The Influence of Self-Efficacy Beliefs in the Academic Performance of Latina/o Students in the United States: A Systematic Literature Review

Abstract

This review had three aims: (a) examine the relationship between self-efficacy (SE) and academic performance (AP) in Latina/o students, (b) identify the types of measures of SE and AP that have been used, and (c) identify the differences between global and specific measures of SE in predicting academic performance in Latina/o students. Twenty-seven articles were included in the analysis. Major findings included the following: (a) significant and positive relationship between self-efficacy and academic performance in Latina/o students, at all educational levels, and through different types of measurements of self-efficacy and academic performance; (b) particular performance domains or content, or task-specific variables, were more favorable in the examination of the relationship between self-efficacy and academic performance than global domains; (c) Latina/o is not a homogeneous group; therefore, acculturation process, immigration status, family’ experience on higher education, affective variables, and parents’ and teachers’ support were also factors that significantly correlated with Latina/os’ academic achievement.

Keywords

self-efficacy academic performance Latina/o students systematic literature review

Introduction

Latina/o students have among the lowest rates of high school and college graduation rates in the United States (National Center for Education Statistics, 2015; Sáenz & Ponjuan, 2012). Although college attendance has increased considerably in the last decade, Latina/o students lag behind their White, Black, and Asian peers (Sáenz & Ponjuan, 2012). Among the reasons for the documented lag are the fact that 38% of Latina/os are enrolled at high-poverty schools, 48% attended schools that did not meet adequate yearly progress (AYP), and only 38% of Latina/o students attended schools in which the counseling programs prepared them for postsecondary education (Ross et al., 2012). Many studies have explored the effects of economic hardships, lack of support, and low quality of the school system on the academic performance among Latina/o students (Nora & Crisp, 2012; Padilla, 2006; Schneider, Martinez, & Owens, 2006). However, there are few studies that examine the relationship between Latina/o psychosocial characteristics such as self-efficacy, outcome expectations and academic goals, and academic performance or academic achievement in this ethnic group (Metheny & McWhirter, 2013; Ojeda, Flores, & Navarro, 2011).

Academic performance is strongly linked to educational and occupational behaviors (Brown & Lent, 2006). Among social cognitive factors related to academic performance, self-efficacy has been identified as among the most robust predictors of academic performance (Sheu et al., 2010). Self-efficacy refers to beliefs about personal capabilities to produce results by actions (Bandura, 2006). According to Bandura, people who believe in their capacities are motivated to take actions and persevere despite difficulties. For example, students who believe in their academic capabilities are more willing to take courses and follow the steps to achieve academic pursuits than those who do not believe in their capabilities (Brown & Lent, 2006).

In 1991, Multon, Brown, and Lent conducted the first meta-analysis of the relation of self-efficacy beliefs to academic outcomes. The authors found positive and statistically significant relationships between self-efficacy beliefs and academic performance (effect size = .38), and across different subjects, experimental designs, and assessment methods, self-efficacy beliefs accounted for 14% of the variance in students’ academic performance. Recent meta-analyses conducted with college students have consistently shown a positive correlation between academic self-efficacy and academic performance. Honicke and Broadbent (2016) found that academic self-efficacy moderately correlate with academic performance. Richardson, Abraham, and Bond (2012) found medium-sized correlations between academic self-efficacy and grade point average (GPA). Similarly, Brown et al. (2008) found a direct relationship between self-efficacy on academic performance rather than mediated by academic goals. Robbins et al. (2004) also found that the best predictor, among multiple construct, of GPA was academic self-efficacy. Although single studies and meta-analysis provide reliable support for the relationship between self-efficacy and academic performance, the contribution of self-efficacy in Latina/os’ academic performance is controversial.

The controversy arises out of conceptualizations of self-efficacy. Brown and Lent (2006) pointed out results of previous meta-analysis suggesting that the influence of self-efficacy in educational and occupations behaviors is more related to “Westernized” adolescents than minority’s that adhere strongly to their identified racial/ethnic groups. Sheu et al. (2010) called for research on the effects of socio-cognitive constructs that shape educational and vocational interests and choices with diverse populations (e.g., minorities groups or ethnicities). Adding to the controversy, studies on self-efficacy with Latina/o students provide contradictory results. For instance, Niehaus, Rudasill, and Adelson (2012) found that self-efficacy was not a significant predictor of GPA in Latina/o middle school students. Conversely, Chun and Dickson (2011) found that academic self-efficacy was the only pathway that directly led to better academic performance among other contextual and cognitive variables in Latina/o middle students. One possible reason for mixed findings on Latina/o students’ self-efficacy is the difference between global and specific measures of self-efficacy. Global domains evaluate self-efficacy across academic behaviors whereas specific self-efficacy measures assess students’ perceptions of a particular aspect of their academic endeavors (Honicke & Broadbent, 2016). For example, academic self-efficacy is a global domain and math self-efficacy is a specific domain. Chesnut and Burley (2015) found that the relationship between self-efficacy and academic performance is stronger when specific measures of self-efficacy are used. In addition, Brown and Lent (2006) pointed out that different from other self-constructs (e.g., self-confidence or self-esteem), self-efficacy is a dynamic construct intimately tied to a specific academic or occupational domain. For instance, students with equal levels of self-confidence or self-esteem may differ in their levels of self-efficacy to accomplish academic goals related to science or social fields. Therefore, the purpose of this systematic review was to present the state of literature on the multiple domains in which Latina/o self-efficacy has been measured and the effects of self-efficacy in Latina/os’ academic performance.

Research Questions

Specifically, this review sought to address the following:

Research Question 1: What do the results of research findings suggest about the relationship between self-efficacy and academic performance?

Research Question 2: What types of measures of self-efficacy and academic performance have been used in the context of Latina/o students in the United States?

Research Question 3: What is the difference between global and specific measures of self-efficacy in predicting academic performance in Latina/o students?

This systematic literature review followed the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA; Moher, Liberati, Tetzlaff, Altman, & Prisma Group, 2009).

Method

Eligibility Criteria and Selection

Studies for inclusion in this review were limited to studies that (a) were English-only peer-reviewed journal articles and theses/dissertations; (b) were conducted in the United States; (c) explored the relationship between self-efficacy and academic performance; (d) used qualitative, quantitative, or mixed methods; and (e) included Latina/o, Hispanic, or Mexican American elementary, middle, high, or college students. Studies were excluded if they were not primary reports of research findings (e.g., reviews, commentaries, or program reviews).

Search Strategy

Studies were identified by searching 18 electronic databases from 1991 to 2016: ERIC, Academic Search Complete, Education Full Text (H.W. Wilson), Education Source, Educational Administration Abstracts, Family & Society Studies Worldwide, Family Studies Abstracts, Gender Studies Database, MEDLINE Complete, Professional Development Collection, Psychology and Behavioral Sciences Collection, PsycINFO, Race Relations Abstracts, Social Sciences Full Text (H.W. Wilson), SocINDEX with Full Text, Teacher Reference Center, Vocational Studies Complete, and ProQuest Dissertations and Theses. Search terms included “self-efficacy or academic self-efficacy,” “academic performance or academic achievement or academic success,” “middle school students or high school students or college students or university students,” and “Latina/o or Latina or Hispanic or Mexican-American.”

Data Abstraction

A data abstraction table was created to enable the primary author to collect and tabulate information relevant to this study. For this analysis, the information extracted included author and year, sample characteristics (sample size, proportion of Latina/os in the study, and gender), setting (primarily, middle, high, and higher education), study design and analytic methods, self-efficacy as a predictor measure of academic performance, academic performance as an outcome measure, and relevant findings (correlation, prediction, and path coefficients).

