, 20.6 percent and 19.3 percent, respectively, versus the 14.4 percent explained by the previous direct effects model. Thus, Models 2 and 3 support Hypotheses 2 and 3, indicating that values do partially mediate the effect of active learning on compassionate orientation.
Course Impact
I conducted 39 interviews with participants who had completed both the pretest and posttest surveys during the fall 2019 course. Descriptive statistics for the interview subsample are shown in Table 2. Compared with the full study sample, there was a significantly higher proportion of first-generation and nontraditional students among interview participants. Such students are likely to have already spent time with me in office hours, and due to our rapport, they may have felt more comfortable speaking with me about their course reflections. This possibility aligns with research highlighting the crucial role that personal relationships with faculty and staff play in increasing student engagement (Chambliss and Takacs 2014).
I transcribed the interview recordings and analyzed the generated data through open then axial coding (Strauss and Corbin 1998). I also ran crosstabs to explore whether the interview participants varied in mean levels of values or compassionate orientation compared with the full sample, but the results of these analyses (not shown) yielded no significant findings. My analysis of interview transcripts resulted in two prominent themes regarding the “takeaways” reported by participants: (1) relevance of material and (2) impact of active learning. Each is illustrated by a selection of quotations drawn from the transcripts.
Relevance of material
Many participants described, often with genuine surprise, being pleased to find that the overall course material related to their daily lives. The discovery is not uncommon among first-time sociology students (McKinney 2007), and the utility of stressing real-world connections is frequently emphasized in strategies for teaching sociology (Atkinson and Lowney 2015). This realization is also an important preliminary step toward developing the necessary sociological mindfulness (Schwalbe 2008) connected to awareness and compassion.
This class, this stuff I felt like I could use all the time because we always talked about things going on now. It actually made me start watching the news, which I never really did that much before. So I pay more attention to what’s happening. (white male, 21, low income, first-generation student) [The course] turned out useful. I think I learned quite a few life lessons. It wasn’t all about sociology so that’s what I liked: how it was made to relate to our lives. (black female, 20, low income, first-generation student) It really opened my eyes to the world, and I thought they were pretty open already going into it. I look a lot deeper into everything now, which is sometimes good and sometimes bad. You ruined some favorite shows for my husband but I feel like I try to understand other things much better now. (black female, 30, low income, married, parent, first-generation student) I’ve noticed things that I wouldn’t have thought about before, like “whoosh” right over my head. I don’t know, sometimes that’s uncomfortable and I don’t want to see it. Like there’s so much racism, you know. So much! How did I miss it? (white female, 26, low income, married) I found that situations have many different angles; there is never only one choice. Like when we talked about there’s not just “one big truth with a capital T” in society. (black male, 24, low income, first-generation student) I have improved on my biases towards people and I’ve also tried to correct my family on their biases that they might not realize otherwise. That’s sometimes really hard and we’ve had some arguments about it. But it’s a good thing too. (white female, 29, first-generation student) Mostly this class made me want to be a better person. So we all can be better people. Impossibly idealistic, maybe. But I try to do it every day if I can. Get out my toolbox and do it. (white female, 41, low income, married, parent, first-generation student)
As with the comments shown here, many interviewees did not necessarily cite only a specific event, action, or lesson as something they took away from the course. Rather, they frequently described an overarching way in which they felt changed after taking the course. Comments noting the desire to improve biases, understand others, and be a better person are reflective of self-transcendence values. Most importantly, these participants suggested they had since been thinking or interacting differently and indicated taking action based on these changes.
Impact of active learning
Consistent with my expectations and my teaching goals, active learning left a more persistent impression. When I asked interviewees about activities that made a prominent impact or stuck with them since the completion of the course, any specific activities mentioned were always those that involved some form of active-learning activity.
For instance, one of the lessons I present within the first two weeks of a course covers facts, opinions, misconceptions, and the pitfalls of relying on “common sense” wisdom. I relate an anecdote of a bluffing game my grandma taught me where you attempt to “call bullshit” on other players to reveal why/how they are being deceptive. As a class we briefly discuss the purpose of such a game and try to play one, then we work together through two case studies from the “Calling Bullshit” syllabus (Bergstrom and West 2017). While it is not one of the three specific exercises used only for this study, this lesson is always a standard part of my introduction to the course and is the sole active-learning type of activity that I did in both the experimental and control group courses. The exercise was frequently mentioned in participant interviews as a memorable and helpful lesson.
One thing is to investigate things in a reliable way and not just share them without caring or checking anything. I saw a lot of what we talked about in classes on social media. I “bring the receipts”’ for those now. It has changed a lot of the way I think on certain things like my work or family and stuff. (white female, 22, low income) The phrase “calling bullshit” will always make me think of this class. In a good way, I mean. I would happily take a whole class just about calling bullshit. I’ve been showing my teenagers. (black female, 38, low income, parent, first-generation student) I’m not going to lie, I was a little annoyed after that first thing in the beginning. Annoyed with you. Like, “For real, did this lady just really say my opinion is wrong? Whatever, we’ll see about that.” But I got it, I get it. I did get it. That wasn’t it. It took me a minute to understand. (black female, 25, low income, parent, first-generation student)
Similarly, both the stratified Monopoly game and racial-privilege exercise used as part of the active-learning intervention variable were frequently mentioned as being impactful beyond just that moment in time. These exercises made a point in the classroom as intended, but they also prompted deeper thinking about social context even after the course was over. This is the most essential goal of active learning.
