Running Head: ACADEMIC DISIDENTIFICATION ... - Towson University

Running Head: ACADEMIC DISIDENTIFICATION IN BLACK COLLEGIANS

Academic Disidentification in Black College Students: The Role of Teacher Trust and Gender

Shannon McClain, Ph.D. Department of Psychology Towson University 8000 York Road, Towson, MD 21252 [email protected] and Kevin Cokley, Ph.D. Department of Educational Psychology University of Texas at Austin [email protected]

Keywords: academic disidentification, academic self-concept, teacher trust, Blacks

Please cite as: McClain, S. & Cokley, K. (In press). Academic disidentification in Black college students: The role of teacher trust and gender. Cultural Diversity and Ethnic Minority Psychology. This article may not exactly replicate the final version published in the APA journal. It is not the copy of record. See version published online before print: http://dx.doi.org/10.1037/cdp0000094


2

Author Note This article is based on part of a dissertation submitted by Shannon McClain to the University of Texas at Austin in partial fulfillment of the requirements for the doctoral degree under the supervision of Kevin Cokley. We thank the members of the dissertation committee, Keisha Bentley-Edwards, Christopher McCarthy, Pamela Moore, and Marie-Anne Suizzo.

Correspondence concerning this article should be addressed to Shannon McClain, Department of Psychology, Towson University, 8000 York Road, Towson, MD 21252. Email: [email protected].


3

Abstract Objectives: Research has identified academic disidentification as a phenomenon that appears to uniquely impact Black male students. However, few empirical studies examine what underlies such gender differences. This study examined whether students’ teacher trust is a factor underlying academic disidentification in Black college students and whether this is moderated by gender. Academic disidentification was investigated by examining the strength of the relation between a student’s view of his/her academic abilities in comparison to peers (i.e., academic self-concept or ASC) and the student’s academic outcomes (i.e., GPA). Attribution theory was used as a lens to test a hypothesized multi-group path model that linked age to teacher trust and ASC, and ASC to GPA through teacher trust. Alternative models were also tested. Method: Participants were 319 Black students (120 males and 199 females) recruited from a large, southwestern, predominantly White university. Results: Results revealed the hypothesized model fit the data reasonably well, while the alternative models resulted in a poorer fit. The final model supported our hypothesis that the relation between ASC and GPA is partially mediated by teacher trust and this relation was moderated by gender, such that the indirect effect was significantly stronger for males than females. Several significant differences were also found across gender for direct paths. Conclusions: These findings suggest college students’ trust of faculty may be particularly important for Black males and is likely a contributing factor to academic disidentification. Practical implications for university professionals’ facilitation of Black college students’ academic development are discussed. Keywords: academic disidentification, academic self-concept, teacher trust, Blacks


4

Academic Disidentification in Black College Students: The Role of Teacher Trust and Gender While the U.S. Department of Education has named as a part of its mission the goal of closing the achievement gap (U.S. Department of Education, 2015), disparities are apparent at every level of education. Research often fails to contextualize this within the context of the stark gender disparity that exists, as Black males are particularly likely to fall prey. The gap between Black male and female college enrollment and academic outcomes represents the largest gender imbalance of any racial-ethnic group (Aud, Fox, & KewalRamani, 2010). One line of research is of particular importance in its observation of such gender differences. Existing research has examined a phenomenon uniquely consistent for Black males: academic disidentification (Cokley, McClain, Jones, & Johnson, 2011; Osborne, 1995; 1997). However, scholarship is needed to address the theoretical underpinnings of academic disidentification and test models related to theory. We suggest academic disidentification is related to an attributional mechanism, such that teacher trust plays an important role in students’ understanding their academic abilities and outcomes. Academic Disidentification Conventional wisdom suggests good academic performance will be rewarding, while poor academic performance will be punishing and, thus, students should be academically motivated as a function of seeking reward and protecting their self-esteem (Cokley, 2002). However, academic feedback and performance is not intrinsically rewarding or punishing (Osborne, 2001). Steele (1992) described academic disidentification as a phenomenon in which one’s academic performance does not impact one’s self-views, as it does for others. Thus, for individuals who are disidentified, academic performance will not highly impact their self-esteem


5

and these individuals will theoretically not be motivated to perform well academically (Osborne, 1997; Steele, 1992). Steele (1992) theorized that while all students begin as identified with academics, over time Black students are more likely than other groups to devalue academics as a protective mechanism. In the current study and consistent with prior investigations (Cokley, 2002; Cokley, et al. 2011), academic disidentification is expressly defined as a weak or attenuated relation between a student’s view of his/her academic abilities in comparison to peers (i.e., academic self-concept or ASC) and the student’s academic outcomes (i.e., GPA). When individuals are identified with academics, students’ perceptions of their own academic and intellectual abilities strongly correlate with their academic outcomes. Research suggests ASC is arguably the strongest psychological correlate of academic outcomes in both multi-ethnic and homogenous Black samples (Cokley, 2002; Cokley et al., 2011). However, for those who disidentify with academics, academic performance does not strongly impact self-views such as academic selfconcept. Most empirical research examining academic disidentification has been done with adolescents in middle and high school. These studies suggest disidentification develops over time and is uniquely consistent for Black males, but inconsistent for Black females. While there is literature to suggest disidentification occurs for males and not females (Cokley et al., 2011; Osborne, 1995; 1997), some studies have failed to find such gender differences (Hope, Chavous, Jagers, & Sellers, 2013; McMillian, Carr, Hodnett, & Campbell, 2015). While intuitively it may seem disidentification would not significantly impact students who are able to excel enough to reach college, Major, Spencer, Schmader, Wolfe, & Crocker (1998) provided evidence of disidentification in Black college students in two experimental