Plan of Analysis, Summary Measures, and Synthesis of Results

A qualitative synthesis of relevant literature was conducted to determine (a) the nature of the relationship between self-efficacy and academic performance, (b) the types of measures that have been used to assess self-efficacy and academic performance, and (c) the difference between global and specific measures of self-efficacy in predicting the academic performance of Latina/o students.

Risk of Bias

In order to address publication bias, the present review included both article journals and theses and dissertations. Locating unpublished studies may help to reduce publication bias and to provide a better estimate of effectiveness or association (Song, Hooper, & Loke, 2013).

Results

Study Selection

Through the literature search, a total of 186 articles were identified. After removing duplicates and those that were not relevant to this review, 79 studies remained. After reviewing the 79 articles for inclusionary criteria, 27 met the criteria for this review (23 quantitative and four qualitative studies).

The reasons for the exclusions described above are as follows: over 26% of the screened articles were excluded because they did not provide findings on the relationship between self-efficacy and academic performance for Latinas/os disaggregated from the total sample; 31% of the articles did not include self-efficacy as a predictor of academic performance. In addition, 35% of the studies did not present academic performance as an outcome variable. Of the remaining articles, 8% were excluded because they were not a primary report of research findings (i.e., systematic or meta-analysis, program evaluation, or report article).

The final 27 articles included dissertations (n = 9) and journal articles from various disciplines, including Hispanic or Latina/o studies (n = 6); education, teaching, and learning (n = 5); child, adolescence, and family (n = 2); diversity (n = 2); and communication (n = 1). One article was published in 1998, nine articles between 2001 and 2009, and 17 after 2011 (of which six articles were published in 2012). Figure 1 outlines the PRISMA flow diagram of the studies included in the review as well as the excluded studies and reasons for exclusion.

Study Characteristics

Sample sizes in the studies examined in this review ranged from five to 1,456 participants. In all, 70% were conducted only with Latina/os students (n = 19) and 30% with other comparison races/ethnicities students (n = 8). Approximately 74% of the studies reported gender (n = 20) and 26% studies did not (n = 7). The settings of the screened studies were 48% in higher institutions—community college (n = 3), college/university (n = 5), and Hispanic Serving Institutions (HSI; n = 5)—22% in high schools (n = 6); 22% in elementary, intermediate, and middle schools (n = 6); 8% in others (one study gathered participants from elementary, middle, and high school combination, and one study from an after-school program). For further details of study characteristics, see Table 1 for quantitative studies and Table 2 for qualitative studies.

Table 1.

Summary of Research Quantitative Studies’ Sample Characteristics, Study Design, Self-Efficacy and Academic Measures, and Findings.

