College Self-Efficacy and Academic Performance

in Mexican American Undergraduates

by

Marvyn R. Arévalo Avalos

A Thesis Presented in Partial Fulfillment of the Requirements for the Degree

Master of Arts

Approved May 2017 by the Graduate Supervisory Committee:

Lisa Spanierman, Chair

Lisa Flores Terence Tracey

ARIZONA STATE UNIVERSITY

December 2017

i

ABSTRACT

Grounded in Social Cognitive Career Theory (SCCT; Lent, Brown, & Hackett,

1994, 2000), the current study examines environmental and person-cognitive variables as

predictors of academic performance among a sample of 194 Mexican American

undergraduate students. Specifically, this study used multiple regression analysis to test

the associations between college self-efficacy (course self-efficacy and social self-

efficacy), proximal contextual influences (campus climate and cultural fit), and gender on

the academic performance (self reported grade point average, GPA). Results indicated

that course self-efficacy was a significant predictor of academic performance for

Mexican American undergraduate students. In addition, social self-efficacy, positive

perceptions of the campus climate, and cultural fit were associated with high self-

efficacy. This study contributes to our knowledge of college student development in

general, and academic attainment among Mexican Americans specifically. Practice and

research recommendations are discussed.

Keywords: College Self-Efficacy, Mexican Americans, University Environment,

Cultural Fit

ii

TABLE OF CONTENTS

Page

LIST OF FIGURES …………………………………………………………………….. iii

LIST OF TABLES ……………………………………………………………………… iv

INTRODUCTION ………………………………………………………………………. 1

Theoretical Framework ………………………………………………………….. 1

Self-Efficacy Beliefs …………………………………………………………….. 2

Contextual Influences …………………………………………………………… 3

Campus Climate and Cultural Fit ……………………………………………..… 6

The Current Study ……………………………………………………………..…8

METHOD .……………………………………………………………………………... 10

Participants ………………………………………………………………….…. 10

Measures …………………………………….……………………………….… 10

RESULTS …………………………………………………………….………………... 15

Preliminary Analysis …………………………………………………………… 15

Confirmatory Factor Analysis: College Self-Efficacy ………………….……… 16

Confirmatory Factor Analysis: University Environment Scale ........................... 17

Confirmatory Factor Analysis: Cultural Congruity Scale ................................... 17

Multiple Regression Moderation Analysis …………………………………..… 18

DISCUSSION ………………………………………………………………………….. 20

Implications for Practice and Research …….………………………………....... 22

Limitations and Conclusion …………………………………………….....…… 24

REFERENCES……………………………………………………………….………… 26

iii

LIST OF FIGURES Figure Page

1. Social-Cognitive Career Theory ……………………………………………….. 31

2. Course Self-Efficacy – Model 1 ……………………………………………..… 32

3. Social Self-Efficacy – Model 2 ………………………………………..………. 33

iv

LIST OF TABLES

Table Page

1. Descriptive Statistics and Correlations ………………………………...........…. 34

2. Descriptive Statistics by Institution ..................................................................... 35

3. Self-Efficacy CFA Results ………………………………...……....................… 36

4. Campus Climate CFA Results ............................................................................. 37

5. Cultural Fit CFA Results ..................................................................................... 38

6. Course Self-Efficacy Model 1 Results ………………………………….....…… 39

7. Social Self-Efficacy Model 2 Results ………………………………..………… 40

1

INTRODUCTION

Latinos/as educational attainment has increased exponentially over the last 40

years (Fry & Lopez, 2012) and in recent years this group has become the largest

racial/ethnic minority in college enrollment (Krogstad, 2016). Yet, compared to White,

Asian American, and Black individuals – Latinos/as fall behind in terms of graduating

with a bachelor’s degree. For example, 22% of Blacks, 36% of Whites, and 54% of Asian

adults 25 years or older had a bachelor’s degree or higher in 2015, compared to 15% of

Latinos/as (Ryan & Bauman, 2016). This statistic is concerning given Latinos/as are

projected to represent nearly one third of the total U.S. population by 2060 (Colby &

Ortman, 2015). Because adults with a college degree are more likely than others to be

employed, earn higher wages, have healthier lifestyles, have decreased healthcare costs,

and engage in more civic participation (Baum, Ma, & Payea, 2013), it is important to

promote Latinos/as academic attainment. Grounded in Social Cognitive Career Theory

(SCCT; Lent, Brown, & Hackett, 1994, 2000), in the current study I examine

environmental and person-cognitive variables as predictors of academic performance for

Latino/a undergraduates. Specifically, this study investigates the relations between

college self-efficacy, university contextual variables (i.e., campus climate and cultural

fit), and academic performance among a sample of Mexican American undergraduate

students.

Theoretical Framework

SCCT (Lent, Brown, & Hackett, 1994, 2000) provides a roadmap to

understanding academic and career processes via the examination of person, contextual,

and experiential factors. Based on the work of Bandura’s (1986) social cognitive theory,

2

SCCT posits that person characteristics (e.g., gender, race and ethnicity), social-cognitive

attributes (e.g., self-efficacy beliefs), behaviors (e.g., declaring a major), and contextual

influences (e.g., perceived academic support and barriers) interact to predict

educational/career choice goals, actions, and performance outcomes (e.g., GPA).

Contextual influences may manifest via distal or proximal contextual factors. Distal

contextual factors, such as gender socialization, mold a person’s learning environment

that in turn lead to the development of interests and self-efficacy beliefs. Proximal

contextual factors, such as perceived racism or sexism, can moderate and directly affect

the processes between educational/career interests, choices goals, and actions. Lastly,

SCCT suggests that performance outcomes are thus a function of and predicted by self-

efficacy beliefs, contextual influences, interests, choice goals, and actions. See figure 1

for Lent et al. (1994) SCCT model.

Self-efficacy Beliefs

A central tenet of SCCT is self-efficacy beliefs, which are defined as a person’s

belief in his or her abilities to complete specific tasks that lead to a desired outcome

(Bandura, 1977, 1986). Self-efficacy has received substantial attention in research

examining educational outcomes among college students. For example, using

predominantly White (52%-85%) undergraduate student samples, indicators of self-

efficacy have been found to predict academic performance (Chemers, Hu, & Garcia,

2001; Gore 2006), college satisfaction (DeWitz & Walsh, 2002), and college adjustment

(Ramos-Sánchez & Nichols, 2007). Similar results were also found in research conducted

with Latino/a undergraduates, where college self-efficacy positively predicted academic

performance (Aguayo, Herman, Ojeda, & Flores; 2011; Brady-Amoon & Fuertes, 2010).