6

studies. Further, Cokley (2002) provided strong evidence for academic disidentification among Black male college students, as it appeared Black males showed a progressively reduced positive bi-variate correlation between ASC and GPA (dropping from .52 for underclassmen to .17 for upperclassmen), while Black females and White males and females maintained a significant positive bi-variate correlation. The examination of disidentification remains a critical, although neglected, area of inquiry. There is a dearth of literature contributing to theory related to this phenomenon, as little research has empirically examined the mechanisms underlying this relation. To date, Hope and colleagues (2013) provide the only study to empirically examine a psychosocial variable influencing academic disidentification, reporting strong, positive views of racial identity may help Black students maintain their self-esteem despite poor academic outcomes. More scholarship is needed to help determine for those who disidentify (i.e., who exhibit a weak relation between ASC and GPA), what contributes to the attenuation of what is typically such a strong and consistent relation. Although past research has examined motivational and selfregulatory learning factors that appear to mediate the relation between ASC and academic outcomes (e.g., academic intrinsic motivation, academic extrinsic motivation, effort regulation; Areepattamannil, 2012; Guay, Ratelle, Roy, & Litalien, 2010; Liu, Carmen, & Yeung, 2015), to the authors’ knowledge no study to date has examined factors that may underlie academic disidentification by testing a hypothesized path model. Thus, the present study seeks to contribute to this gap in knowledge by proposing a psychosocial variable that may partially explain disidentification and have a unique influence on Black males: teacher trust. Attributional Mechanism


7

Attribution theory in educational research assumes individuals attempt to explain the academic outcomes and behaviors of themselves and others. The theory categorizes causal attributions based on locus (i.e., internal or external), stability (i.e., stable or unstable), and controllability (i.e., controllable or uncontrollable; Weiner, 1986). Several researchers have theorized possible causal mechanisms underlying disidentification in Black students. Major and Schmader (1998) considered two possibilities: 1) reduction of the centrality or importance of the domain or 2) discounting of the “diagnosticity,” or validity, of their academic feedback and, as such, rejecting the feedback as an accurate indicator of their competency. In this study, we suggest a framework consistent with Major and Schmader’s (1998) second hypothesis: Black students who disidentify are disregarding the academic feedback they receive as accurate indicators of their academic and intellectual ability, which may impact academic motivation and achievement. Thus, it is proposed the weak relation between academic self-concept and academic outcomes found in those who exhibit academic disidentification may be related to an attribution of their academic outcomes to stable, external causes—specifically, teachers whom they do not trust. As such, we suggest these students disregard feedback from faculty as indicative of their true academic ability. Prior evidence lends support to our hypotheses, as Black students are found to make more external attributions compared to their White counterparts: Crocker, Voelkl, Testa, and Major (1991) reported when Black college students performed poorly on an experimental task, they were more likely to blame teachers’ bias (i.e., an external attribution). Stable, external attributions have been linked to reduced academic motivation (Graham, 1991). Further, Crocker et al. (1991) reported Black college students’ self-esteem was buffered by an external attribution of evaluator bias when given negative feedback by a White individual on an experimental task.


8

Interestingly, Black students were more likely than White students to attribute both positive and negative feedback to evaluator prejudice. As such, their self-esteem was not impacted by evaluator feedback, which was likely disregarded. Moreover, research by Yeager and colleagues (2014) suggests using the lens of attribution theory can allow scholars to partially explain poorer performance among Black students and successfully intervene for those low in school trust. The authors hypothesized that in ambiguous situations students low in school trust might perceive critical feedback as reflective of teacher bias. The authors demonstrated that particularly for Black students low in school trust, students’ perception of teacher feedback had a significant effect on academic outcomes. In particular, in the authors’ third in a series of experimental studies, a brief intervention was conducted to shift high school students’ attributions to critical feedback of their academic work. Students in the experimental group were encouraged to attribute such feedback to teachers’ high standards and teachers’ belief that the student can meet these standards. The students receiving this intervention showed significantly improved grades over the course of the semester, such that the achievement gap between Black and White students in this study reduced by 39%. Teacher Trust. The importance of teacher-student relations is not unique to secondary school students, as faculty-student relationships have been found to be an important part of the college experience, impacting academic outcomes (Gloria & Robinson Kurplus, 2001; Maton et al., 2011). However, for Black students, particularly those attending predominantly White colleges and universities (PWCUs), mistrust and social distance between Black students and White faculty may create barriers between student-faculty relationships (Cole & Griffin, 2013). We suggest teacher trust, an aspect of student-faculty relationships, is an important construct that may influence academic achievement in Black college students.


9

Scholars suggest interpersonal trust is important when individuals are in interdependent relationships, relying on one another to achieve their goals (Hoy & Tschannen-Moran, 1999). Such relationships are evident in schools, where student outcomes are dependent on evaluation by instructors. Teacher trust is defined as a student’s trust toward his/her instructor, as evidenced by a student’s perception that instructors embody benevolence, honesty, reliability, openness, and competence (Adams & Forsyth, 2009). Although few studies have explored the impact of students’ trust of teachers on academic achievement, recent research suggests higher teacher trust is related to better academic outcomes (Romero, 2015; Yeager et al., 2014). While research has found Black adolescents report lower levels of teacher trust (Adams, 2010; Zirkel, 2005; Yeager et al., 2014), no study to date has examined whether Black college students’ teacher trust is related to academic outcomes. Purpose of the Study The primary purpose of the present study is an exploratory investigation of the role of teacher trust and gender in the academic disidentification of Black collegians. We theorize that teacher trust plays a significant role in the academic disidentification of Black college students, particularly for males. To examine this phenomenon, we proposed a path model examining the relations between age, teacher trust, ASC, and GPA in a sample of Black male and female collegians. Using attribution theory and prior research as a guide, we tested the following hypotheses: 1. Males will report significantly lower GPAs than females, consistent with prior research (Cokley, 2001; Cokley & Moore, 2007), and significantly lower teacher trust. Research suggests Black males are more likely to perceive discrimination (Fischer & Shaw, 1999)