Author and year	Sample characteristics	Setting	Study design and analytic methods	Self-efficacy measure	Academic performance measure	Findings
Acoach and Webb (2004)	n = 89; Latinos = 100%; M = 32; F = 53; not respond = 4	High School	Design: Cross-sectional; analytic methods: t tests, factor analysis, Spearman’s rho analyses	Academic Self-Efficacy scale (Solberg, O’Brien, Villareal, Kennel, & Davis, 1993); 9 items; Global domain; Cronbach’s alpha = .84	Self-report GPA	Nonparametric correlation between academic self-efficacy and GPA (p = .46,p² = .21, p < .02)
Aguayo, Herman, Ojeda, and Flores (2011)	n = 408; Latinos = 100%; M = 171; F = 237	College—Hispanic Service Institution	Design: cross-sectional; analytic methods: multiple regression analyses	College Self-Efficacy Inventory (Solberg, O’Brien, Villareal, Kennel, & Davis, 1993); 20 items; Global domain; Cronbach’s alpha = .90	Self-report GPA	CSE was significantly correlated with GPA (r = .24, p > .01). CSE was significantly correlated with GPA for second generation and beyond students (r = .29, p < .001. But CSE was not significantly correlated to GPA for immigrants (r = .13; p = .26)
Buriel, Perez, De Ment, Chavez, and Moran (1998)	n = 122; Latinos = 100%; M = 59; F = 63	High School	Design: cross-sectional; analytic methods: t tests, stepwise multiple regression	Academic self-efficacy scale was created based on Solberg et al., (1993); 10 items; Global domain; Cronbach’s alpha = .87	Self-reported grades	Academic self-efficacy was significantly correlated with academic performance(r = .26, p < .01). Self-efficacy was a significant predictor of academic performance(β = .45, p < .001), accounting for 23% of the explained variance.
Castro-Villarreal, Guerra, Sass, and Hseih (2014)	n = 128; Latinos = 54%; M = 36; F = 92	College—Hispanic Service Institution	Design: cross-sectional; analytic methods: confirmatory factor analysis, SEM.	Academic Self-Efficacy scale from the PALS (Midgley et al., 2000); 6 items, Global domain; Cronbach’s alpha = .96.	Final course grades was obtained from official university records	Path coefficient in Hispanic students from ASE to final course grades (β = 9.16, unstandarized coefficient and = 9.16, β = .45, standarized coefficient)
Chun and Dickson (2011)	n = 478; Latinos = 100%; M = 227; F = 246; not respond = 5	Middle school, seventh grade	Design: cross-sectional; analytic methods: structural equation model	Academic self-efficacy was measured using the Jonson-Reid et al. (2005) adaptation of Bachman’s (1970) School Ability Self-Concept Index; 8 items; Global domain; Cronbach’s alpha = .95.	Self-reported grades English, math, and science	Path coefficient from ASE to academic performance (β = .70, p < .001)
Chun, Marin, Schwartz, Pham, and Castro-Olivo (2016)	n = 289; Latinos = 100%; M = 84; F = 205	College—Hispanic Service Institution	Design: cross-sectional; analytic methods: structural equation model	College Self-Efficacy scale (Solberg et a., 1993); 9 items; Global domain; Cronbach’s alpha = .88	Semester GPA was obtained from official university records	College self-efficacy was significantly correlated with GPA (r = .021, p < .01). Path coefficient from CSE to GPA (β = .24 p < .01)
DeFreitas and Bravo (2012)	n = 249; Latinos = 58%; gender was not reported	College—Hispanic Service Institution	Design: cross-sectional; analytic methods: t tests, hierarchical linear regression	Self-Regulated Learning scale of the Multidimensional Scales of Perceived Self-Efficacy (Bandura, 1990); 11 items; Global domain; Cronbach’s alpha for Latinos = .87	GPA was obtained from official university records	Self-efficacy for Latina/o students was not a statistically significant predictor of GPA
Dopke (2001)	n = 354; Latinos = 100%; gender was not reported	Elementary and intermediate schools	Design: cross-sectional; analytic methods: structural equation model	Two multidimensional scales of perceived self-efficacy (Bandura, 1990): Self-Efficacy for Academic Achievement, and Self-Efficacy for Self-Regulated Learning; 9 items and 11 items; Global domain; Cronbach’s alpha did not report	Three observed variables were used to measure final grade: prior grades, the average of the final two 9-week quarters, final grade	Path coefficient from Student SE Academic Achievement to final grade by family styles: authoritative (β = .53, p < .05) authoritarian (β = .29, p < .05) indulgent (β = .69, p < .05), neglectful (β = .55, p < .05)
Edman and Brazil (2009)	n = 475; Latinos =15%; M = 171; F = 304	Community college	Design: cross-sectional; analytic methods: ANOVA, Pearson correlations	Academic Self-Efficacy Scale from Child’s Self-Efficacy Scale (Bandura, 2006); 13 times, Global domain; Cronbach’s alpha = .89	Semester and cumulative GPAs were obtained from official university records	Academic efficacy was positively associated with GPA cumulative, r(66) = .328, p < .001. Academic self-efficacy was not statistically correlated with GPA semester
Fong, Zientek, Yetkiner Ozel, and Phelps (2015)	n = 804; Latinos = 19%; gender was not reported	Community college	Design: cross-sectional; analytic methods: DDAs, SEM: ordinal logistic regression	Four subscales of students’ efficacy beliefs (Marat, 2005) for Motivational strategies, 5 items; Cognitive Strategies, 15 items; Resource Management Strategies, 12 items; and Self-Regulated Learning, 12 items; Specific Domain: math; Cronbach’s alpha = .97	Grade of math course letter grade	Self-regulated learning strategies was a statistically significant predictor of math grade, Wald = 5.82, p = .016. Motivational, cognitive, resource-management strategies were not statistically significant predictors of math grade.
Garriott and Flores (2013)	n = 90; Latinos = 100%; M = 27; F = 56; not respond = 7	Community college	Design: longitudinal; analytic methods: post hoc power analyses, MANOVA	College self-efficacy measured by 1 item asking students to rate their perceived confidence in their ability to complete a degree at a 4-year college; Global domain; Cronbach’s alpha did not report	Self-reported GPA was assessed as a proxy of educational performance.	College self-efficacy was a significant positive predictor of GPA, F(1, 76) = 5.24, p < .05, η = .07 (η represents the univariate effect size)
Hewitt (2009)	n = 79; Latinos = 100%; M = 29; F = 50	High School	Design: two pronged design; analytic methods: power analysis, ANOVA, t test, ANCOVA, hierarchical regression	SEQ-C (Muris, 2001); 24 items; Global domain; Cronbach’s alpha .88	Student grades (2007-2008 school year) were obtained from official high school records	Academic self-efficacy was predictor of academic achievement (F change = 4.637, p = .02, R = .32, R² = .10, ΔR² = .06
Lewis et al. (2012)	n = 1,456; Latinos = 100%; M = 680; F = 776	Elementary and intermediate schools	Design: multilevel modeling (two-level) with maximum likelihood estimation; analytic methods: multiple group confirmatory factor analysis	Math self-efficacy (Karabenik & Maehr, 2007); 4 items, Specific domain; Cronbach’s alpha = .84	The CST-M was used to assess student math achievement. CST-M were obtained from the schools records	The main effect of math self-efficacy on math achievement was significant, t(1,366) = 7.816, p < .001
McNamara (2012)	n = 98; Latinos = 100%; M = 34; F = 64	College—Hispanic Service Institution	Design: cross-sectional; analytic methods: mixed methods model; hierarchical multiple regression, logistic regression, focus groups	Science Self-Efficacy Scale (Britner & Pajares, 2001); 5 items; Specific domain; Cronbach’s alpha = .91	GPA was obtained from official university records	Bivariate correlations between science self-efficacy and earned credit hours (r = .22, p = .03). SSE was statistically significant predictor of earned credit hours (F = 6.47, p =.01) and accounted for 5.60% of the variation. Bivariate correlation between self-efficacy and GPA was not statistically significant. (r = .06, p = .56).
Mejia Arias (2007)	n = 88; Latinos = 100%; M = 22; F = 66	College—Private university	Design: cross-sectional; analytic methods: multiple regression with hierarchical and bivariate correlations, ANOVAs	CASES (Owen & Froman, 1988); 33 items; Global domain; Cronbach’s alpha did not report	Self-report cumulative GPA	College self-efficacy was moderate correlated with GPA (r = .34, p < .01). Academic self-efficacy was a strong predictor of GPA (β = .48, p < .001)
Migray (2003)	n = 651; Latinos = 16.6%; M = 327; F = 321; not respond = 3	Middle school, sixth and seventh grade	Design: cross-sectional; analytic methods: Pearson product–moment correlations, linear regression, covariance procedures	Math self-efficacy by this researcher; 20 items; Specific domain; Cronbach’s alpha = .88	Math performance, the Woodcock–Johnson Psychoeducational Battery (1977)	Regression coefficient for prediction of math achievement by math self-efficacy (β = .26, p < .01, n = 108). The regression coefficients for grade level: sixth grade was not significant, n = 68; seventh grade Hispanic students (β = .47 p = .00, n = 40)
Niehaus, Rudasill, and Adelson (2012)	n = 47; Latinos = 100%; M = 34; F = 13	After-school program in a middle school	Design: longitudinal; analytic methods: ordinary least squares regression, multilevel growth models	MJSES (Jinks & Morgan, 1999); 30 items; Global domain; Cronbach’s alpha = .85	GPA, standardized test scores for the current and previous school years (reading and math) were obtained from official school records	Self-efficacy was not related to GPA (β = .20, p = .14). Self-efficacy was not a statistically significant predictor of students’ reading achievement scores (β = .18, p = .25). However, students who had higher self-efficacy at the beginning of the year (β = .39, p = .01) and attended the after-school program more frequently (β = .20, p = .02) achieved higher scores on the end-of-year standardized math assessment.
Riconscente (2014)	n = 326; Latinos = 100%; M = 140; F = 186	High school, ninth and 10th grade	Design: multiwave design; analytic methods: one-way and 2 × 2 ANOVA, post hoc analyses with Bonferroni correction	Math self-efficacy adapted from the Motivated Strategies for Learning Questionnaire (Pintrich, Smith, Garcia, & McKeachie, 1991) and was administered at T1 and T3; 3 items; Specific domain; Cronbach’s alpha = .84 and .78 at T1 and T3, respectively.	Math grades for the academic year were obtained from official school records.	T1: Self-efficacy was significantly correlated to math achievement at T1 (r = .27) and T3 (r = .37, p < .001). Standardized beta coefficients for predicting end-of-year mathematics achievement: only self-efficacy at T1 was a marginally significant predictor of math achievement (β = .169, p = .001)
Rivera (2012)	n = 99; Latinos = 100%; M = 57; F = 42	Elementary and middle schools second, fifth, eighth grades	Design: cross-sectional; analytic methods: linear and hierarchical linear multiple regression analysis	Competence Beliefs Questionnaire (Wigfield et al., 1997); only math and reading competence beliefs; 10 items; Specific domain; Cronbach’s alpha = .74	The Woodcock–Johnson III Tests of Achievement: Batería math, Batería Spanish-Reading, and English reading—WJ-III (Woodcock, McGrew, & Mather, 2001); 76 items.	SASE was not correlated with batteries (math, Spanish, and English reading). SASE significantly predicted battery reading (β = −.22, p < .05). SASE did predict neither batteries nor English reading at any grade. MSE was significantly correlated to Batería math (r = .21,p < .05). By grades, MSE was significantly related to Batería math in fifth grade (r = .48, p < .05) and eighth grade (r = .52, p < .01). MSE did not predict math achievement across grade levels. However, MSE significantly predicted math achievement in fifth (β = −.51, p < .05) and eighth grade (β = −.59, p < .05). RSE did not predict reading achievement (both Spanish and English reading) across grade levels. However, RSE significantly predicted reading English in eighth grade (β = −.44, p < .05).
Roberts (2014)	n = 83; Latinos = 27.8%; M = 49; F = 34	High school	Design: cross-sectional; analytic methods: Pearson’s r correlation tests, Crosstabs chi-square tests, Spearman’s rho correlation tests, and multiple regression tests	Student Efficacy Scale (Jinks & Morgan, 1999); 30 items; Global domain; Cronbach’s alpha from previous study = .81	GPA was obtained from official school records	Self-efficacy was significantly correlated with GPA (r = .488, p < .018). A cross tabulation chi-square suggests nonstatistically significant values for the strength of relationship between academic achievement and self-efficacy among all three racial groups.
Stevens, Olivárez, Lan, and Tallent-Runnels (2004)	n = 358; Latinos = 53%; gender was not reported	High school, ninth and 10th grade	Design: cross-sectional; analytic methods: SEM path model	Mathematics self-efficacy (Pajares & Graham, 1999); 20 items; Specific domain; Cronbach’s alpha for Hispanics = .87	Math performance: students were asked to complete 20 problems that were similar to those on which they had previously rated their confidence in solving them.	Path coefficient from math self-efficacy to math performance (β = .22, p < .01). Adequate model fit (CFI = .98, SRMR = .06)
Stevens, Olivárez, and Hamman (2006)	n = 610; Latinos = 52%; gender was not reported	Elementary, middle, and high schools above fourth grade	Design: cross-sectional; analytic methods: SEM path model	Mathematics self-efficacy by authors; 19 items; Specific domain; Cronbach’s alpha = .92	Two subtests were employed to assess mathematics performance: calculation and mathematics fluency, of the Woodcock–Johnson III Tests of Achievement	Path coefficient from math self-efficacy to math performance (β = .27, p < .05). Poor model-to-data fit for the Hispanic sample (CFI = .90, SRMR = .10)
Zychinski and Polo (2012)	n = 133; Latinos = 100%; M = 77; F = 56	Elementary and middle schools, fifth to seventh grades	Design: cross-sectional; analytic methods: hierarchical regression models, tests for mediational effects	PALS (Midgley et al., 2000); Academic Self-Efficacy, Mastery Learning, and Performance-Avoidance subscales; 5 items; Global domain; Cronbach’s alpha = .78	GPA students’ final grades for the 2007-2008 school year were obtained from official school records	Academic self-efficacy was significantly correlated with ISAT reading/math standard score (r = .27, p < .0). Academic self-efficacy was significantly correlated to reading/math GPA (r = .31, p < .0)