3

In fact, meta-analysis results indicate that self-efficacy is a moderate predictor of

academic performance, persistence, and retention outcomes (Multon, Brown, & Lent,

1991; Richardson, Abraham, & Bond, 2012; Robbins et al., 2004). It is important to note

however, that self-efficacy is most useful as a predictor of outcomes when these

constructs are domain specific, rather than general. Constructs of self-efficacy that are

matched along key dimensions of the outcome variable, such as beliefs in one’s math

abilities and performance on math based tasks, yield the most significant results as

predictors of performance (Multon et al., 1991; Lent, Brown, & Gore, 1997; Pajares

1996).

Few studies have used SCCT framework to explain academic attainment of

college students. For instance, using SCCT among a predominantly White sample (3%

Latino), researchers found that college self-efficacy predicted academic persistence and

success after controlling for gender, ethnicity, and high school GPA (Wright, Jenkins-

Guarnieri, & Murdock, 2012). Among a sample of 457 Mexican American college

students attending a Hispanic-Serving Institution, Ojeda, Flores, and Navarro (2011)

found college self-efficacy to positively predict academic goal progress and academic

satisfaction. To this end, I will utilize SCCT to examine college self-efficacy as a

predictor of college GPA among Mexican American Undergraduate students.

Contextual Influences

SCCT hypothesizes that proximal contextual influences may moderate the

relationships between interests and goals and exert a direct influence on

career/educational goals and actions (Lent et al., 2000). In contrast to this theory, recent

findings suggest that proximal contextual variables do not have a direct effect on

4

interests, choices, and/or goals – but rather have an indirect effect on these variables via

self-efficacy beliefs. For example, Byars-Winston, Estrada, Howard, Davis, and Zalapa

(2010) found that perceived campus climate predicted academic self-efficacy beliefs,

which in turn predicted academic interests and goals among a sample of Black, Latino,

Southeast Asian, and Bicultural students in science and engineering. Similarly, in a multi

group comparison using three samples of White, Asian, and Latino/a college students,

Sheu, Mejia, Rigali-Oiler, Primé, and Chong (2016) found support for a fully mediated

model in which academic supports predicted academic self-efficacy, which in turn

predicted academic goal progress among their participants. The findings suggest that

students’ learning environment and development of self-efficacy beliefs are influenced

directly by proximal contextual factors, such as sense of belonging. It is possible that if

students’ learning environment were challenged by lack of educational/personal support

or perceived racism and sexism, their self-efficacy belief would be impacted negatively.

Thus, leading to adverse consequences in educational goals as self-efficacy beliefs are

directly linked to academic performance.

In an effort to improve academic attainment and college retention rates of

Latinos/as researchers have turned their attention toward examining contextual influences

that are associated with academic performance and persistence intentions. For example,

research has examined the protective role of ethnic identity and parental support as

predictors of GPA, finding that greater psychological and family resources are associated

with greater academic achievement (Ong, Phinney, & Dennis, 2006). In terms of gender

and ethnic identity, Barajas and Pierce (2001) found that Latinas use their positive view

of their ethnic identity and group membership as tools for navigating college; yet, Latinos

5

are not able to do that as well as their female counterparts. Similarly, Gloria and

colleagues have found positive associations between cultural congruity and psychological

well-being among samples of Latinas (Gloria, Castellanos, & Orozco, 2005) and Latinos

(Gloria, Castellanos, Scull, & Villegas, 2009). These associations, the authors note, can

help predict college persistence intentions and improved coping strategies.

Additional Latinos/as educational research has explored family and economic

factors in relation to academic performance and persistence. For example, among first

generation Latino/a immigrants, socioeconomic status and enculturation positively and

significantly predict academic outcomes; yet, gender did not have a significant effect

(Aguayo et al., 2011). Another study found using a sample of Mexican American

undergraduate men that Familismo positively predicted parental encouragement, which in

turn was positively associated with college persistence intentions (Ojeda, Navarro, &

Morales, 2011). Collectively, these studies suggest that for Latino/a students their cultural

background and family support experiences are influential in the process of pursuing

higher education.

These findings are congruent with SCCT framework, in that distal contextual

influences mold a person’s learning experiences, which in turn can impact self-efficacy

beliefs and interests. Yet, these studies offer limited information on the mechanisms for

how proximal contextual influences predict educational outcomes. Two proximal

contextual influences that have received significant attention in the academic attainment

literature with Latinos/as are perceptions of the university climate and cultural fit.

Examining the role of perceptions of the university environment and cultural fit in

6

predicting self-efficacy of Latinos/as can lead to the development of interventions that

enhance academic attainment of these students.

Campus Climate and Cultural Fit

Due to the importance of culture and support systems in educational endeavors,

another area of research involves examining how perceptions of the campus climate and

cultural fit predict academic outcomes. Campus climate refers to student perceptions of

their university environment and whether they feel welcomed, valued, and supported as a

racial/ethnic minority student (Gloria & Kurpius Robinson, 1996). Cultural fit is the

degree to which a student feels their values align with their university (Gloria & Kurpius

Robinson, 1996). Drawing from literature that examined the experiences of students of

color in predominantly White universities, Gloria and Kurpius Robinson (1996)

developed two measures for assessing Chicano/a undergraduate students’ perceptions of

the campus climate and cultural fit.

Given the U.S. history of racial discrimination and anti-immigrant it is important

to consider the contextual factors that inform Latino/a student educational experiences,

specifically non-persistence decisions or poor performance. Scholars note that institutions

may hold the values of the dominant culture and unintentionally create an unwelcoming

or discriminatory environment for Latino/a students (Bordes & Arredondo, 2005; Castillo

et al., 2006; Gloria & Robinson Kurpius, 1996; Gloria, Castellanos, & Herrera, 2016).

Thus, understanding the interplay of personal variables, such as gender or self-efficacy

beliefs, within the larger educational context is of utmost importance for the development

of interventions that promote educational attainment among Latinos/as.

7

For Latinos/as, researchers found that positive perceptions of the campus climate

are associated with positive measures of academic success (Gloria et al., 2016); college

persistence intentions (Anguinaga & Gloria, 2015; Castillo et al., 2006; Gloria &

Kurpius, 1996); and sense of belonging and academic self-efficacy (Gloria, Castellanos,

Scull & Villegas, 2009). Similarly, cultural fit is positively associated with persistence

intentions among Latino male undergraduates (Gloria et al., 2009) and cumulative GPA

among Latinos/as attending a predominantly white institution (Cerezo & Chang, 2013).

Finally, perceptions of the university environment and cultural fit are positively related

(Anguinaga & Gloria, 2015; Gloria & Kurpius, 1996; Gloria et al., 2016), meaning that

students who perceive they belong on campus are also endorsing a positive view of their

campus climate. Collectively, this data suggests that students who perceive a positive

campus climate and cultural fit are more likely to endorse high academic outcomes or to

remain engaged in their academic endeavors.