10

and report differential treatment by teachers (Marcus, Gross, & Seefeldt, 1991). Thus, it seems such perceptions may contribute to decreased teacher trust in Black males. 2. Age will have a significantly negative relation with teacher trust among males, but a significantly positive relation among females. With older students’ increased interaction and experience with faculty, the aforementioned gendered perceptions of discrimination and teacher relationships may decrease teacher trust in older males, while allowing increased trust among females. 3. Age will have a significantly positive relation with ASC among males, but a nonsignificant relation among females. The literature suggests academic disidentification may develop over time and uniquely impact Black males (Cokley et al., 2011). Thus, while we expect females’ ASC to remain constant with age, we expect males’ ASC will increase with age as a protective mechanism regardless of whether GPA increases. 4. The relation between ASC and teacher trust will be significantly positive and weaker among males than females. As we theorize that males may perceive faculty as less trustworthy than females, we predict males’ ASC may be less related to trust. 5. The relation between teacher trust and GPA will be significantly positive and weaker among males than females. Research suggests trust may play a particularly important role in the academic achievement of Black adolescents (Yeager et al., 2014). Given males’ more negative experience in school (Wood, Kaplan, & McLoyd, 2007), perceptions of faculty as untrustworthy may lead to more maladaptive reactive coping mechanisms as compared to females. Thus, the relation between teacher trust and academic outcomes may be stronger for males.


11

6. Teacher trust will serve as a significant mediator of the relation between ASC and GPA and the indirect effect of ASC on GPA through teacher trust will be stronger among males than females. Attribution theory provides a basis to expect decreased teacher trust may result in a buffered ASC despite poor grades, such that students low in teacher trust may attribute academic outcomes to stable, external conditions that are out of their control. Thus, in line with one of Major and Schmader’s (1998) hypotheses, we suggest disidentification represents students’ disregard of academic feedback as indicative of their true academic abilities. Again, given Black males’ increased perceptions of racism and unique experiences in schools, we theorize male students will be significantly more likely to have their ASC buffered by a lack of teacher trust. Method Participants and Procedures Our sample consisted of 319 self-identified Black college students, including 120 males and 199 females. Participants ranged in age from 18 to 28 years (M = 20.12, SD = 1.86, median = 20.00). There were 81 freshman, 74 sophomores, 71 juniors, and 91 seniors. All participants included in the sample identified as Black: 89.7% Black, 10.0% biracial, and 0.3% other (i.e., Nigerian-American). Participants reported a mean grade point average of 2.87 (SD = 0.60), ranging from 1.0 to 4.0. Participants were recruited from a large, Southwestern university during the course of a spring semester and into the summer session. Convenience sampling occurred via recruitment through the subject pools of the Educational Psychology and Psychology departments, as well as recruitment postings to a university website and snowball sampling. Students participated in the study by completing a series of online questionnaires via Qualtrics. Prior to participating, all


12

participants were provided informed consent and told they could refuse to answer any question. Data were collected as part of a larger study and participants also completed inventories of parental attitudes and socialization. All participants of department subject pools received course credit in exchange for participation and participants outside of these subject pools were entered into a series of drawings for $50 Visa gift cards. Approval was granted for this study by a university institutional review board. Measures Academic Self-Concept Scale (ASCS; Reynolds, Ramirez, Magrina, & Allen, 1980). The ASCS is a 40-item Likert-type scale with response options ranging from 1 (strongly disagree) to 4 (strongly agree). The ASCS is designed as a measure of an academic facet of general self-concept in college students (Reynolds, 1988). The current study computed a mean score of all 40-items. Studies consisting of Black samples report adequate reliability: .90 (Awad, 2007), .91 (Cokley, Komarraju, King, Cunningham, & Muhammad, 2003). Construct validity with Black samples has also been demonstrated with reported positive correlations between ASC and both GPA and self-esteem (Cokley, 2002). Items are keyed positively, with higher scores representative of a more positive academic self-concept. In this study, a Cronbach’s alpha of .92 was obtained. Teacher Trust Scale (Adams & Forsyth, 2009). The teacher trust scale is a 13-item Likert-type scale with response options ranging from 1 (strongly disagree) to 5 (strongly agree). This scale is designed as a measure of students’ perceptions of teachers’ trustworthiness. Internal consistency (Cronbach’s alpha) was reported to be .90 (Adams & Forsyth, 2009) and .83 (Adams, 2010) for the scale. Internal structure validity has been supported by an exploratory factor analysis with factor loadings ranging from .62 to .85 (Adams, 2010). Construct validity