Note. GPA = grade point average; CSE = college self-efficacy; SEM = structural equation model; PALS = Patterns of Adaptive Learning Scales; ASE = academic self-efficacy; SE = self-efficacy; DDA = descriptive discriminant analysis; SEQ-C = Self-Efficacy Questionnaire for Children; CST-M = California Standards Test for Mathematics; SSE = science self-efficacy; CASES = College Academic Self-Efficacy Scale; MJSES = Student Efficacy Scale; SASE = students’ academic self-efficacy; MSE = math self-efficacy; RSE = reading self-efficacy; CFI = comparative fit index; SRMR = standardized root mean square residual; ISAT = Illinois Standards Achievement Test.

Table 2.

Summary of Research Qualitative Studies’ Sample Characteristics, Study Design, Self-Efficacy and Academic Measures, and Findings.

Author and year	Sample characteristics	Setting	Study design and analytic methods	Self-efficacy measure	Academic performance measure	Findings
Ayro (2012)	n = 5; Latinos = 100%; M = 5	College	Design: cross-sectional qualitative; analytic methods: ethnographic interviews	Self-efficacy is a factor linked to academic success. Other factors: relationships, engagement, persistence, and ethnic identity	Academic success was defined as graduating from high school and passing academic classes	Student’s self-efficacy was linked to educational success. Self-efficacy allowed the participant to make decisions which assisted them in improving grades, applying for scholarships, preparing for college, and steering from negative temptations (drugs and gangs).
Cavazos et al. (2010)	n = 11; Latinos = 100%; gender was not reported	College—Hispanic Service Institution	Design: cross-sectional qualitative; analytic methods: interviews, open-ended questions	High self-efficacy is one of the individual factors that play a role in developing resilience	High academic achievement: high grade point averages and prestigious academic credentials	Participants belief that their efforts and hard work would result in high academic achievement. Effort, perseverance, hard work, and self-belief were qualities to become successful in college. These factors facilitate the deep conviction that they were in control of their academic future.
Salas (2001)	n =8; Latinos = 100%; M = 4; F = 4	College	Design: cross-sectional; analytic methods: narrative analysis, in-depth, open-ended, flexibly structured interviews	Self-efficacy as a belief in their capacity to succeed through education. Their goals, initiative, tenacity, and self-efficacy helped them successfully negotiate school and family difficulties, and provided a base for the development of leadership skills.	School success: honor’s and/or advanced placement classes during high school	Participants used creative strategies for coping with barriers to their academic success. The students had a high sense of self-efficacy which resulted in good school performance. The students academic successes helped build a sense of academic competence that transferred into leadership and initiative in other areas of their lives.
Zell (2010)	n = 15; Latinos = 100%; gender was not reported	Community college	Design: cross-sectional qualitative; analytic methods: one-on-one, in-depth, semistructured interviews	Self-efficacy emerged as a theme that examined the psychological and subjective experiences of Latina/o community college students.	Academic success was related to academic goals: the desire to go to college, to obtain solid academic preparation, and to achieve professional goals.	Participants reported feeling confident that they could accomplish their academic goals. Although some participants may have doubted their ability to succeed prior to enrolling in college, all had come to trust their ability to navigate an academic environment. Self-efficacy was attributed to having tested themselves in community college and having succeeded at something that was not expected of them. All sacrifices (personal, financial, emotional) were made worthwhile by the belief that they had achieved a major milestone in development.

Study Design and Analytic Methods

The sample consisted of 18 studies using a cross-sectional research design (Acoach & Webb, 2004; Aguayo, Herman, Ojeda, & Flores, 2011; Buriel, Perez, De Ment, Chavez, & Moran,1998; Castro-Villarreal, Guerra, Sass, & Hseih, 2014; Chun & Dickson, 2011; Chun, Marin, Schwartz, Pham, & Castro-Olivo, 2016; DeFreitas & Bravo, 2012; Dopke, 2001; Edman & Brazil, 2009; Fong, Zientek, Yetkiner Ozel, & Phelps, 2015; McNamara, 2012; Mejia Arias, 2007; Migray, 2003; Rivera, 2012; Roberts, 2014; Stevens, Olivárez, Lan, & Tallent-Runnels, 2004; Stevens, Olivárez, & Hamman, 2006; Zychinski & Polo, 2012). Four studies used a qualitative research design (Ayro, 2012; Cavazos et al., 2010; Salas, 2001; Zell, 2010). Two studies used a longitudinal design (Garriott & Flores, 2013; Niehaus et al., 2012). Three used a two pronged design (Hewitt, 2009): a one multilevel design (Lewis et al., 2012) and one multiwave design (Riconscente, 2014), respectively.

For data analyses, 15 studies utilized multiple regressions (e.g., confirmatory factor analysis, logistic regression, structural equation model), four studies used ethnographic and narrative analysis (e.g., coding and thematic analysis), five studies used multivariate statistics (e.g., t tests, hierarchical and linear regression), three studies used ANOVA, and one study used bivariate statistics.

Academic Self-Efficacy Measures

In all studies, academic self-efficacy or any form of self-efficacy was the primary predictor of academic performance. The following constructs were measured in self-efficacy scales: academic self-efficacy, college academic self-efficacy, math self-efficacy, science self-efficacy (SSE), and self-efficacy. Table 3 presents constructs of self-efficacy and the instruments used to measures these constructs.

Table 3.

Constructs of Self-Efficacy and Instruments for Measurements.