Yet, examining the relations between campus climate and cultural fit as proximal

contextual variables in predicting college self-efficacy warrant further attention. In a

study of 115 Latinas’ academic persistence decision making, Delgado-Guerrero and

Gloria (2013) did not find significant correlations between measures of college self-

efficacy and perceptions of the university environment or cultural fit. It is possible the

lack of significant findings is an artifact of attenuation given the multiple measures

included in the study and relatively small sample size. Additionally, it is uncertain

whether these findings would be replicated with a sample of Latinos. Because few studies

have examined the direct link between these contextual variables and college self-

8

efficacy and academic performance, additional, theory derived research, is needed in this

area to arrive to clearer conclusions.

The Current Study

In the current study I will test two SCCT-informed models examining the

association between proximal contextual influences and two constructs of college self-

efficacy (Course Self Efficacy and Social Self-Efficacy) on academic performance of a

sample of Mexican American undergraduates. Specifically, I will use a measure of

college self-efficacy (Solberg, O’Brien, Villareal, Kennel, & Davis, 1993) to represent

students’ socio-cognitive attributes, measures of perceptions of the campus climate and

cultural fit (Gloria & Kurpius Robinson, 1996) as indicators of proximal contextual

influences, and students’ self-reported GPA as an indicator of performance outcome.

According to SCCT as originally hypothesized by Lent and colleagues (1994, 2000),

proximal contextual factors moderate the relationships between, and directly impact

interests-goals-actions. Because this study is not measuring those variables, I predict that

contextual variables will moderate the association between indicators of college self-

efficacy on and academic performance (GPA). See figure 2 and 3 for the hypothesized

models. Model 1 will test Course SE and moderating variables, and model 2 will test the

Social SE and moderating variables on GPA. Finally, due to the limited availability of

findings examining the role of gender differences on Latino/a students’ academic

attainment, the current study will also test the direct and moderating effect of gender on

academic performance. The following hypotheses will be tested: (a) campus climate and

cultural fit will be positively associated with measures of college self-efficacy; (b) both

measures of college self-efficacy will be positively associated with academic

9

performance; and, (c) gender will be significantly associated with measures of cultural fit

but not associated with other study variables, such as academic performance.

10

METHOD

Participants

Participants were 194 self-identified Mexican American undergraduate students,

ages 18-26 years old (M = 19.36, SD = 1.26), attending a Hispanic serving institution

(HSI; n=167, 86%) and a Historically White institution1 (HWI; n=27, 14%) located in

the southwest region of the U.S. At the HSI, Latina/o students represented approximately

73% of the student body (The University of Texas Pan American, n.d.), whereas at the

HWI they represented approximately 11% of the student population (Texas Tech

University, n.d.) at the time of data collection. Participants were recruited to represent a

diverse segment of the Mexican American undergraduate population in terms of

generational status and year in school. This sample included 92 (47%) males and 102

female students (53%). Regarding year in college, the sample was restricted to first year

(n=77, 40%) and second year (n=116, 60%) less than 1% of students did not report their

year in school. Liberal Arts and Humanities academic majors (e.g., bilingual education,

psychology, and business) were the most endorsed by this sample (n=125, 64%), and

STEM majors (e.g., biology and electrical engineering) were the least endorsed by the

sample (n=54, 28%). The remaining participants reported their major as ‘undecided.’

Lastly, 23% of the sample was first generation immigrant, 39% second generation, and

the remaining 37% third generation or greater.

Measures

College self-efficacy. The College Self-Efficacy Instrument (CSEI; Solberg,

O’Brien, Villareal, Kennel, & Davis, 1993) assesses students’ perception of their ability

1 Term adapted from Eduardo Bonilla Silva, as cited in Roediger (2005).

11

to complete college related tasks. The original CSEI includes 19 items rated on a Likert-

scale ranging from 0 (not at all confident) to 10 (extremely confident). Using a principal

component analysis, Solberg et al. (1993) found support for a three-factor structure

examining questions related to academics, roommate interactions, and

interpersonal/social adjustment self-efficacy while attending college; yet, only reported

validity and reliability data for the total scale. Using a diverse sample, Brady-Amoon and

Fuertes (2011) demonstrated adequate internal consistency of CSE total scores,

Cronbach’s alpha = .91. Research with Latino/a samples has also yielded adequate

internal consistency reliability ranging from .92 to .93 (Delgado-Guerrero & Gloria,

2013; Solberg & Villareal, 1997; Torres & Solberg, 2001) for the total scale.

Barry and Finney (2009) conducted a psychometric investigation and validation

study of the CSE and provide empirical and theoretical support for a shortened version of

the CSE and with greater attention to specific subscales. For instance, the authors posit

that the roommate self-efficacy subscale and some items in the social self-efficacy

subscale (e.g., “get a date if I wanted one”) may not be an adequate representation of

College Self-Efficacy as these items may not apply to all students (Barry & Finney,

2009). Thus, the current study utilizes the course and social self-efficacy subscales as

outlined by Barry and Finney (2009).

Course SE. This subscale of College Self-Efficacy includes 7-items and was

measured on a Likert-type scale ranging from 0 (totally unconfident) to 8 (totally

confident). Scale items were related to academic performance. Sample items included

“Do well on your exams” and “Write a course paper.” This subscale has demonstrated

strong reliability indices, ranging from .84 - .91 across three independent samples of

12

college students (Barry & Finney, 2009). The alpha coefficient for this measure of

campus climate was .88 in the current study.

Social SE. This subscale of College Self-Efficacy includes 4-items and was

measured on a Likert-type scale ranging from 0 (totally unconfident) to 8 (totally

confident). Scale items were related to interpersonal dynamics in the classroom. Sample

items included “Participate in class discussions” and “Talk to your professor.” This

subscale has demonstrated strong reliability indices, ranging from .88 - .90 across three

independent samples of college students (Barry & Finney, 2009). The alpha coefficient

for this measure of campus climate was .78 in the current study.

Campus climate. The University Environment Scale (UES; Gloria & Robinson

Kurpius, 1996) measures participants’ academic concerns and perceptions of the campus

climate. The original UES includes 14-items measured on a Likert-type scale ranging

from 1 (not at all) to 7 (very true). Scale items explore perceptions regarding classes,

faculty, university services, and overall sense of belonging. Sample items include

“University staff have been warm and friendly” and “I feel comfortable in the university

environment” The UES was developed with a sample of Chicano/a students (Gloria &

Robinson Kurpius, 1996). Research with Latino/a students in community college and

university settings found significant positive correlations between cultural fit and campus

climate as measured by the UES (Anguinaga & Gloria, 2015; Bordes & Arredondo,

2005; Gloria et al., 2016). Furthermore, the UES is negatively associated with college

persistence intentions among Latino/a undergraduates (Castillo et al., 2006; Gloria &

Kurpius, 1996), indicating that a negative perception of the campus climate negatively

impacts academic attainment. This scale has been used widely in educational outcomes

13

research with diverse Latino/a undergraduate samples demonstrating adequate internal

consistency ranging from .75 to .85 (Anguinaga & Gloria, 2015; Castillo et al., 2006;

Gloria et al., 2009; Gloria & Kurpius, 1996). In the current study, I used a seven-item

version of the UES as supported by a confirmatory factor analysis. The alpha coefficient

for this measure of campus climate was .86 in the current study.