13

has been supported by factor analysis, showing an alignment of items with the discernment facets of trust (Hoy & Tschannen-Moran, 1999), as well as concurrent and predictive validity measures (Forsyth, Adams, & Hoy, 2011). Although this instrument has not been used with a homogenous Black sample, adequate reliability and validity has been reported with a racially diverse sample including a 32% Black makeup (Adams & Forsyth, 2009). Items are keyed positively, with higher scores representative of more teacher trust. For the purposes of the current research, items were slightly altered to make language more appropriate for use with a college sample. The Cronbach’s alpha for this study was found to be .89 for this scale. Demographic Questionnaire. Demographic information included, race/ethnicity, gender, year in school, socio-economic status, mother’s/father’s highest level of education, and self-reported cumulative GPA. Data Analysis Path models were fit and all variables in the model were treated as observed variables. Analyses were conducted via SPSS AMOS version 21 and the covariance matrices were analyzed using maximum likelihood estimation procedures. The following were used as criteria for acceptable model fit: a non-significant chi-square statistic (χ²), confirmatory factor index (CFI) > .90, Standardized Root Mean Square Residual (SRMR) < .10, root mean square error of approximation (RMSEA) < .08, and a lower Akaike Information Criterion (AIC) as indicative of a better fitting model (Keith, 2015). Additionally, the comparison between two nested models was tested through significant ∆ χ² value, and gender invariance of paths was tested using multigroup moderation analysis. Mediation analysis was conducted using bias-corrected bootstrapping (Preacher & Hayes, 2008), computing the means of 5,000 estimated indirect effects by creating


14

5,000 bootstrap samples using random sampling with replacement. Significant mediation effects are detected if the 95% confidence intervals (CIs) of the indirect effects do not include zero. Results Data Screening and Preliminary Analysis To compute missing data for cases on a given subscale, we used a subscale-mean substitution procedure. Intercorrelations among all variables by gender are presented in Table 1. A multivariate analysis of variance (MANOVA) was conducted on the four predictor variables and one criterion variable to examine mean differences by gender. A significant main effect for gender was found, F (4,309)=4.54, p < .01, λ = .94, η2 = .06. Consistent with our hypotheses, significant differences were found for GPA and teacher trust. Males reported significantly lower GPAs and significantly lower levels of trust. No significant gender differences were found for age or ASC. Table 2 includes the results of the MANOVA. Testing the Hypothesized Path Model The primary aim of this study was to test the hypothesized links between age, teacher trust, ASC, and GPA. The following steps were conducted: 1) an unconstrained multi-group model across gender was examined in which the same pattern of structural paths was tested without constraints across groups; and (2) a constrained multi-group model was examined, where structural paths were constrained to be equal across groups. Analysis of the unconstrained multigroup model indicated this model fit the data adequately: χ² = 5.37, CFI = .98, SRMR = .031, RMSEA = .07. It also provided a significantly better fit to the data (AIC = 41.37) than the constrained multi-group model (AIC = 55.96, ∆ χ² = 24.59, ∆ df = 5, p < .001). This significant chi-square difference test indicates one or more structural paths are different across gender. When this statistic is significant, the model with the smaller chi-square is chosen (Yuan &


15

Bentler, 2004). Therefore, we selected the baseline model as the final model. A summary of fit statistics is presented in Table 3. The final model for males accounted for 28% of variance in GPA, 17% of variance in trust, and 4% of variance in ASC. The final model for females accounted for 27% of variance in GPA, 17% of variance in trust, and 0% variance in ASC. Test of Direct Effects: Path Coefficient Comparison by Gender All paths were found to be significant in the hypothesized path models across gender, with the exception of females’ path from age to ASC. By testing structural invariance, we were able to see how each path coefficient differed by gender. First, age significantly predicted ASC for males only: older males were more likely to report higher ASC, while age did not significantly predict ASC for females (males, β = .21, p < .05; females β = .02, p > .05). Second, age significantly predicted trust for both groups: older males reported lower levels of trust (β = .23, p < .01), while older females reported higher levels of trust (β = .19, p < .01). Third, ASC significantly predicted trust in both groups: higher ASC was related to higher trust in both males and females (males, β = .40, p < .001; females β = .35, p < .001). Fourth, trust significantly predicted GPA in both groups: males and females with higher trust reported higher GPAs (males β = .42, p < .001; females β = .20, p < .01). Finally, ASC significantly predicted GPA in both groups: males and females with higher ASC reported significantly higher GPAs (males, β = .20, p = .01; females β = .41, p < .001). These results support our hypotheses. Standardized coefficients for males and females are shown in the baseline model depicted in Figure 1. Multi-group moderation analysis was conducted using a sequential constraints approach in order to test whether paths were significantly variant by gender. Individual pathways were tested for invariance by successively constraining each path to be equal across groups to locate the variances in the model. Corresponding chi-square difference tests were then used to


16

determine whether gender significantly moderated the paths. Results revealed that gender moderated the relations between age and trust (∆ χ²[1] = 15.02, p < .001), age and ASC (∆ χ²[1] = 3.86, p < .05), trust and GPA (∆ χ²[1] = 6.98, p < .001, and ASC and GPA (∆ χ²[1] = 4.85, p < .05). These findings further supported our hypotheses. We did not find support for our hypothesis that gender differences would exist in the relation between ASC and trust, as this path was found to be statistically invariant (∆ χ²[1] = 0.92, p > .05). Test of Indirect Effect Finally, we examined our mediation hypothesis. Results indicate trust was a significant partial mediator of the relation between ASC and GPA for both males and females: males β = .17, p = .001, CI = .097-.250, S.E. = .045, t = 3.78; females β = .07, p < .001, CI = .003-.115, S.E. = .025, t = 2.80. As hypothesized, this indirect effect was significantly stronger for males (b = .32) than females (b = .09), t(317) = 2.71, p < .01. Test of Alternative Models Three alternative models were also tested. First, we considered the possibility that age may directly predict GPA, and GPA may predict trust (i.e., age may have an indirect effect on trust through GPA). Testing this alternative model yielded the following results: χ² = 43.81, CFI = .79, SRMR = .068, RMSEA = .10, AIC = 61.81. Second, we considered the possibility that GPA may predict ASC and thus have an indirect effect on trust. Testing this alternative model yielded the following results: χ² = 43.81, CFI = .792, SRMR = .068, RMSEA = .10, AIC = 61.81. Third, we considered the possibility that trust may predict ASC and thus may have both a direct and indirect effect on GPA. Testing this alternative model yielded the following results: χ² = 43.29, CFI = .80, SRMR = .061, RMSEA = .10, AIC = 61.30. Each alternative model yielded a poorer fit than our hypothesized baseline model.