Construct of self-efficacy	Number of studies	Instruments for measurements
Academic self-efficacy	6	Academic Self-Efficacy Scale (Solberg et al., 1993; n = 2); Self-efficacy from Patterns of Adaptive Learning Scales (PALS; Midgley et al., 2000; n= 2); Academic Self-Efficacy Scale from Child’s Self-Efficacy Scale (Bandura, 2006; n = 1), and academic self-efficacy (Jonson-Reid et al., 2005; n = 1)
College academic self-efficacy	4	College Self-Efficacy Inventory (Solberg et al., 1993; n = 2); College Academic Self-efficacy Scale (CASES; Owen & Froman, 1988; n = 1). One question to measure college self-efficacy was created by the authors of one study (Garriott & Flores, 2013; n = 1).
Math self-efficacy	7	Math self-efficacy (n = 1); math self-efficacy adapted from the Motivated Strategies for Learning Questionnaire (Pintrich, Smith, Garcia, & McKeachie, 1991; n = 1); mathematics self-efficacy (Pajares & Graham, 1999; n= 1); Competence Beliefs Questionnaire (only math competence beliefs; Wigfield et al., 1997; n = 1). Students’ efficacy beliefs–specific math (Marat, 2005; n = 1). Two scales were created by the authors of the study: Math Self-Efficacy (Migray, 2003); Mathematics Self-Efficacy (Stevens, Olivárez, & Hamman, 2006)
Reading self-efficacy	1	Competence Beliefs Questionnaire (only reading competence beliefs; Wigfield et al., 1997)
Science self-efficacy	1	Science Self-Efficacy Scale (Britner & Pajares, 2001)
Self-efficacy	5	Multidimensional Scales of Perceived Self-efficacy (Bandura, 1990; n = 2); Student Efficacy Scale (MJSES; Jinks & Morgan, 1999; n = 2); Self-Efficacy Questionnaire for Children (SEQ-C; (Muris, 2001; n = 1)

Global and Specific Measures of Academic Self-Efficacy

Global measures assessed self-efficacy across academic behaviors whereas specific self-efficacy measures assessed students’ perceptions of a particular aspect of their academic endeavors (Honicke & Broadbent, 2016). Of the 23 quantitative studies reviewed, 15 used measures classified as global and eight studies as specific domain. The three types of global domains focused on academic self-efficacy (Acoach & Webb, 2004; Buriel et al., 1998; Castro-Villarreal et al., 2014; Chun & Dickson, 2011; Edman & Brazil, 2009; Zychinski & Polo, 2012), college self-efficacy (Aguayo et al., 2011; Chun et al., 2016; Garriott & Flores, 2013; Mejia Arias, 2007), and self-efficacy (DeFreitas & Bravo, 2012; Dopke, 2001; Hewitt, 2009; Niehaus et al., 2012; Roberts, 2014). Similarly, three types of specific domains were extracted and included reading self-efficacy (RSE; Rivera, 2012), SSE (McNamara, 2012), and math self-efficacy (Fong et al., 2015; Lewis et al., 2012; Migray, 2003; Riconscente, 2014; Rivera, 2012; Stevens et al., 2004; Stevens et al., 2006). The Cronbach’s alpha of the global measures of SE ranged from .78 to .96 and the specific measures of SE ranged from .74 to .97. Three instruments did not report the reliability of their SE measures (Dopke, 2001; Garriott & Flores, 2013; Mejia Arias, 2007).

In regard to the global domains, of the 15 studies, 10 used similar instruments to measure self-efficacy. The most common scales utilized were the Multidimensional Scales of Perceived Self-Efficacy (Bandura, 1990; n = 2); the Student Efficacy Scale (Jinks & Morgan, 1999; n = 2); the Patterns of Adaptive Learning Scales (PALS; Midgley et al., 2000, n = 2); the Academic Self-Efficacy Scale (Solberg, O’Brien, Villareal, Kennel, & Davis, 1993; n = 2); and the College Self-Efficacy Inventory (Solberg et al., 1993; n = 2). On the contrary, in specific domain, there was not a similar instrument utilized to measure self-efficacy; each study used a specific instrument related to a specific domain under study (see Table 3).

Academic Performance Measures

Academic performance is the outcome achieved by a student and measured by tests and examinations. The measures of academic performance that were found are subject specific grades, tests for specific fields, and GPA. Of the 23 qualitative studies reviewed, the most commonly reported outcome variable was GPA (n = 12). Within those 12, eight used GPA obtained from official records and four from self-reported GPA scores. The 11 remaining outcome variables reported were battery tests (n = 4), grades from official records (n = 2), self-reported grades (n = 2), grades that did not report the source (n = 2), and the math standard test results (n = 1). Both battery tests and the standard test measured specific academic subjects (e.g., math, English reading, and Spanish reading).

Synthesis of Results of Quantitative Studies

This section presents the results of the relationships between self-efficacy and academic performance according to the specificity level in which self-efficacy and academic performance were measured by the authors included in this review.

Academic self-efficacy and academic performance

Four of the five studies found correlations between academic self-efficacy (ASE) and GPA and ASE and grades. Acoach and Webb (2004) found a nonparametric correlation between ASE and GPA (p = .46, p < .02) in Latina/o high school students. Edman and Brazil (2009) found a positive correlation between ASE and GPA cumulative (r = .33, p < .001) in Latina/o community college students. Buriel et al. (1998) found a significant correlation between ASE and AP (r = .26, p < .01) in Latina/o high school students. Zychinski and Polo (2012) found a significant correlation between ASE and Illinois Standards Achievement Test (ISAT; r = .27, p < .0) and ASE and GPA (r = .31, p < .00), in fifth and seventh grade Latina/o students.

Although Edman and Brazil (2009) found a correlation between ASE and GPA cumulative, they did not find a statistically significant correlation between ASE and GPA semester. Another study also did not find a statistically significant correlation between ASE and math, Spanish, or English batteries, in second, fifth, and eighth grade Latina/o students (Rivera, 2012).

Regarding the prediction of ASE in AP, two studies found that ASE predicted AP and one study found mixed results. Buriel et al. (1998) found that ASE was a significant predictor of AP (β = .45, p < .001), accounting for 23% of the explained variance, in Latina/o high school students. Hewitt (2009) found that ASE was a predictor of Academic Achievement (r = .32), accounting for 5% of the variance in academic achievement. Rivera (2012) found that ASE only negatively predicted battery Spanish reading (β = −.22, p < .05) while ASE did not predict math, Spanish reading, and English at any grade (second, fifth, eighth) in Latina/o students.

The studies that used path analysis found significant relationship between ASE and grades. Castro-Villarreal et al. (2014) found significant path coefficient from ASE to final course grades (β = 9.16) in college Hispanic students. Chun and Dickson (2011) found significant path coefficient from ASE to grades (math and English; β = 0.70, p < .001), in middle school students.

College self-efficacy and academic performance

The four studies that examined the relationship between college self-efficacy (CSE) and AP in Latina/os students found positive and significant correlations. Mejia Arias (2007) found a correlation between CSE and cumulative GPA (r = .34, p < .01) in a private university, and Aguayo et al. (2011) and Chun et al. (2016) found correlations (r = .24, p < .01, and r = .21, p < .01, respectively) in HSIs. However, Aguayo et al. (2011) found differences between Latina/os who were born overseas (named first generation) and Latina/os who were born in the United States (named second, third, and so on). CSE was significantly correlated with GPA for second and beyond generations (r = .29, p < .001), but it was not significantly correlated for first generation (r = .13, p = .26).

Two studies found that CSE was a significant predictor of GPA: Mejia Arias (2007) reporting a β = .48 (p < .001) and Garriott and Flores (2013) reporting a F(1, 76) = 5.24 (p < .05, η = .07). In a similar way, Chun et al. (2016) found a path coefficient from CSE to GPA (β = 0.24, p < .01). The population of these four studies was Latina/o students in two HSIs, a private university, and a community college, respectively.

