Abstract
This study examines the changes in self-efficacy among Latinos during the freshman year in a highly selective institution. Results indicate that gender differences exist during this period. Males rate themselves high in self-efficacy at the beginning of the year, while females rate themselves low. An interaction effect occurs at the end of the year, and the ratings of self-efficacy are similar for both males and females. Possible explanations for the interaction effect are discussed.
College student self-efficacy has been the topic of emerging research over the past decade. Current indicators suggest a strong correlation between high levels of self-efficacy and persistence (Chemers, Hu, & Garcia, 2001; DeWitz & Walsh, 2002; Torres & Solberg, 2001). Despite significant research in the area, little research has been conducted on the topic of self-efficacy and academic persistence for Latino students. This study addresses the topic of Latino self-efficacy and explores gender as a factor at an elite, predominantly White institution (PWI).
Self-efficacy theory posits that a target behavior will be produced if people believe they are able to organize behavioral, cognitive, and social skills in a manner that will produce a desired objective (Bandura, 1977, 1997). Self-efficacy appraisals are influenced in four ways: performance accomplishments (previous performance), vicarious experiences (observing others), social persuasion (suggestions from self or others), and emotional arousal (stress or anxiety; Bandura, 1977). Self-efficacy beliefs affect levels of confidence and approach/avoidance of an academic task as well as persistence and performance in academic domains (Bandura, 1997; DeWitz & Walsh, 2002;Solberg, O’Brien, Villarreal, Kennel, & Davis, 1993; Suarez-McCrink, 2002; Torres & Solberg, 2001). Social adjustment, psychological adjustment, satisfaction in college, and academic achievement are also related to self-efficacy (Bandura, 1986; DeWitz & Walsh, 2002; Solberg et al., 1998; Zimmerman, 1989).
When exploring what may influence persistence for Latino college students, it is important to examine social, cultural, and institutional factors (Gloria, Castellanos, Lopez, Rosales, 2005). Latino college students may experience cultural incongruity (Gloria & Pope-Davis, 1997; Gloria & Robinson-Kurpius, 1996; N. Rodriguez, Myers, Morris, & Cardoza, 2000) and a hostile campus racial climate, which may affect student attitudes, aspirations, and behaviors (Crosson, 1988; Hurtado, 1992). Latino students must also contend with invalidation (Rendon, 1994) and feelings of being undervalued by peers, faculty, and/or administration (Byrd-Gunnings, 1982; Fuentes, Sedlacek, & Westbrook, 1993; Hurtado, 1994). Subtle discrimination can also occur in the form of microaggressions (Sue et al., 2007), and has been found to occur in PWIs (Harwood, Hunt, Mendenhall, Lewis, 2012). Microaggressions can affect the campus racial climate (Yosso, Smith, Ceja, & Solórzano, 2009) and subsequently adjustment patterns for college students (Blume, Thyken, Lovato, & Denny, 2012). Such experiences are exacerbated in elite academic environments (Hurtado & Faye-Carter, 1997; Hurtado, Faye-Carter, & Spuler, 1996). Thus, Latino student self-efficacy appraisals may be negatively associated with the aforementioned factors.
Self-efficacy has been postulated to be an integral factor in Latino mental health and college student adjustment (Cervantes & Castro, 1985; Vega, Hough, & Miranda, 1985). Self-efficacy for Latino college students has been found to be positively associated with social and familial support (Solberg & Villarreal, 1997; Torres & Solberg, 2001) and negatively associated with academic stress (Solberg, Valdez, & Villarreal, 1994; Solberg & Villarreal, 1997).
Gender differences are reported among Latinos with regard to the experience in institutions of higher education. Research indicates that despite having higher levels of parental support (Ceja, 2004), Latina undergraduates experience higher stress levels than their male counterparts (Munoz, 1985). This may be indicative of various academic and familial roles that Latinas are expected to successfully maintain (Sy, 2006). Conversely, males have been found to report greater academic discouragement and racial discrimination (Lopez, 1995). However, a dearth of literature exists with regard to gender and self-efficacy among Latino college students (Gonzalez, Stoner, & Jovel, 2003).