17

Discussion This study sought to gain a better understanding of academic disidentification among Black collegians. While several studies have documented the presence of academic disidentification over time in Black collegians (Cokley, 2002; Hope et al., 2013), few empirical studies build on this knowledge by examining factors that may underlie this phenomenon. The goal of this study was to hypothesize a path model that might explain whether teacher trust underlies academic disidentification, defined as a weak relation between ASC and GPA, and whether this mechanism differs based on students’ gender. Results of the study largely supported our hypotheses. In line with prior research, results suggested the relation between ASC and GPA was significantly weaker for Black males than females (Cokley, 2002; Cokley et al., 2011). A goal of this study was to examine why this may be the case. Using attribution theory as a theoretical lens, teacher trust was examined as a critical variable that might partially explain this phenomenon (i.e., a partial mediator of the relation between ASC and GPA). Our path model included the variable age, as research suggests that disidentification is a process that occurs over time (Osborne, 1997). Our results demonstrated that while age did not predict females’ perceptions of their own academic ability (ASC), older females were more likely to trust their faculty. On the other hand, males’ ASC was shown to increase with age, despite their significantly lower academic performance compared to females, while older males also reported less trust of their faculty. In other words, older males are feeling better about their academic capabilities, while trusting their professors less. This finding has important implications for our theory that the buffering of ASC in the face of poor academic outcomes may be the result of older male students discounting the feedback of their instructors. Males and


18

females’ contrasting association between age and trust may be particularly impactful in reducing the relation between ASC and GPA for males, while allowing females to maintain strongly identified with academics. As predicted, trust had a significantly stronger effect on GPA for males than females; the relation between ASC and GPA was partially mediated by teacher trust; and this relation was moderated by gender, such that the indirect effect of teacher trust was significantly stronger for males than females. Thus, it appears teacher trust may partially explain why males more consistently experience academic disidentification. Also of importance, gender differences in mean scores were found for both GPA and teacher trust. Not only is teacher trust functioning differently for Black males and females (i.e., trust has a stronger influence on males’ GPA than females’), but also males are less likely to trust their professors. It appears Black male collegians’ relationship to teacher trust may create the perfect storm allowing for the maintenance of their ASC in the face of poorer academic outcomes. Our findings failed to support the hypothesis that the relation between ASC and teacher trust would be significantly weaker among males than females. Although males appear to perceive faculty as less trustworthy than females, this did not seem to impact the relation between ASC and teacher trust differently across gender. Given that few studies have examined teacher trust in Black collegians, further research is needed to shed light on factors that predict trust and how this might vary across gender. This study makes a significant contribution to the literature in beginning to build a strong theoretical basis via attribution theory for why academic identification may occur in Black collegians and how teacher trust may partially explain this phenomenon. One possible explanation for our findings is that students who have a tendency to distrust their instructors may


19

have a tendency to attribute their academic outcomes to teacher bias. This would be an example of an external, stable, and uncontrollable attribution, which research suggests is a formula for poor academic motivation (Graham, 1991). Thus, for such students, a reduced relation between ASC and academic outcomes might represent students’ disregard for academic feedback from teachers as indicative of their true academic and intellectual abilities. Such reasoning is in line with Major and Schmader’s (1998) hypothesis regarding the cause of academic disidentification. It may be that Black males weigh trust more heavily than females. Potentially, trust may be more important to academic motivation for students who are particularly marginalized or stereotyped in academic domains, as a lack of trust may have a greater tendency to foster the belief that their effort will not result in better academic outcomes (Yeager et al., 2014). While Black males and females both face challenges in the classroom at PWCU’s, Black males face unique educational challenges (i.e., stereotypes, implicit bias, increased risk of punishment for behavioral problems), often undermining the privilege that would typically come with being male (Thomas & Stevenson, 2009). While some might argue such issues are of less concern because such students were successful or resilient enough to reach post-secondary education, successful persistence in college is influenced by perceptions of the campus racial climate (Locks, Hurtado, Bowman, & Oseguera, 2008). Further, it seems plausible that past experiences in education may continue to impact Black male students’ potential for academic success as they transition to college. The stress inherent in this transition or, for some students, being in a very different racial-ethnic environment for the first time, may heighten the importance of interpersonal relationships and trust for academic motivation and outcomes. Scholars suggest such life transitions can cause unique crises around social identities such as race and gender (Cross & Cross, 2008). Given that