Cultural fit. The Cultural Congruity Scale (CCS; Gloria & Robinson Kurpius,

1996) measures students’ sense of campus belonging and perceptions of how much their

cultural values fit within the campus environment. The original CCS includes 13-items

measured on a Likert-type scale ranging from 1 (not at all) to 7 (a great deal). Scale

items explore perceptions regarding sense of belonging as an ethnic minority in school

and negotiating cultural values with family and peers. Sample items include: “I feel that I

have to change myself to fit in at school” and “I try not to show the parts of me that are

‘ethnically’ based.” Gloria and Robinson Kurpius (1996) reported using the CCS and

UES jointly provides a “more complete profile” (p. 542) of factors that impact Latino/a

students’ sense of belonging and cultural fit during college. As previously reported, the

CCS and UES are conceptually and empirically linked as evidenced by positive

correlations between these variables (Anguinaga & Gloria, 2015; Bordes & Arredondo,

2005; Gloria et al., 2016). Further, among a sample of Latino undergraduates. Since its

development, the CCS has been widely used with diverse Latino/a undergraduate

samples, yielding internal consistency estimates ranging from .80 to .89 (Anguinaga &

Gloria, 2015; Cerezo & Chang, 2013; Gloria et al., 2009; Gloria & Kurpius, 1996). In the

current study, I used a five-item version of the UES as supported by a confirmatory factor

14

analysis. The alpha coefficient for this measure of campus climate was .81 in the current

study.

Demographics. A survey was used to obtain student demographic and academic

performance information. Demographic information included questions about student’s

self-reported race/ethnicity, sex, age, generational status, and year in school.

Furthermore, participants’ self-reported college GPA, rated on a 4.0 scale, will be used as

an indicator of academic performance. College GPA has been used in prior research

examining academic attainment among Latino/a students (Aguayo et al., 2011; Ong et al.,

2006).

15

RESULTS

Preliminary Analysis

The final sample (N=194) represented 65% of the participants who completed the

survey and met study criteria (e.g., first or second year college student, reported GPA).

Specifically, 49 cases were dropped due to missing GPA data. A missing value analysis

using IBM SPSS 23 on GPA indicated data were not missing at random (Little MCAR

Test p < .05). To diagnose the missing data pattern, I coded GPA as 0=missing and 1=not

missing and correlated it with academic year standing (First or Second year = 0; Junior or

Third year = 1). GPA missing pattern and academic standing were positively correlated (r

= .21, p < .01) with the majority of missing GPA occurring among first year students

(n=42), these cases were removed. Finally, 9 cases were excluded due to being under 18

or over 30 years old. After removing these cases, the sample included 238 first and

second year undergraduate students. Forty-four of those participants did not have data

available for any of the 14-items on the UES. Missing value analysis on the UES

variables indicated data were missing at random (Little’s MCAR Test p > .05). To

diagnose the missing data pattern, I coded the UES as 0=missing and 1=not missing and

ran descriptive statistics by institutions. Results indicated that 98% of the missing UES

data occurred at the HWI. It appears that during an administration of the study, the UES

scale was unintentionally excluded from the study, as it was the last question on the

survey instrument. Thus, I removed those data from additional preliminary analysis and

from the regression moderation analysis. The means, standard deviations, ranges, alpha

coefficients, and bivariate correlations among the study variables are presented in Table

1, whereas descriptive information by institution for these items is presented in Table 2.

16

Confirmatory Factor Analysis: College Self-Efficacy

To test whether the CSE factor structure of the Social SE subscale proposed by

Barry and Finney (2009) replicated in the current study, I conducted a confirmatory

factor analysis using Mplus (version 7.4) statistical software (Muthén & Muthén, 2015).

The Course SE subscale structure is the same in both Solberg et al., 1993 and Berry and

Finney (2009) validation study; yet, I decided to keep it in the analysis in order to

replicate previous studies as much as possible. Specifically, I conducted two CFA – First,

testing the subscale structures as proposed by Solberg et al., 1993 (i.e., seven item Course

SE and eight item Social SE). The second model tested the shortened CSE version (i.e.,

seven item Course SE and four item Social SE). To test model fit, I used the following fit

indices and acceptable fit cut off criteria: Comparative fit index (CFI) with values > .90

for acceptable fit; the standardized root mean-square residual (SRMR) with values < .08

indicating good fit; and the root mean square error of approximation (RMSEA) with

values < .10 for acceptable fit (Hu & Bentler, 1999; Schreiber, Stage, King, Nora, &

Barlow; 2006). The CFA results for both models and item loadings on the final version of

the scales are reported in Table 3.

The shortened, 11-item, version of the CSE (seven items Course SE subscale and

four items Social SE subscale) had overall better fit indices than the 15-item version

originally proposed by Solberg et al., (1993). All items loaded into their intended factors.

Factor loadings for the seven items on Course SE ranged from .66 to .73. Factor loadings

for the four items on the Social SE ranged from 0.57 to 0.81. Course and Social SE were

significantly positively correlated (r = .73, p < .01). Thus, in light of these findings the

four-item Social SE was used in the main analysis.

17

Confirmatory Factor Analysis: University Environment Scale

Since its’ development, the UES – this study’s measure of campus climate – has

been widely used in educational research with Latinos/as. However, when this scale was

first developed, the authors did not use factor analytic techniques. Thus, in the current

study I tested the UES factor structure using CFA in Mplus (version 7.4) statistical

software (Muthén & Muthén, 2015). The 14-item UES had poor fit indices (see previous

section for discussion regarding acceptable fit cut off criteria). Seven of the 14 items had

factor loadings less than 0.05. Researchers recommend that ideal factor loadings should

be greater than 0.50, as these items contribute the most to the factor structure (McCoach,

Gable, & Madura, 2013). These items may not have loaded onto the single factor

structure for two reasons. First, five of the items were reversed scored (i.e. they were

originally negatively worded items) leading them to cluster together (McCoach et al.,

2013). Whereas, the remaining two items (one related to values and one related to

financial aid) did not make conceptual sense with the remaining scale items. Thus, given

this empirical data and conceptual justification, I decided to test a seven-item factor

structure and all fit indices improved. The shortened, 7-item, version of the UES had

overall acceptable fit indices, beyond those of the original 14-item version. Factor

loadings ranged from .55 to .86. The CFA results for both models, as well as the final

scale factor loadings, are reported in Table 4.