Self-efficacy was defined in this study as the degree of confidence the students have in their ability to successfully perform college-related tasks in coursework, academic, and roommate domains (Solberg, O’Brien, Villarreal, Kennel, & Davis, 1993). The goal of this study was to examine gender differences in appraisals of self-efficacy among Latino students during the transition into a predominately White, elite institution.
Method
Site
The research site was a high selective, private university where the admission rate was indicated to be 13% relative to the number of applications the institution received. The undergraduate student body consisted of approximately 6,000 students. The racial makeup of the student body is 49% White/Caucasian, 21% Asian/Asian American, 4% African American/Black, 10% Mexican American/Latino, and approximately 16% Other.
Participants
Participants were Latino freshmen between the ages of 17 and 19 years and were identified via the University Registrar’s office. All participants from the entering class of 2000 were identified as Latinos using their self-identified ethnicity from the student body list.
Measure
College Self-Efficacy Inventory
The measure is a 19-item survey with three subscales: Coursework Efficacy, Roommate Efficacy, and Social Efficacy (Solberg, Hale, Villarreal, & Kavanagh, 1993). The measure was validated with a Hispanic population and subscales were found to have strong internal consistency and demonstrated good convergent and discriminant validity. Reliability was established for internal consistency. Coefficient alpha estimates were .93 for the total College Self-Efficacy Instrument and .88 for each subscale, respectively (Solberg, O’Brien, Villarreal, Kennel, & Davis, 1993).
Factor analysis with this sample indicated a three-factor solution that accounted for 64.3% of the variance. The first factor is Social Efficacy. Seven items loaded high on this factor. The factor loadings ranged from .820 to .577. The alpha reliability for this factor is .90. The second factor is Coursework Efficacy. Eight items loaded high in this factor. The factor loadings range from .823 to .682. The alpha reliability for this factor is .90. The third factor is Roommate Efficacy. Four items loaded high on this factor. The factor loadings range from .873 to .830. The alpha reliability for this factor is .88.
Survey Administration
The survey was administered 2 times to determine change. The first administration was performed utilizing 73 participants in the autumn quarter (Time 1). The sample consisted of 41 males and 32 females. The final administration was performed utilizing 95 participants during the spring quarter (Time 2). The sample consisted of 57 females and 38 males. Data collected during both survey administrations were used to analyze change. There were participants who completed only the first or second survey administration. Moreover, some of the surveys returned were incomplete, which is reflected in the varying degrees of freedom indicated in the results section.
Results
Time 1
ANOVAs were conducted on the self-efficacy subscales for gender. The Univariate results were significant. Males reported higher levels of Social Efficacy and Coursework Efficacy, Univariate F(1, 68) = 5.755, p = .019 and 4.090, p = .047, respectively.
Time 2
ANOVAs were conducted on the self-efficacy subscales for gender. The results indicated that males reported higher Roommate Efficacy than females, Univariate F(1, 43) = 6.292, p = .016.
Change Analysis
A repeated-measures ANOVA was conducted on the entire sample; however, the results were not significant. A repeated-measures ANOVA was conducted while selecting students based upon gender. Males reported a decrease in Social Efficacy, Univariate F(1, 20) = 8.422, p = .009. Males reported a decrease in Coursework Efficacy, Univariate F(1, 20) = 3.904, p = .062; whereas females reported an increase, Univariate F(1, 24) = 6.653, p = .016.
Interaction effects were found for both Coursework Self-Efficacy and Social Self-Efficacy in regard to gender. Males rated themselves higher on Coursework Efficacy at Time 1 and report lower levels at Time 2. Females report the opposite trend, Univariate F(1, 44) = 10.266, p = .003. The same trend was found for Social Efficacy. Males at Time 1 reported higher levels of Social Efficacy than at Time 2, while female scores demonstrated an increase, Univariate F(1, 45) = 6.229, p = .016.