Self-efficacy and academic performance

Two studies reported a significant relationship between self-efficacy (SE) and AP. Roberts (2014) found that SE was significantly correlated with GPA (r = .488, p < .018) in Latina/o high school students. Dopke (2001) found significant path coefficient from SE to final grade by family style: authoritative (β = .53, p < .05), authoritarian (β = .29, p < .05), indulgent (β = .69, p < .05), and neglectful (β = .55, p < .05). Two additional studies were unable to find a statistically significant relationship between SE and GPA (DeFreitas & Bravo, 2012; Niehaus et al., 2012). The first was comprised of Latina/o students from a HSI, and the second one included Latina/o middle school students from an after-school program.

Math self-efficacy and academic performance

The findings of the relationship between math self-efficacy (mean square error [MSE]) and grades were mixed—of the seven studies, three found significant relationship, three found mixed results, and one did not identify a good model fit to explain this relationship. At the college level, Fong et al. (2015) found that self-regulated learning strategies, a dimension of self-efficacy, was a significant predictor of math grade (Wald = 5.82, p = .016). Lewis et al. (2012) found that MSE was the main effect on math achievement in elementary and intermediate Latina/o students, t(1364) = 7.03, p < .001. Stevens et al. (2004) found a significant path coefficient with an adequate model fit (comparative fit index [CFI] = .98, standardized root mean square residual [SRMR] = .06) from MSE to math performance (β = .22, p < .01) in Latina/o high school students. The studies that found mixed findings were Migray (2003) who found MSE predicted math achievement in Latina/o middle school students (β = .26, p < .01); however, when the prediction was estimated by grades, MSE predicted math achievement only for seventh graders (β = .47 p = .00) and not for sixth grade Latina/o students. Riconscente (2014) using a multiwave design found that MSE was significantly correlated to math achievement at T1 and T3 (r = .27 and r = .37, p < .001, respectively). Riconscente (2014) also found that MSE at T1 was marginally a significant predictor of math achievement (β = .169, p = .001), but it was not a predictor at the final time—T3 in Latina/o high school students. Rivera (2012) found that MSE was significantly correlated with battery math (r = .21, p < .05) in Latina/o students in general. However, Rivera (2012) found differences by grades; MSE was significantly related to battery math in both fifth (r = .48, p < .05) and eighth grades (r = .52, p < .01), but it was not significant in second grade. Rivera (2012) also found MSE predicted negatively math achievement in both fifth (β = −.51, p < .05) and eighth grades (β = −.59, p < .05), but MSE did not predict math grades in second grade.

Finally, Stevens et al. (2006) tested a theoretical model explaining the relationships between cognitive, motivational, and emotional variables in their prediction of mathematics performance in both White and Latina/o students attending elementary, middle, and high school. The findings for Latina/os were a significant path coefficient from MSE to math performance (β = .27, p < .05); however, the model fit was not good (CFI = .90, SRMR = .10).

RSE and academic performance

Only one study examined the relationship between RSE and AP and the results were mixed. Rivera (2012) found that RSE significantly predicted English reading achievement in the eighth grade (β = −.44, p < .05). However, RSE was not correlated with battery reading both across grade level and by grades. In a similar way, RSE predicted neither English nor Spanish reading achievement across grade levels.

SSE and academic performance

Similarly, only one study examined the relationship between SSE and AP and the findings were also mixed. McNamara (2012) reported correlations between SSE and credit hours (r = .22, p = .03) and SSE and GPA (r = .06, p < .56). Moreover, SSE was a significant predictor of credit hour (F = 6.47, p = .01) accounting for 6% of the explained variance. However, the contribution of self-efficacy in explaining the variation in GPA was not statistically significant.

Synthesis of Results of Qualitative Studies

The four qualitative studies included in this review found that self-efficacy helped students to make decisions which increased their academic performance and college readiness (Ayro, 2012). The use of creative strategies for coping with barriers to their academic success was another characteristic of Latina/o students’ self-efficacy (Salas, 2001). In addition, students’ academic successes helped them to build a sense of academic competence that can be transferred into leadership and initiative in other areas of their lives (Salas, 2001). Self-efficacy also was attributed to success at something—in that case, college environment—that was not expected of them (Zell, 2010). Latina/o perceived that sacrifices (personal, financial, emotional) that were made were worthwhile due to the belief that they had achieved a major milestone in their personal development (Zell, 2010). One study found that similar to self-efficacy, effort, hard work, and perseverance were factors that influenced students’ academic achievement (Cavazos et al., 2010).

Discussion

This literature review sought to identify studies that examine the relationship between self-efficacy and academic performance in Latina/o students, identify the types of measures of SE and AP that have been used, and identify the differences between global and specific measures of SE in predicting academic performance in Latina/o students. A systematic review identified 27 articles. Of the 23 quantitative studies reviewed, 14 used global measures of self-efficacy (self-efficacy beliefs, academic self-efficacy, and college self-efficacy) and nine used specific domain (math self-efficacy, reading self-efficacy, and science self-efficacy). The most common measure of academic performance was GPA; the remaining measures of AP were grades from battery tests and grades from specific subjects. Settings of the studies were 13 in higher institutions and 13 in elementary, middle, and high schools, and one in an after-school program.

Based on these findings, it is evident that there is a significant and positive relationship between self-efficacy and academic performance in Latina/o students, at all educational levels, and through different types of measurements of self-efficacy and academic performance. This relationship supports the hypothesis that students’ efficacy beliefs play a pivotal role in cognitive development and accomplishments as suggested by Bandura (2006). The findings suggest that Latina/o students with high levels of self-efficacy attain higher grades and succeed academically. Self-efficacious Latina/o students make efforts to improve their grades, to prepare for college, and to avoid negative temptations leading to risk behaviors (Ayro, 2012). These Latina/o students believe that their effort and hard work will lead to achievement of academic goals and thereby affect their academic future (Cavazos et al., 2010). In addition, self-efficacious Latina/o students use creative strategies for coping with barriers to hinder their academic success (Salas, 2001). However, more research is needed to inquire the characteristics of self-efficacious Latina/o students and strategies they use to increase their academic performance and to establish academic goals.

In addition, findings have shown that academic, math, and college self-efficacy were the domains in which more research has been done on Latina/o students. Persistent gaps between the educational attainment of White and Hispanic students have encouraged the inquiry of Latina/os’ beliefs related to their academic success (Chun & Dickson, 2011; Zychinski & Polo, 2012). In a similar way, the low rates of college attendance and college retention have been reasons to inquire the role of Latina/os’ self-efficacy beliefs in their confidence to successfully complete college-related tasks (Aguayo et al., 2011; Chun et al., 2016; Garriott & Flores, 2013; Mejia Arias, 2007). Findings in these domains (academic, math, and college) suggest that measures of self-efficacy focusing on a particular performance domain or task-specific domain were more favorable to the examination of the relationship between self-efficacy and academic performance.

More research is also needed to evaluate the appropriate applicability of self-efficacy belief scales—as a global domain—to inquire the relationships to academic performance. Of the five studies that used self-efficacy belief instruments, three were unable to find significant relationships or nonsignificant values for the strength of the relationship between self-efficacy and GPA (DeFreitas & Bravo, 2012; Niehaus et al., 2012; Roberts, 2014). Future research should also focus on evaluating the impact of variables in the selected domain of self-efficacy.