20

so little is known about the impact of teacher trust on academic outcomes, particularly among Black students, it is integral that future studies build on this knowledge, considering how gender appears to influence this relation. One concept that this study’s findings highlight is the influence of interpersonal relationships on academic motivation and achievement. Martin and Dowson (2009) stress the importance of relatedness to academic motivation and achievement. The scholars note that relationships, particularly feedback from significant others, have the ability to impact the types of attributions students are capable of making. For instance, teachers have the ability to influence such attributions via their relationship with the student and the student’s witnessing of the teacher’s relationship with other students. Research has found when instructors deliver feedback equivalent to students’ performance, students perceive an increased amount of control over academic achievement (Thompson, 1994). Conversely, if feedback given by teachers is inconsistent, students might feel a decreased sense of control. Limitations Despite the contributions of this study, its limitations must be addressed to better understand the applicability of these findings. As research has not explored all of the relations examined in this study, much of this research is exploratory in nature. However, all hypotheses are based on prior findings or a theoretical basis and analyses were conducted with the hope that this research would address a serious dearth of research in several areas of Black college student achievement. Further, while the goodness of fit criteria for our model were acceptable, some were lower than desirable (i.e., an RMSEA below .05 would indicate a strong, rather than adequate fit; Keith, 2015). Methodologically, causal inferences in this study cannot be made temporally (i.e.,


21

about directional relations occurring due to changes over time) as one could in a longitudinal study. Future research would benefit from collecting data when students are in their first-year and recollecting data in the fourth year, as a longitudinal methodology would provide the strongest evidence for change in the relation of ASC and GPA over time. Theoretically, this study uses attribution theory as a lens to understand academic disidentification. However, this study did not utilize methods intended to measure the kinds of attributions students were making. Speculation that attributions of teacher bias contribute to the outcomes of this study is theoretical in nature. As such, future research might benefit from including measures intended to capture academic outcome attributions to better understand how attributions might play a role in academic disidentification. In examining the sample in this study, all participants are recruited from a large, southwestern, predominantly White university. Future research must be conducted to determine whether such results can be generalized to students in primary or secondary education, other regions, and historically black colleges/universities. Finally, this sample consisted entirely of Black students. Thus, our results may not generalize to students to other racial-ethnic groups. Future research should be conducted to better understand how these variables might impact students of other backgrounds who disidentify with academics. Implications Practitioners in Higher Education The present study has noteworthy theoretical and practical implications for university administrators, faculty, and student affairs professionals. The finding that for males, in particular, teacher trust contributes to the buffering of academic self-concept in the face of poorer academic outcomes has important practical implications. From a clinical and educational intervention perspective, there are several important points to consider that are relevant to the findings of this


22

study. First, it should not be assumed that Black students who appear apathetic about academics devalue education. Rather, it may be such students experience heightened levels of teacher mistrust, which impact their levels of identification with academics. For such students at PWCU’s, in particular, interventions should focus on acknowledging the presence of racial stress and building trusting, positive relationships among students and faculty. Given the role of teacher trust in academic disidentification, future research should seek to examine factors influencing students’ teacher trust. When considering the impact of teacher trust, educators might wonder how they can gain the trust of racial-ethnic minority students in the face of racial stress and the reality of racial discrimination. Scholars have suggested using “wise” strategies and interventions seeking to convey to students that they are not being judged on the basis of stereotypes, but appreciated as capable individuals (Cohen & Steele, 2002; Yeager et al., 2014). Yeager and colleagues demonstrated psychological interventions could decrease the negative impact of mistrust on academic outcomes, as they found encouraging students to perceive they are not being reduced to stereotypes by instructors improved achievement, particularly for Black students’ with chronically low trust. They distinguish such interventions from instructors who simply overpraise mediocre work or assure students they are not prejudiced, as such behavior does not dispel negative stereotypes and may have a negative impact on academic outcomes (Croft & Schmader, 2012; Harber et al., 2012). Such interventions providing “attributional retraining” show great promise for effectively diminishing school mistrust. However, it also seems important to push instructors to actively examine and challenge their racial attitudes (Bentley-Edwards, Thomas, & Stevenson, 2013). Thus, providing university interventions challenging faculty to recognize racial stress, while also


23

utilizing more “stealthy” psychological interventions to alter the ways students make attributions about feedback may be particularly efficacious. It is important to acknowledge the unique experience of racial-ethnic minority students at PWCU’s and the reality of explicit and implicit racial bias, while seeking ways to provide students with coping methods that are productive to their academic achievement. Professional development seeking to explore topics of race with faculty may help reduce implicit bias (Bentley-Edwards et al., 2013), while also informing instructors on how their academic feedback may impact student attributional styles. Professors should understand the importance of teacher trust and, as such, seek to build trust by serving as allies to students. Interventions should also focus on connecting students with faculty mentors and building trusting, positive relationships among students and faculty. This study provides a basis for the continued study of teacher trust and gender as they relate to academic disidentification, which may inform interventions seeking to reduce the collegiate racial-ethnic and gender achievement gap. As we work toward reducing such disparities, it is clear that increasing enrollment of Black students in college is not enough. We must also seek to achieve the end goal of graduating Black students.


24

References Adams, C. (2010). Social determinants of student trust in high-poverty elementary schools. In W.K. Hoy & M. DiPaola (Eds.), Analyzing school contexts: Influences of principals and teachers in the service of students (pp. 254-280). New York: Information Age Publishing. Adams, C., & Forsyth, P. (2009). Conceptualizing and validating a measure of student trust. In Wayne Hoy and Michael DiPaola (Eds). Improving Schools: Studies in Leadership and Culture (pp. 263-279). Charlotte, NC: Information Age Publishing. Areepattamannil, S. (2012). Mediational role of academic motivation in the association between school self-concept and school achievement among Indian adolescents in Canada and India. Social Psychology of Education, 15(3), 367-386. doi: 10.1007/s11218-012-9187-1. Aud, S., Fox, M., and KewalRamani, A. (2010). Status and Trends in the Education of Racial and Ethnic Groups (NCES 2010-015). U.S. Department of Education, National Center for Education Statistics. Washington, DC: U.S. Government Printing Office. Awad, G. (2007). The role of racial identity, academic self-concept, and self-esteem in the prediction of academic outcomes for Black students. Journal of Black Psychology, 33, 188–207. doi: 10.1177/0095798407299513. Bentley-Edwards, K. L., Thomas, D. E., & Stevenson, H. (2013). Raising Consciousness: Promoting healthy coping among Black boys at school. In C. Clauss Ehlers, Z. Serpell & M. Weist (Eds.), Handbook of Culturally Responsive School Mental Health: Advancing Research, Training, Practice, and Policy (pp. 121-133). New York, Springer. doi: 10.1007/978-1-4614-4948-5_9. Cohen, G.L., & Steele, C.M. (2002). A barrier of mistrust: How stereotypes affect cross race mentoring. In J. Aronson (Ed.), Improving academic achievement: Impact of