Confirmatory Factor Analysis: Cultural Congruity Scale

Similar to the UES, the CCS has been widely used with Latinos/as and

educational research; but has not been validated using factor analytic techniques. The

results from the CFA conducted on the 13-item CCS indicated the scale had poor fit

18

indices. Five of the items (which were also reverse scored in the original survey) had

factor loadings lower than the recommended cut off value of 0.50 (McCoach et al., 2013).

After removing those five items, The 8-item CCS fit indices had improved; yet, still

failed to achieve adequate fit indices. Due to this empirical data, additional

respecification of the CCS factor structure was needed. While model respecification may

help improve overall fit, it can also leads to a more exploratory analysis which may

require additional data collection and testing of the stability of the scales’ factor structure

(McCoach et al., 2013). After careful examination of the remaining scale items, I chose to

remove three items related to values and value conflicts. These items, at face value, were

not consistent with the remaining items on the scale. The shortened, 5-item, version of

the CCS exclusive includes items about fitting in at school in light of one’s own ethnicity.

This scale version had overall acceptable fit indices, beyond those of the original 13-item

version. Factor loadings ranged from .58 to .82. The CFA results for the three models, as

well as the final scale factor loadings, are reported in Table 5.

Multiple Regression Moderation Analysis

Two practically identical multiple regression models were used to examine the

associations between Course SE, moderating variables, and academic performance

(model 1) and between Social SE, moderating variables, and academic performance

(model 2). All analyses were conducted in Mplus (version 7.4) and were modeled after

Stride, Gardner, Catley, and Thomas (2015) Mplus adaptation of Andrew Hayes

PROCESS analysis examples (Hayes, 2013). An added benefit of utilizing Mplus is that

it allows the researcher to specify which type of estimation analysis to use given the

research question and quality of the data. For the purposes of this study, I conducted the

19

multiple regressions utilizing a Maximum Likelihood Estimator with Robust Standard

Errors (MLR), which accounts for missing data and is robust for non-normality and non-

independence of observations (Muthén & Muthén, 2015). Each model included and

tested for the direct effects of the study variables: Course SE or Social SE, campus

climate, cultural fit, and gender. To test for moderation, I computed interaction terms by

centering and multiplying the predictor (e.g., Social SE) with each of the moderating

variables (i.e., campus climate, cultural fit, and gender). The results of the regression

analysis for Course SE are presented in Table 6 and for Social SE in Table 7. Results

indicated that none of the moderating variables or interaction terms had a direct or

moderating effect on academic performance. Course SE, but not Social SE, was a

significant predictor (β = .354, p < .05) of academic performance. The Course SE model

accounted for 8.8% of the variance in academic performance among this sample of

Mexican American first and second year undergraduate students.

20

DISCUSSION

The current study examined the associations between college self-efficacy

(Course SE and Social SE), proximal contextual influences (Campus climate and cultural

fit), and gender on the academic performance (self reported GPA) of Mexican American

first and second year undergraduate students. This study contributes to our knowledge of

college student development in general, and academic attainment among Mexican

Americans specifically. This study contributes to our understanding of the utility of

SCCT in predicting academic attainment among Latinos/as as it includes constructs (e.g.,

cultural fit) particularly designed to represent the experiences of Latinos/as in higher

education. Furthermore, the study findings offer insight into factors that could help

promote educational attainment among Latino/a students. Results indicated that course

self-efficacy was a significant predictor of academic performance for Mexican American

undergraduate students. In addition, social self-efficacy, positive perceptions of the

campus climate, and cultural fit were associated with high self-efficacy.

Results from the multiple regression analysis indicated that only Course self-

efficacy was a significant predictor of academic performance among this sample of

Mexican American undergraduate students. These findings are consistent with prior

research that has demonstrated academic and college self-efficacy are positively

associated with a variety of educational outcomes, including academic satisfaction

(DeWitz & Walsh, 2002) and college performance (Aguayo et al., 2011; Brady-Amoon

& Fuertes, 2010; Chemers et al., 2001; Gore 2006). In this study, the course self-efficacy

measure involved tasks related to time management, exam and note taking, completing

assignments, and understanding course content. Not surprisingly, students who felt

21

confident in performing the tasks listed above had higher self-reported GPA than those

with lower sense of course-self efficacy.

Course self-efficacy is only one element of global college self-efficacy. An

additional aspect of global college self-efficacy is social self-efficacy, or the ability to

deal with interpersonal tasks while in college (Solberg et al., 2013). In this study, social

self-efficacy pertained to interpersonal interactions that occur in the classroom – such as

participating in class discussions, talking with professors or school staff, and asking

instructors questions outside of the classroom. Against my hypothesis, social self-efficacy

did not predict academic performance among this sample of Mexican American

undergraduates. It is possible that this construct is associated with other educational

outcomes, such as persistence intentions (Wright, Jenkins-Guarnieri, & Murdock, 2012)

and not just GPA. Furthermore, it is possible that the measure I utilized for social self-

efficacy underrepresented the intended construct (Barry and Finney, 2009) and it may be

that other indicators of interpersonal self-efficacy are better predictors of academic

performance in college.

Perceptions of campus climate and cultural fit were not significant predictors of

academic performance or moderated the association between college self-efficacy and

performance. These results are not totally unexpected, as previous research with

measures of perceptions of the university environment has yielded mixed results. For

instance, where as Cerezo & Chang (2013) found cultural congruity and college GPA to

be positively correlated among Latino/a college students, Delgado-Guerrero and Gloria

(2013) did not find those associations among their sample of Latina undergraduates.

Thus, it was important to test whether gender was a significant predictor among these

22

associations. However, results from this study did not provide any significant

associations between gender and all other study variables. Though prior studies (Gloria et

al., 2005; 2009) had found associations between gender and cultural fit, it is possible that

the modified cultural fit scale used in this study attenuated the associations between these

variables. It would be important for future research to examine additional factors or

models that may better explain the link between contextual variables and academic

performance. For example, some educational SCCT-based research suggests

environmental variables predict educational outcomes via indirect (mediation)

relationship of self-efficacy (Byars-Winston, et al., 2010; Lent et al., 2003; Sheu et al.,

2016). Future research with Mexican American undergraduate students might explore

mediation models involving the person-cognitive and environmental constructs included

in this study.

Implications for Practice and Research

Promoting educational attainment of Latinos/as is important for the social and

economic development of the United States and for individual Latino/a community

members. The findings of the current study reinforce our knowledge about the role of

academic self-efficacy in promoting academic attainment. Educational institutions, as

well as psychologists, can work to develop brief interventions that promote individual’s

increase in self-confidence in their abilities. For example, having access to quality

tutoring services and encouraging college students to utilize this resource can help to

foster academic self-efficacy for students. Establishing learning communities, where

students can engage in vicarious learning opportunities, may be another strategy for

helping students develop increased self-efficacy.