Immigration status was a common theme among studies which did not find a relationship between self-efficacy and academic performance. Within Latina/os, there are differences in self-efficacy beliefs of academic success between young and older generations. Self-efficacy beliefs were not related to academic performance in first-generation Latina/o immigrants, or those who were born outside of the United States. This review found three studies in which nonstatistically significant correlation between academic achievement and self-efficacy was found for migrant Hispanic students (Roberts, 2014), first-generation Latina/o students (Aguayo et al., 2011), and a study in which the 66% of the sample size were first-generation immigrants (Niehaus et al., 2012). Another study found mixed results, academic efficacy was positively associated with GPA cumulative, but it was not correlated with GPA semester. This finding is not surprising because first-generation Latina/o students face culture shock between their family culture and mainstream American culture (Aguayo et al., 2011; Murillo & Schall, 2016).

In a similar way, some students feel discriminated or isolated by their ethnicity or country of origin (McWhirter, Luginbuhl, & Brown, 2014; McWhirter, Ramos, & Medina, 2013). Therefore, these previous experiences may affect Latina/os’ self-efficacy beliefs about their capabilities to succeed academically (Aguayo et al., 2011) and to navigate suitable pathways in the U.S. educational system. More research also is needed to evaluate the effect of immigrant status on self-efficacy beliefs and their relationships to Latina/os’ academic success as well as other emotional and contextual variables that affect Latina/os’ self-efficacy beliefs.

A specific characteristic that affects the relationship between self-efficacy and academic performance is being first-generation college student. In the United States, half of all Hispanic students are first-generation college students; they are more likely to come from low socio-economic and academically disadvantaged backgrounds and have a lack of familiarity with college options, processes, and norms (Reyes & Nora, 2012). In addition, first-generation college students “perceived strong contrasts between the literacies of home and school and between literacy instruction and expectations for reading and writing in high school and college” (Murillo & Schall, 2016, p. 321). This situation may considerably affect Latina/os’ success in academic performance and academic literacies (DeFreitas & Bravo, 2012; Murillo & Schall, 2016). More research is also needed to identify the effect of being first-generation college student in self-efficacy beliefs about both their capabilities to achieve academic goals in higher education and their skills to navigate successfully in campus social life/college involvement, in particular on a large campus with a small presence of Latina/o populations.

The contrast between the level of academic requirements between high school and college might also affect students’ self-efficacy. First-year college students expect to earn high grades like they used to have at the high school level without considering the differences between high school level and college level in the academic rigor (McNamara, 2012). As a result, first-year college student with previous positive academic experiences in high schools with low academic rigor might feel overconfident in being able to accomplish tasks for academic success. Elevated levels of self-efficacy might also be self-defeating because students might establish unrealistic academic goals in unknown contexts that exceed their capabilities, generating a situation of failure or discouragement (Brown & Lent, 2006). As a result, inauthentic high school academic feedback and ineffective motivational goals may also affect the relationship between academic self-efficacy and GPA semester in students enrolled in introductory-level courses (Edman & Brazil, 2009). Therefore, in order to prevent an overestimation of self-efficacy in first-year college students, researchers may consider collecting data later in the semester or after the students have experienced their first university-level examination (McNamara, 2012). In a similar way, better correlation between self-efficacy and exam performance has been found when the data collection occurs from the mid-point of a course to the end (Galyon, Blondin, Yaw, Nalls, & Williams, 2012).

An extensive literature has found that not only motivational but also cognitive, emotional, and contextual variables influence Latina/os’ academic performance. Although several studies in this review found a significant relationship between self-efficacy and academic achievement, there were four studies in which the inclusion of other variables modified the effect of self-efficacy on the outcome variable or found better relationships between academic performance and other variables. For example, the inclusion of several affective variables affected the goodness of fit of a theoretical model explaining the relationships between cognitive, motivational, and emotional variables in their prediction of mathematics performance for Latina/o students (Stevens et al., 2006). Academic self-efficacy was related to involvement with faculty and mentoring, and it was not related to GPA in college students (DeFreitas & Bravo, 2012). Teacher variables (e.g., caring, interest promotion, and content explanations) contributed to Latina/os math achievement similar to math self-efficacy (Riconscente, 2014). Cultural congruity and college environment were also other variables associated with academic outcomes (Edman & Brazil, 2009). More research is also needed to determine the influence of contextual variables (e.g., teachers caring, teachers’ support, mentoring, college environment, cultural congruity) in Latina/os’ academic performance and success.

Additional research is needed to examine the relationship between self-efficacy and academic achievement during early childhood. In this review, only one study focused on this relationship during this developmental period, and it was unable to find a relationship or prediction between self-efficacy (academic, math, and reading) and academic achievement (Rivera, 2012). This may be due to the fact that second-grade students might be too young to develop interpretations and perceptions associated with self-efficacy (Rivera, 2012). Findings of previous research with elementary students are mixed. In a longitudinal study that began in first grade and finished in third grade, academic self-efficacy was positively correlated with reading and math achievement within and across time (Liew, McTigue, Barrois, & Hughes, 2008). Another study, with first- through third-grade students at risk for reading failure, also found a significant impact but small effect of academic self-efficacy on academic reading (Lee & Jonson-Reid, 2016). However, other studies have found the strength of the relationship between reading self-efficacy and reading achievement was not particularly strong in third-grade students (Smith, Smith, Gilmore, & Jameson, 2012). Smith et al.,’s (2012) findings indicated that third-grade students did not have a strong sense of their reading abilities or they were more influenced by classroom factors rather than an objective sense of their abilities as readers. Although children in early ages can answer questions of self-efficacy beliefs related to academic performance domains (Lee & Jonson-Reid, 2016), there are also environmental factors that can affect a proper interpretation of children’ efficacy beliefs.

The construction or modification of scales to measure specific or global domains of self-efficacy should consider content or task-specific self-efficacy (Bandura, 2006) and characteristics of the Hispanic culture (Dopke, 2001). Global measures of self-efficacy that include questions from different fields may affect the predictability of self-efficacy because a student might feel very confident in one content area and less confident in another one, affecting the predictability of this variable (Rivera, 2012). Also, “applying principles that have been effective in explaining the performance of samples of predominately White students does not appear to be acceptable” (Stevens et al., 2006, p. 181) for Hispanic students. Conversely, modified instruments or including variables or re-writing questions that correspond with an ethnic language may improve the accuracy of the instrument. For example, Dopke (2001) changing the item “English grammar” to “Spanish grammar” could lead to Latina/o students feeling more confident when they recognize that characteristics of their culture are valued.

Limitations

There are some limitations of the current review. First, this study resulted from the selection of articles that have been published in journal and theses/dissertations. However, it is possible that some articles relevant to this review were not identified in this search. Second, this review focused on studies that inquired about the relationship between Latina/os’ self-efficacy beliefs and academic performance exclusively in the United States. More reviews are needed to look at these relationships in Hispanic/Latina/o countries. Latina/o is not a homogeneous culture; therefore, to explore Latina/os’ efficacy beliefs about their capacities in academic matters in other countries and local cultures may provide insightful information. In addition, in-depth analysis of Latina/o cultural differences within the United States based on country of origin may provide stronger findings on self-efficacy and academic performance. Third, this review only examined the influence of self-efficacy as a motivational variable on academic performance. The study of cognitive, affective, and contextual variables may reveal differences or similarities of the contributions of these factors on Latina/os’ academic performance. Finally, this review did not examine the influence of cultural identity and acculturation processes on Latina/os’ academic performance. Several studies presented how immigrant status, cultural identity, and acculturation process influence the relationships between self-efficacy and academic performance. Future research may inquire these relationships.