Playing real-world Monopoly for one made me mad! I still remember that thing. Fun but real eye-opening, too, you know. Like poverty or gender and race stuff outside of a textbook. Like it made you notice things you usually don’t ever think about even though they’re right there all the time and you’re just used to it. It still does now. (white female, 34, married, first-generation student) That horrible fake quiz, too. It made me so mad and upset that I did so bad I was ready to leave, but then finding out it was built that way on purpose, it really blew my mind. That was systemic. Like just because of where I sat I couldn’t do as good as the guy next to me who’s never ready for class? So much like the actual world.” (white female, 27, first-generation student) Definitely the tricky fake quiz I remember a lot. That was a trip. I’ve always hated that whole word privilege, like I’m supposed to feel guilty or something for being a white man? That’s unfair. But that was the first time I ever understood what privilege meant without being an accusation. I think there’s still way more to success than just that, but that day helped think about it a little different and try to remember structure. (white male, 25, married)
These interviewees were not necessarily representative of all participants in the sample or of all students in any single course section. Much like end-of-semester evaluations, the interviewees’ responses may overrepresent those students most likely to have strong feelings about the course or about me as its instructor (Goos and Salomons 2017; Stark and Freishtat 2014). Nonetheless, their general reflections remain valuable as a window into what students take away and what may stick with them beyond the conclusion of a course.
Discussion
The presented findings indicate that active learning is associated with a measurable shift away from self-enhancement values and toward self-transcendence and a compassionate orientation. The direct-effects hypotheses for these outcomes were supported. Exposure to active-learning exercises as part of sociology course material is associated with lower self-enhancement values, greater self-transcendence values, and increased compassionate orientation. Furthermore, the mediation hypotheses were also supported. Self-enhancement and self-transcendence values partially mediate the effect of active learning on compassionate orientation.
One consistent finding was that being male is a significant predictor in every regression model. As mentioned, this is not an unexpected result in the context of a patriarchal culture. If the gendered meanings of manhood, particularly those prevalent in hegemonic masculinity (Connell and Messerschmidt 2005; Schwalbe 2014), dictate that compassion and focusing on others’ well-being are feminine qualities, then it makes sense that being male is associated with significantly greater self-enhancement values and lower compassionate orientation. Prior research on gender as an ideology (Kroska 2000) and as a status characteristic (Ridgeway 2011) supports such a conclusion. It is important to reiterate, however, that the collapsed dummy variable for male I used in this study is reflective of sex, not gender identity. It does not capture the complexity of gender identities or what it may mean to a participant and how that subsequently affects their values or feelings. This remains a rich potential area for future study to examine.
The qualitative interview data provide additional support and insight into the overall influence of active learning as well as general student perceptions of any sociology course. While the interviews did not ask explicitly about other-focused values or specific acts of compassion and cannot be used as evidence of a direct effect, the participant responses do suggest the overall growth of self-transcendence values and a more compassionate orientation to their social lives. I interpret their statements as demonstrating the “long game” of the sociological imagination at work: encouraging sustained critical thought about social context and, potentially, an increased probability of engaging in behaviors that lessen social inequalities. This supports the premise of active learning producing the small but meaningful shifts in values and compassionate orientation that are evidenced in the regression results. It is reasonable to suggest those shifts might operate akin to planted seeds that may eventually grow into a consistently higher likelihood of compassionate social action in the future.
Quasi-experimental designs allow research to occur in a natural setting, offer wider applicability of results, and help examine the feasibility of an intervention (Marsden and Torgerson 2012). These were important concerns for me. Yet it is also important to recognize that one limitation is this research design could not eliminate the possibility of alternative explanations connected with history, maturation, or statistical regression effects (Gravetter and Forzano 2011). Future research of this nature would need to reduce such potential concerns.
A second limitation lies in my assumption that each of the active-learning exercises prompted the experience of compassionate feelings, in part because they were specifically designed with that end in mind. While the results clearly indicated that active learning has both direct and indirect effects on compassionate orientation, the measure was not intended to assess compassion as a product of any specific course exercise. That remains an opportunity area for research to explore the connection between presentation methods of specific material and the experience of compassion as they may relate to values change.
A final limitation is that this was a small, exploratory study, and I was unable to expand upon any other patterns hinted at in the data. Future studies with a broader range of students from a variety of institutions would be needed to provide more insight. I also collected survey data at only two limited points in time. Prior research shows that individuals continue to change slowly throughout the life course as a reflection of experience, education, and maturity (Falletta and Dannefer 2014; George 1993). For considering complex long-term shifts in values and compassion and their connection to active learning, longitudinal studies designed to capture multiple observations over the breadth of a college career would be most useful.
Even with these issues and limitations in mind, this study can have important practical implications for the sociology classroom. The findings with respect to the development of self-transcendence values and a compassionate orientation are promising. They suggest that if the goal is not only to cover important theoretical concepts but to give students “takeaway” tools that can be used outside of the classroom, then an effective approach includes incorporating active learning. Teaching with strategies geared to affect values change toward self-transcendence can provide a key space for sociology students to grapple with social issues critically and compassionately and to continue doing so beyond the duration of a course.
Footnotes
Acknowledgements
I would like to thank my students for generously sharing their experiences and making this research possible. I am also grateful to Rebecca Erickson, Kathy Feltey, Matt Lee, John Zipp, Gary Holliday, and the anonymous reviewers for their insightful comments and helpful feedback on this article.
Editor’s Note
Reviewers for this manuscript were, in alphabetical order, Jay Howard, Kathleen Korgen, Yvonne Luna, and Matthew McLeskey.