23

Regarding university context, though the findings of this study did not support a

direct link between campus climate and cultural fit to academic performance; the findings

did show that academic self-efficacy is positively correlated to these variables. In other

words, students who perceive their campus climate as a supportive and welcoming

environment have higher confidence in their academic abilities than those who do not

view their campus climate positively. Thus, it behooves colleges and universities to

continue allocating resources targeting the social integration of racial/ethnic minority

students in their respective campuses. Promoting a sense of belonging is likely to help

Latino/a students perform better academically via an increased sense of social self-

efficacy.

Future research should consider how generation status (both as an immigrant and

college student) might moderate perceptions of the campus climate and cultural fit.

Longitudinal studies would be uniquely positioned to test the efficacy of interventions or

the mediating effect of attributes such as academic self-efficacy on academic

performance. Psychometric validation studies of measures commonly used with

Latinos/as are also needed. This study has taken a step toward advancing the

psychometric validation of the UES and CCS, which are commonly used scales in

Latinos/as educational research. However, the results presented in this study are

somewhat exploratory due to extensive model respecification process that was needed to

improve scale model fit (McCoach et al., 2013). Further, it is possible that measures that

were developed 20-30 years ago do not represent the nuanced experiences of Latino/a

students in the current socio-political climate.

24

Limitations and Conclusions

The current study sought to include a representative sample of Mexican American

undergraduate students across two institutions. However, students attending the HWI

were not well represented in the final sample for this study. The lack of variability

regarding where data were collected is a limitation of this study because this could have

limited the variability of responses reported on the campus climate and cultural fit

measures. Future researchers should be mindful of recruiting diverse student samples and

to consider how their studies measures will be impacted by participant demographic

variables.

As noted earlier, it is important to examine psychometric properties of measures

commonly used with Latinos/as, especially if these were developed with non-Latino

samples. Based on the current findings it is difficult to assess whether the measures of

campus climate and cultural fit are psychometrically sound instruments. Thought these

measures correlated with college self-efficacy in the intended directions, the lack of

significant findings as predictors of GPA may be a warning sign about the underlying

factor structure of these measures or due to an underrepresentation of the construct.

Finally, as common with many cross-sectional research designs, the current study

does not help establish causality. Findings should be interpreted with caution because

important demographic and cultural variables (e.g., generational status, and family source

of support) were not included in this study’s design. Furthermore, the study was limited

to self-reported GPA and without having access to student records it is uncertain if this

outcome variable accurately represented student’s academic performance. Furthermore,

these data were not weighted by institution or type of major, which may increase

25

measurement error bias into the statistical analysis. Finally, GPA may not be the best

predictor of academic persistence and retention.

In summary, the current study explored the role of course and social self-efficacy

in predicting academic performance of Mexican American first and second year

undergraduate students. Results indicated that course self-efficacy was a significant

predictor of self-reported GPA. Furthermore, positive perceptions of the campus climate

and sense of belonging as measured by cultural fit were positively associated with college

self-efficacy. Supporting Latino/a students academic achievement via interventions that

promote increased academic confidence and a positive campus climate are likely to

increase the higher education graduation rate of this highly underrepresented group.

26

REFERENCES

Aguayo, D., Herman, K., Ojeda, L., & Flores, L. Y. (2011). Culture predicts Mexican Americans’ college self-efficacy and college performance. Journal of Diversity in Higher Education, 4(2), 79-89. doi: 10.1037/a0022504

Aguinaga, A., & Gloria, A. M. (2015). The effects of generational status and university environment on Latina/o undergraduates’ persistence decisions. Journal of Diversity in Higher Education, 8(1), 15–29. doi:10.1037/a0038465

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215. doi:10.1037/0033-295x.84.2.191

Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice Hall.

Barajas, H. L., & Pierce, J. L. (2001). The significance of race and gender in school success among Latinas and Latinos in college. Gender & Society, 15(6), 859-878.

Barry, C. L., & Finney, S. J. (2009). Can we feel confident in how we measure college confidence? Measurement and Evaluation in Counseling and Development, 42(3), 197–222. doi:10.1177/0748175609344095

Baum, S., Ma, J., & Payea, K. (2013). Education Pays 2013. The benefits of higher education for individuals and society. College Board.

Bordes, V. & Arredondo, P. (2005). Mentoring and 1st-Year Latina/o College Students. Journal of Hispanic Higher Education, 4(2), 114–133. doi:10.1177/1538192704273855

Brady-Amoon, P., & Fuertes, J. N. (2011). Self-Efficacy, Self-Rated Abilities, Adjustment, and Academic Performance. Journal of Counseling & Development, 89(4), 431–438. doi:10.1002/j.1556-6676.2011.tb02840.x

Byars-Winston, A., Estrada, Y., Howard, C., Davis, D., & Zalapa, J. (2010). Influence of social cognitive and ethnic variables on academic goals of underrepresented students in science and engineering: A multiple-groups analysis. Journal of Counseling Psychology, 57(2), 205–218. doi:10.1037/a0018608

Castillo, L. G., Conoley, C. W., Choi-Pearson, C., Archuleta, D. J., Phoummarath, M. J., & Van Landingham, A. (2006). University environment as a mediator of Latino ethnic identity and persistence attitudes. Journal of Counseling Psychology, 53(2), 267–271. doi:10.1037/0022-0167.53.2.267

Cerezo, A., & Chang, T. (2013). Latina/o Achievement at Predominantly White Universities. Journal of Hispanic Higher Education, 12(1), 72–85. doi:10.1177/1538192712465626

27

Chemers, M. M., Hu, L., & Garcia, B. F. (2001). Academic self-efficacy and first year college student performance and adjustment. Journal of Educational Psychology, 93(1), 55–64. doi:10.1037/0022-0663.93.1.55

Colby, S. L. & Ortman, J. M. (2015). Projections of the size and composition of the U.S. population: 2014 to 2060. United States Census Bureau. Report: P25-1143

Delgado-Guerrero, M., & Gloria, A. M. (2013). La importancia de la hermandad Latina: Examining the psychosociocultural Influences of Latina-based sororities on academic persistence decisions. Journal of College Student Development, 54(4), 361–378. doi:10.1353/csd.2013.0067

DeWitz, S. J., & Walsh, W. B. (2002). Self-Efficacy and College Student Satisfaction. Journal of Career Assessment, 10(3), 315–326. doi:10.1177/10672702010003003

Gloria, A. M., Castellanos, J., & Herrera, N. (2016). The reliability and validity of the cultural congruity and university environment scales with Chicana/o community college students. Community College Journal of Research and Practice, 40(5), 426–438. doi:10.1080/10668926.2015.1066276