Discussion
The interaction effect with gender and time on Coursework and Social Efficacy was the most salient finding of this study. Males reported higher levels of Social and Coursework Efficacy at the beginning of the academic year. Toward the end of the freshman year, males reduced their levels of reported efficacy, while females reported an upward trend, which resulted in similar levels of self-efficacy in these respective domains. These results were consistent with the research of Mednick and Thomas (1993), which show that, among beginning college students, males tend to overestimate and females tend to underestimate their academic abilities.
Students admitted into highly selective institutions are likely to have a positive academic self-efficacy. However, there may be factors that affect Latinos differentially by gender, which may have influenced lower appraisals of self-efficacy at the beginning of college for Latinas. Latinas have been found to experience negative feedback with regard to college attendance due to familial obligations and gender role stereotyping (Gandara, 1994; A. Rodriguez, Guido-DiBrito, & Torres, 2000). Moreover, Latinas experience discrimination based upon race and gender, which may foster negative perceptions of their academic ability (Reyes, 2001; Rodriguez, Myers, Morris, & Cardoza, 2000). Emotional arousal may be higher for Latina college students as well as they have been found to report higher levels of stress and anxiety (Chacon, Cohen, & Strover, 1986; Munoz, 1985), which affects self-efficacy appraisals (Bandura, 1977).
Values prior to matriculation may differentially affect the way in which the college environment is experienced by gender (Arbona & Novy, 1991; N. Rodriguez, Myers, et al., 2000). Latinas have been found to be socialized toward modesty, whereas males have been socialized toward machismo (Arredondo, 1991) or excessive masculinity. While this explanation relies on stereotypes that do not apply to all Latinos (Murillo, 1976), it applies more to Latino families who have traditional values (Shirley, 1981; Vasquez-Nuttall & Romero-Garcia, 1989). However, high-achieving Latinos have been found to come from families whose parenting style is nonauthoritarian (Gandara, 1982, 1994). Therefore, the effect of familial socialization on appraisals of self-efficacy among high-achieving Latinos is an important predictor of how self-efficacy develops during the first year of college.
During the year, Latinas became more efficacious over time. The most stressful time of the transition into college is during the first few months (Garg, 1992; Tinto, 1975, 1988). Clearly, after the initial stress of leaving home and entering a highly selective institution, Latinas appraised their self-efficacy at a higher level.
Males report a decrease in self-efficacy over the course of the first year of college attendance. When examining the causes of this decline, it is important to note that Latino males have been found to experience more difficulty with academic and social integration in comparison with their female counterparts (Stage, 1989). Similar results have been found with regard to African American males (Fleming, 1985). These findings indicate that minority males have more negative experiences on predominately White college campuses. These experiences may be overt or may be covert as in the form of microaggressions. Blume et al. (2012) found that Latino males are more likely to experience heightened stress due to microaggressions, which have an effect on their attitudes, aspirations, and behaviors. Furthermore, microaggressions have an almost immediate effect on the individual and their self-efficacy (Sue et al., 2007). In addition, microaggressions lead to further alienation and isolation on college campuses (Yosso et al., 2009). Therefore, the downward trend in reported self-efficacy for males in this elite, predominately White university context can be explained in part by a reaction to microaggressions.
Conclusion
It is clear that gender differences exist in self-efficacy appraisals for Latino college students during the first year of university attendance. However, a few limitations must be acknowledged. While most of the findings were statistically significant, it is possible that a larger sample size would have produced a more robust significance. Another limitation to this study was the self-selected nature of the sample. Out of approximately 540 Latino students, only 95 participated. While results would have been more significant with a larger sample, it would have not been possible to require participation from the total Latino population at the chosen university. Therefore, there may have been a systematic difference with regard to those who chose to participate in the study. In future research, it may be beneficial to survey students from a variety of similarly elite universities. Despite the limitations to the study, the findings of this study may be useful for helping student development personnel, faculty, staff, and administration in addressing intervention strategies for incoming Latino freshmen as they encounter obstacles in higher education. Programs can be designed to promote self-efficacy and student success during the first year and should consider the effect of gender and self-efficacy appraisals over time, so as to guide students toward positive academic and social outcomes.
Footnotes
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.