25

psychological factors on education (pp. 305-331), Oxford, England: Academic Press. doi: 10.1016/B978-012064455-1/50018-X. Cokley, K. (2001). Gender differences among African American students in the impact of racial identity on academic psychosocial development. Journal of College Student Development, 42, 480-487. Cokley, K. (2002). Ethnicity, gender and academic self-concept: A preliminary examination of academic disidentification and implications for psychologists. Cultural Diversity and Ethnic Minority Psychology, 8(4), 378-388. doi: 10.1037//1099-9809.8.4.378. Cokley, K., Komarraju, M., King, A., Cunningham, D., Muhammad, A. (2003). Ethnic differences in the measurement of academic-self concept in a sample of African American and European American college students. Educational and Psychological Measurement, 63(4), 707-722. doi: 10.1177/0013164402251055 Cokley, K., McClain, S., Jones, M., & Johnson, S. (2011). A preliminary investigation of academic disidentification, racial identity, and academic achievement among Black adolescents. The High School Journal, 95(2), 54-68. doi: 10.1353/hsj.2012.0003 Cokley, K. and Moore, P. (2007). Moderating and mediating effects of gender and psychological disengagement on the academic achievement of African American college students. Journal of Black Psychology, 33(2), 169-187. http://dx.doi.org/10.1177/0095798407299512 Cole, D., & Griffin, K. A. (2013). Advancing the study of student-faculty interaction: A focus on diverse students and faculty. In Higher education: Handbook of theory and research (pp. 561-611). Netherlands: Springer. http://dx.doi.org/10.1007/978-94-007-5836-0_12


26

Crocker, J., Voelkl, K., Testa, M., & Major, B. (1991). Social stigma: The affective consequences of attributional ambiguity. Journal of Personality and Social Psychology, 60(2), 218-228. http://dx.doi.org/10.1037/0022-3514.60.2.218 Croft, A., & Schmader, T. (2012). The feedback withholding bias: Minority students do not receive critical feedback from evaluators concerned about appearing racist. Journal of Experimental Social Psychology, 48(5), 1139-1144. doi:10.1016/j.jesp.2012.04.010 Cross, W. E., & Cross, T. B. (2008). Theory, research, and models. In S. M. Quintana & C. McKown (Eds.), Handbook of race, racism, and the developing child (pp. 154-181). Hoboken, NJ US: Wiley. Fischer, A.R., & Shaw, C.M. (1999). Blacks’ mental health and perceptions or racist discrimination: The moderating effects of racial socialization experiences and self esteem. Journal of Counseling Psychology, 46(3), 395-407. http://dx.doi.org/10.1037/0022-0167.46.3.395 Forsyth, P. B., Adams, C. M., Hoy, W. K. (2011). Collective trust: Why schools can’t improve without it. New York: Teacher’s College Columbia University. Gloria, A. M., & Robinson Kurpius, S. E. (2001). Influences of self-beliefs, social support, and comfort in the university environment on the academic nonpersistence decisions of American Indian undergraduates. Cultural Diversity And Ethnic Minority Psychology, 7(1), 88-102. doi: 10.1037//1099-9809.7.1.88 Graham, S. (1991). A review of attribution theory in achievement contexts. Educational Psychology Review, 3(1), 5-39. doi: 10.1007/BF01323661.


27

Guay, F., Ratelle, C. F., Roy, A., & Litalien, D. (2010). Academic self-concept, autonomous academic motivation, and academic achievement: Mediating and additive effects. Learning and Individual Differences, 20, 644–653. doi: 10.1016/j.lindif.2010.08.001. Harber, K. D., Gorman, J. L., Gengaro, F. P., Butisingh, S., Tsang, W., & Ouellette, R. (2012). Students' race and teachers' social support affect the positive feedback bias in public schools. Journal of Educational Psychology, 104(4), 1149-1161. doi: 10.1037/a0028110. Hope, E. C., Chavous, T. M., Jagers, R. J., & Sellers, R. M. (2013). Connecting self-esteem and achievement diversity in academic identification and dis-identification patterns among black college students. American Educational Research Journal, 50(5), 1122-1151. doi: 10.3102/0002831213500333. Hoy, W.K., & Tschannen-Moran, M. (1999). Five faces of Trust: An empirical confirmation in urban elementary schools. Journal of School Leadership, 9(3), 184-208. Keith, T. Z. (2015). Multiple regression and beyond: An introduction to multiple regression and structural equation modeling, 2nd ed., New York: Routledge. Liu, E. S., Carmen, J. Y., & Yeung, D. Y. (2015). Effects of approach to learning and self perceived overall competence on academic performance of university students. Learning and Individual Differences, 39, 199-204. doi: 10.1016/j.lindif.2015.03.004. Locks, A. M. & Hurtado, S. & Bowman, N. A. & Oseguera, L. (2008). Extending notions of campus climate and diversity to students' transition to college. The Review of Higher Education, 31(3), 257-285. The Johns Hopkins University Press. http://dx.doi.org/10.1353/rhe.2008.0011.