Gloria, A. M., Castellanos, J., & Orozco, V. (2005). Perceived Educational Barriers, Cultural Fit, Coping Responses, and Psychological Well-Being of Latina Undergraduates. Hispanic Journal of Behavioral Sciences, 27(2), 161–183. doi:10.1177/0739986305275097

Gloria, A. M., Castellanos, J., Scull, N. C., & Villegas, F. J. (2009). Psychological Coping and Well-Being of Male Latino Undergraduates. Hispanic Journal of Behavioral Sciences, 31(3), 317–339. doi:10.1177/0739986309336845

Gloria, A. M., & Kurpius, S. E. R. (1996). The Validation of the Cultural Congruity Scale and the University Environment Scale with Chicano/a Students. Hispanic Journal of Behavioral Sciences, 18(4), 533–549. doi:10.1177/07399863960184007

Gore, P. A. (2006). Academic Self-Efficacy as a Predictor of College Outcomes: Two Incremental Validity Studies. Journal of Career Assessment, 14(1), 92–115. doi:10.1177/1069072705281367

Hu, L. T., & Bentler, P. MM. (1999). Cutoff criteria for fit indices in covariance structure analysis: Conventional criteria versus alternatives. Structural Equation Modeling, 6, 1-55. doi: 10.1080/10705519909540118

Krogstad, J. M. (2016). 5 facts about Latinos and education. Pew Research Center. Retrieved from: http://www.pewresearch.org/fact-tank/2016/07/28/5-facts-about-latinos-and-education/

28

Lent, R. W., Lopez, F. G., Brown, S. D., & Gore, Jr., P. A. (1996). Latent Structure of the Sources of Mathematics Self-Efficacy. Journal of Vocational Behavior, 49(3), 292–308. doi:10.1006/jvbe.1996.0045

Lent, R. W., Brown, S. D., & Hackett, G. (1994). Toward a Unifying Social Cognitive Theory of Career and Academic Interest, Choice, and Performance. Journal of Vocational Behavior, 45(1), 79–122. doi:10.1006/jvbe.1994.1027

Lent, R. W., Brown, S. D., & Hackett, G. (2000). Contextual supports and barriers to career choice: A social cognitive analysis. Journal of Counseling Psychology, 47(1), 36–49. doi:10.1037/0022-0167.47.1.36

McCoach, D. B., Gable, R. K., & Madura, J. P. (2013). Instrument development in the affective domain – School and corporate applications (3rd ed). New York: Springer

Multon, K. D., Brown, S. D., & Lent, R. W. (1991). Relation of self-efficacy beliefs to academic outcomes: A meta-analytic investigation. Journal of Counseling Psychology, 38(1), 30–38. doi:10.1037/0022-0167.38.1.30

Muthén, L. K. & Muthén, B. O. (1998-2015). Mplus user’s guide (7th ed). Los Angeles: Muthén & Muthén

Ojeda, L., Flores, L. Y., & Navarro, R. L. (2011). Social cognitive predictors of Mexican American college students’ academic and life satisfaction. Journal of Counseling Psychology 58(1), 61–71. doi: 10.1037/a0021687

Ojeda, L., Navarro, R. L., & Morales, A. (2011). The role of la familia on Mexican American men’s college persistence intentions. Psychology of Men & Masculinity, 12(3), 216–229. doi: 10.1037/a0020091

Ong, A. D., Phinney, J. S., & Dennis, J. (2006). Competence under challenge: Exploring the protective influence of parental support and ethnic identity in Latino college students. Journal of Adolescence, 29(6), 961–979. doi:10.1016/j.adolescence.2006.04.010

Pajares, F. (1996). Self-Efficacy Beliefs in Academic Settings. Review of Educational Research, 66(4), 543. doi:10.2307/1170653

Ramos-Sánchez, L., & Nichols, L. (2007). Self-Efficacy of First-Generation and Non-First-Generation College Students: The Relationship With Academic Performance and College Adjustment. Journal of College Counseling, 10(1), 6–18. doi:10.1002/j.2161-1882.2007.tb00002.x

Richardson, M., Abraham, C., & Bond, R. (2012). Psychological correlates of university students’ academic performance: A systematic review and meta-analysis. Psychological Bulletin, 138(2), 353–387. doi:10.1037/a0026838

29

Robbins, S. B., Lauver, K., Le, H., Davis, D., Langley, R., & Carlstrom, A. (2004). Do Psychosocial and Study Skill Factors Predict College Outcomes? A Meta-Analysis. Psychological Bulletin, 130(2), 261–288. doi:10.1037/0033-2909.130.2.261

Roediger D. (2005) What’s wrong with these pictures? Race, narratives of admission, and the liberal self-representations of historically white colleges and universities. Journal of Law and Policy, 18, 203–222.

Ryan, C. L. & Bauman, K. (2016). Educational attainment in the United States: 2015. United States Census Bureau. Report: P20-578

Schreiber, J. B., Stage, F. K., King, J., Nora, A., & Barlow, E. A. (2006). Reporting structural equation modeling and confirmatory factor analysis results: A Review. The Journal of Educational Research, 99, 323-327. doi: 10.3200/JOER.99.6.323-338

Tabachnick, B. G., & Fidell, L. S. (2013). Using multivariate statistics (6th ed.). New York: Harper Collins.

The University of Texas Pan American (n. d.). 2005-2007 Undergraduate Catalog . Retrieved from: http://www.utrgv.edu/_files/documents/academics/legacy-catalogs/utpa%202005-2007%20undergraduate%20catalog.pdf

Texas Tech University (n. d.). Fall enrollment by ethnicity since 1975. Retrieved from: https://www.depts.ttu.edu/irim/ARCHIVE/ENR/FALLETHNIC.php

Torres, J. B., & Solberg, V. S. (2001). Role of Self-Efficacy, Stress, Social Integration, and Family Support in Latino College Student Persistence and Health. Journal of Vocational Behavior, 59(1), 53–63. doi:10.1006/jvbe.2000.1785

Sheu, H.-B., Mejia, A., Rigali-Oiler, M., Primé, D. R., & Chong, S. S. (2016). Social cognitive predictors of academic and life satisfaction: Measurement and structural equivalence across three racial/ethnic groups. Journal of Counseling Psychology, 63(4), 460–474. doi:10.1037/cou0000158

Stride, C.B., Gardner, S., Catley, N. & Thomas, F. (2015) 'Mplus code for the mediation, moderation, and moderated mediation model templates from Andrew Hayes' PROCESS analysis examples'. Retrieved from: http://www.offbeat.group.shef.ac.uk/FIO/mplusmedmod.htm

Solberg, V. S., O’Brien, K., Villareal, P., Kennel, R., & Davis, B. (1993). Self-Efficacy and Hispanic College Students: Validation of the College Self-Efficacy Instrument. Hispanic Journal of Behavioral Sciences, 15(1), 80–95. doi:10.1177/07399863930151004