28

Major, B, & Schmader, T. (1998). Coping with stigma through psychological disengagement. In J. Swim (Ed.), Prejudice: The target's perspective (pp. 219 241). San Diego, CA: Academic Press. doi: 10.1016/B978-012679130-3/50045-4. Major, B., Spencer, S., Schmader, T., Wolfe, C., & Crocker, J. (1998). Coping with negative stereotypes about intellectual performance: The role of psychological disengagement. Personality and Social Psychology Bulletin, 24(1), 34-50. doi: 10.1177 /01461 67298241003. Marcus, G., Gross, S., & Seefeldt, C. (1991). Black and White students' perceptions of teacher treatment. Journal of Educational Research, 84(6), 363-367. doi: 10.1080/00220671. 1991.9941817. Martin, A. J., & Dowson, M. (2009). Interpersonal relationships, motivation, engagement, and achievement: Yields for theory, current issues, and educational practice. Review of Educational Research, 79(1), 327-365. doi: 10.3102/0034654308325583. Maton, K. I., Wimms, H. E., Grant, S. K., Wittig, M. A., Rogers, M. R., & Vasquez, M. J. (2011). Experiences and perspectives of African American, Latina/o, Asian American, and European American psychology graduate students: A national study. Cultural Diversity and Ethnic Minority Psychology, 17(1), 68. http://dx.doi.org/10.1037/a0021668. McMillian, M. M., Carr, M., Hodnett, G., & Campbell, F. A. (2015). A longitudinal study of academic identification among African American males and females. Journal of Black Psychology. doi: 10.1177/0095798415603845. Osborne, J. W. (1995). Academics, self-esteem, and race: A look at the underlying assumptions of the disidentification hypothesis. Personality and Social Psychology


29

Bulletin, 21, 449 455. doi: 10.1177/0146167295215003. Osborne, J. W. (1997). Race and academic disidentification. Journal of Educational Psychology, 89(4), 728-735. http://dx.doi.org/10.1037/0022-0663.89.4.728. Osborne, J. W. (2001). Testing stereotype threat: Does anxiety explain race and sex differences in achievement? Contemporary Educational Psychology, 26, 291-310. http://dx.doi.org/10.1006/ceps.2000.1052. Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879–891. doi:10. 3758/BRM.40.3.879. Reynolds, W. M. (1988). Measurement of academic self-concept in college students. Journal of Personality Assessment, 52, 223–240. doi: 10.1207/s15327752jpa5202_4. Reynolds, W. M., Ramirez, M. P., Magrina, A., & Allen, J. E. (1980). Initial development and validation of the academic self-concept scale. Educational and Psychological Measurement, 40, 1013–1016. doi: 10.1177/001316448004000428. Romero, L. S. (2015). Trust, behavior, and high school outcomes. Journal of Educational Administration, 53(2), 215-236. http://dx.doi.org/10.1108/JEA-07-2013-0079. Steele, C. (1992). Race and the schooling of African-Americans. Atlantic Monthly, 269(4), 6878. Thomas, D. E., & Stevenson, H. C. (2009). Gender risks and education: The particular classroom challenges for urban low-income Black boys. Review of Research in Education, 33, 160180. doi: 10.3102/0091732X08327164. Thompson, T. (1994). Self-worth protection: Review and implications for the classroom. Educational Review, 46, 259–274. doi: 10.1080/0013191940460304.


30

US Department of Education (2015). Mission. Retrieved from www2.ed.gov/about/overview/focus/goals.html Weiner, B. (1986). An attribution theory of motivation and emotion. New York: Springer Verlag. doi: 10.1007/978-1-4612-4948-1. Wood, D., Kaplan, R., & McLoyd, V. C. (2007). Gender differences in the educational expectations of urban, low-income African American youth: The role of parents and the school. Journal of Youth and Adolescence, 36(4), 417-427. http://dx.doi.org/10.1007/s10964-007-9186-2 Yeager, D.S., Purdie-Vaughns, V., Garcia, J., Pebley, P., Master, A., Hessert, W., Williams, M. & Cohen, G.L. (2014). Breaking the cycle of mistrust: Wise interventions to provide critical feedback across the racial divide. Journal of Experimental Psychology: General. 143(2), 804-824. http://dx.doi.org/10.1037/a0033906 Yuan, K. H., & Bentler, P. M. (2004). On chi-square difference and z-tests in mean and covariance structure analysis when the base model is misspecified. Educational and Psychological Measurement, 64, 737–757. doi:10.1177/0013164404264853. Zirkel, S. (2005). Ongoing issues of racial and ethnic stigma in education 50 years after Brown v. Board. The Urban Review, 37(2), 107-126. doi: 10.1007/s11256-005-0004-4.

Running Head: ACADEMIC DISIDENTIFICATION IN BLACK COLLEGIANS Table 1 Intercorrelations for Study Variables Disaggregated by Gender Variable 1 2

31

3

4

1. GPA

1

.49**

.35**

-.05

2. ASC

.35**

1

.36**

.02

3. Trust

.49**

.35**

1

.20**

4. Age

.10

.21*

-.15

1

Note. Males: n = 120. Females: n = 199. GPA = Grade Point Average. ASC = Academic SelfConcept. Trust = Teacher Trust. The male sample results are below the diagonal; the female sample results are above the diagonal. *p