30

Wright, S. L., Jenkins-Guarnieri, M. A., & Murdock, J. L. (2012). Career Development Among First-Year College Students: College Self-Efficacy, Student Persistence, and Academic Success. Journal of Career Development, 40(4), 292–310. doi:10.1177/0894845312455509

31

FIGURES

Figure 1

SCCT as presented in Lent, Brown, and Hackett (1994)

32

Figure 2

Model 1 - Course Self-Efficacy and Moderating Variables on Academic Performance

CampusClimateUES

CulturalFitCCS

AcademicPerformance

GPACourseSelf-Efficacy

Gender

33

Figure 3

Model 2 - Social Self-Efficacy and Moderating Variables on Academic Performance

CampusClimateUES

CulturalFitCCS

AcademicPerformance

GPASocial

Self-Efficacy

Gender

34

TABLES

Table 1 Correlations, Reliability Coefficients, Means, Standard Deviations and Range among Measured Variables Variable 1 2 3 4 5 6 1. Academic Performance –

2. Course Self-Efficacy .24** – 3. Social Self-Efficacy .063 .66** – 4. Campus Climate .076 .48** .37** – 5. Cultural Fit -.017 .30** .29** .40** – 6. Gender (0=male,

1=female) .097 -.058 -.135 .063 .075 –

Cronbach’s Alpha –

.88 .79 .86 .81 – Mean 2.91 5.85 5.94 5.49 6.17 .53

Standard Deviation .517 1.44 1.56 1.16 1.13 .50 Range 1.5–4.0 1.0–8.0 0–8.0 1.0–7.0 1.8–7.0 0–1.0

Note. N = 194; *. p < .05; **. p < .01

35

Table 2 Means and Standard Deviations among Measured Variables by Institution Hispanic Serving

Institution (HSI) n=167 Historically White

Institution (HWI) n=27 M SD M SD Academic Performance 2.88 .49 3.11 .63

Course Self-Efficacy 5.86 1.48 5.8 1.2

Social Self-Efficacy 5.98 1.55 5.69 1.6

Campus Climate 5.52 1.20 5.32 .87

Cultural Fit 6.26 1.08 5.59 1.3

Gender (0=male, 1=female) .49 .50 .74 .45

36

Table 3 Summary of Model-Fit Statistics and Factor Loadings on measure of College Self-Efficacy

Model χ2 df CFI SRMR RMSEA 90% Confidence

Interval for RMSEA

15-item CSE, 2 factors (8 items on Social Self-Efficacy)

273.78 89 .829 .073 .093 (0.081, 0.106)

11-item CSE, 2 factors (4 items on Social Self-Efficacy

104.99 43 .908 .063 .078 (0.059, 0.097)

Note. CSE = College Self-Efficacy; CFI = comparative fit index; SRMR = standardized root mean squared residual; RMSEA = root mean square error of approximation. Factor 1 – Course Self Efficacy Items B (SE) β

Take good class notes 1.00 (--) .73 Research a term paper 1.09 (.13) .73

Understand your textbooks .99 (.13) .69 Write a course paper .95 (.09) .72

Do well on your exams .90 (.12) .72 Manage your time effectively 1.02 (.12) .66

Keep up to date with your school work 1.01 (.11) .71

Factor 2 – Social Self Efficacy Items B (SE) Β Talk to your professors/instructors 1.00 (--) .77

Ask a professor or instructor a question outside of class .97 (.07) .81 Talk with a school academic and support (e.g., advising) staff .93 (.12) .68

Participate in class discussions .84 (.09) .57

37

Table 4 Summary of Model-Fit Statistics and Factor Loadings on measure of Campus Climate

Model χ2 df CFI SRMR RMSEA 90% Confidence

Interval for RMSEA

14-item UES, 1 factor 248.88 77 .692 .101 .118 (0.103, 0.133)

7-item UES, 1 factor 37.03 14 .927 .046 .092 (.057, .129)

Note. UES = University Environment Scale; CFI = comparative fit index; SRMR = standardized root mean squared residual; RMSEA = root mean square error of approximation.

Items B (SE) β The library staff is willing to help me find materials/books. 1.00 (--) .72

University staff have been warm and friendly. 1.12 (.08) .86 The university encourages/sponsors ethnic groups on campus. .78 (.12) .64

There are tutoring services available for me on campus. .59 (.12) .59 Faculty have been available for help outside of class. .81 (.11) .74

Faculty have been available to help me make course choices. .91 (.12) .68 I feel comfortable in the university environment .65 (.10) .55

38

Table 5 Summary of Model-Fit Statistics and Factor Loadings on measure of Cultural Fit

Model χ2 df CFI SRMR RMSEA 90% Confidence

Interval for RMSEA

13-item CCS, 1 factor 397.46 65 .519 .135 .147 (0.133, 0.161)

8-item CCS, 1 factor 86.35 20 .80 .076 .118 (.093, 0.144)

5-item CCS, 1 factor 6.402 5 .991 .025 .034 (.000, 0.101)

Note. CCS = Cultural Congruity Scale; CFI = comparative fit index; SRMR = standardized root mean squared residual; RMSEA = root mean square error of approximation.

Item B (SE) β

I feel that I have to change myself to fit in at school. 1.00 (--) .78

I try not to show the parts of me that are “ethnically” based. .93 (.15) .71

I often feel like a chameleon, having to change myself depending on the ethnicity of the person I am with at school. .94 (.13) .82

I feel that my ethnicity is incompatible with other students. .76 (.14) .58

I feel that my language and/or appearance make it hard for me to fit in with other students .70 (.09) .59

39

Table 6 Summary of Multiple Regression of Course Self-Efficacy and Moderating Variables on Academic Performance Variable B (SE) β Course Self-Efficacy .125 (.037) .349** Campus Climate -.025 (.034) -.057 Cultural Fit -.02 (.039) -.045 Gender (0=male, 1=female) .123 (.072) .119 Course Self-Efficacy x Campus Climate -.002 (.015) -.008 Course Self-Efficacy x Cultural Fit -.004 (.019) -.013 Course Self-Efficacy x Gender -.046 (.047) -.096 Model R2 adjusted .077* Note. N = 194; *. p < .05; **. p < .01;

40

Table 7 Summary of Multiple Regression of Social Self-Efficacy and Moderating Variables on Academic Performance Variable B (SE) β Social Self-Efficacy .041 (.041) .123 Campus Climate .029 (.031) .065 Cultural Fit -.012 (.038) -.027 Gender (0=male, 1=female) .114 (.076) .111 Course Self-Efficacy x Campus Climate -.019 (.015) -.079 Course Self-Efficacy x Cultural Fit -.013 (.018) -.047 Course Self-Efficacy x Gender -.054 (.048) -.123 Model R2 adjusted .033 Note. N = 194; *. p < .05; **. p < .01