Conclusion

This is the first review that focuses on the relationships between self-efficacy and academic performance in Latina/o students. Despite its limitations, this study confirms the overwhelming contribution of self-efficacy on Latina/o students’ academic performance at all educational levels. The positive impact of self-efficacy beliefs on students’ academic achievement should consider the development of educational environments and programs that foster Latina/os’ self-efficacy beliefs in their capacities to success academically. Findings from this study confirm that Latina/o is not a homogeneous group; therefore, further research is required to inquire ethnic identity, acculturation process, immigration status, family’ experience on higher education, affective variables, and parents’ and teachers’ support which significantly correlate with Latina/os’ academic achievement and/or academic success. Finally, longitudinal studies that focus on changes in academic self-efficacy in the transition from elementary to middle school, middle to high school, and high school to college are necessary to illuminate patterns of self-efficacy and the correlation with an effect on Latina/o students’ academic achievement and academic success.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Harold Manzano-Sanchez is an auxiliary professor in the Institute of Education and Pedagogy at the Universidad del Valle in Cali Colombia. His current research focuses on barriers that hinder college aspirations and supports that foster career choices in Latino high school students. He also focuses on the impact of recreation programs in the development of young people who live in low-income areas.

Corliss Outley is an associate professor in the Department of Recreation, Parks and Tourism Sciences at Texas A&M University and serves as the associate head for the Department. She conducts research on youth with particular interest in ethnic identity, built and physical environmental influences, and community-based participatory research involving youth. She considers herself an “engaged scholar” who focuses on real problems within communities. She earned her BS in biology at Grambling State University, Louisiana; her MS in forestry from Southern Illinois University–Carbondale; and her PhD in recreation and natural resources development from Texas A&M University.

Jorge E. Gonzalez is an associate professor in the School Psychology Program at the University of Houston. His scholarship focuses on developing and accelerating diverse young children’s oral vocabulary through interactive content-rich and explicit instructional approaches. His research contributes to a body of work on early childhood instructional approaches that promote and accelerate vocabulary development for low-income and/or English or dual language learners (ELL/DLL) through thematic content-enriched explicit interactive book reading. He is also interested in ecocultural models of Latino children’s home literacy environments. Two themes underlie this work: interventions that develop and accelerate vocabulary and background knowledge for Latino, especially Spanish-speaking children, and the role and measurement of the home literacy environment in relation to children’s vocabulary development.

David Matarrita-Cascante is an associate professor in the Department of Recreation, Park, and Tourism Sciences at Texas A&M University. His research uses community development and natural and environmental sociology frameworks. His work sits at the intersection of the above interests and various literatures including international development, amenity migration, tourism, protected areas, natural events, and community health.

References

*Acoach

C. L.

Webb

L. M.

(2004). The influence of language brokering on Hispanic teenagers’ acculturation, academic performance, and nonverbal decoding skills: A preliminary study. Howard Journal of Communications, 15, 1-19. doi:10.1080/10646170490275459

*Aguayo

Herman

Ojeda

Flores

L. Y.

(2011). Culture predicts Mexican Americans’ college self-efficacy and college performance. Journal of Diversity in Higher Education, 4, 79-89. doi:10.1037/a0022504

*Ayro

(2012). A case study of five Latina/o male college students: What contributes to their educational success? ProQuest LLC, Ed.D. Dissertation, University of Southern California.

Bachman

J. G.

(1970). Youth in transition, Vol. 2. The impact of family background and intelligence on tenth-grade boys. Ann Arbor, MI: Institute for Social Research.

Bandura

(2006). Adolescent development from an agentic perspective. In Pajares

Urdan

(Eds.), Self-efficacy beliefs of adolescents (pp. 1-43). Greenwich, CT: Information Age.

Bandura

(1990). Multidimensional Scales of Perceived Self-efficacy. Stanford, CA: Stanford University.

Britner

S. L.

Pajares

(2001). Self-efficacy beliefs, motivation, race, and gender in middle school science. Journal of women and Minorities in Science and Engineering, 7. DOI: 10.1615/JWomenMinorScienEng.v7.i4

Brown

S. D.

Lent

R. W.

(2006). Preparing adolescents to make career decisions. A social cognitive perspective. In Pajares

Urdan

(Eds.), Self-efficacy beliefs of adolescents (pp. 201-223). Greenwich, CT: Information Age.

Brown

S. D.

Tramayne

Hoxha

Telander

Fan

Lent

R. W.

(2008). Social cognitive predictors of college students’ academic performance and persistence: A meta-analytic path analysis. Journal of Vocational Behavior, 72, 298-308.

10.

*Buriel

Perez

De Ment

T. L.

Chavez

D. V.

Moran

V. R.

(1998). The relationship of language brokering to academic performance, biculturalism, and self-efficacy among Latino adolescents. Hispanic Journal of Behavioral Sciences, 20, 283-297.

11.

*Castro-Villarreal

Guerra

Sass

Hseih

(2014). Models of pre-service teachers’ academic achievement: The influence of cognitive motivational variables. Journal of the Scholarship of Teaching and Learning, 14, 71-95.

12.

*Cavazos

Jr. Johnson

M. B.

Fielding

Cavazos

A. G.

Castro

Vela

(2010). A qualitative study of resilient Latina/o college students. Journal of Latinos and Education, 9, 172-188.

13.

Chesnut

S. R.

Burley

(2015). Self-efficacy as a predictor of commitment to the teaching profession: A meta-analysis. Educational Research Review, 15, 1-16.

14.

*Chun

Dickson

(2011). A psychoecological model of academic performance among Hispanic adolescents. Journal of Youth and Adolescence, 40, 1581-1594.

15.

*Chun

Marin

M. R.

Schwartz

J. P.

Pham

Castro-Olivo

(2016). Psychosociocultural structural model of college success among Latina/o students in Hispanic-serving institutions. Journal of Diversity in Higher Education, 9, 385-400. doi:10.1037/a0039881

16.

*DeFreitas

S. C.

Bravo

Jr. (2012). The influence of involvement with faculty and mentoring on the self-efficacy and academic achievement of African American and Latino college students. Journal of the Scholarship of Teaching and Learning, 12(4), 1-11.

17.

*Dopke

S. A.

(2001). Self-efficacy, goal setting, and academic achievement: Mexican-American parents and their children (Doctoral Dissertation) Northern Illinois.

18.

*Edman

J. L.

Brazil

(2009). Perceptions of campus climate, academic efficacy and academic success among community college students: An ethnic comparison. Social Psychology of Education: An International Journal, 12, 371-383.

19.

*Fong

C. J.

Zientek

L. R.

Yetkiner Ozel

Z. E.

Phelps

J. M.

(2015). Between and within ethnic differences in strategic learning: A study of developmental mathematics students. Social Psychology of Education: An International Journal, 18, 55-74.

20.

Galyon

C. E.

Blondin

C. A.

Yaw

J. S.

Nalls

M. L.

Williams

R. L.

(2012). The relationship of academic self-efficacy to class participation and exam performance. Social Psychology of Education, 15, 233-249.

21.

*Garriott

P. O.

Flores

L. Y.

(2013). The role of social cognitive factors in Mexican American students’ educational goals and performance: A longitudinal analysis. Journal of Latina/o Psychology, 1, 85-94. doi:10.1037/a0032378

22.

*Hewitt

L. A. L.

(2009). The effects of bilingual education on language, achievement, and self-efficacy of Hispanic students (2009-99150-539) Texas A&M University, ProQuest Dissertations Publishing, 2008. 3347925.

23.

Honicke

Broadbent

(2016). The influence of academic self-efficacy on academic performance: A systematic review. Educational Research Review, 17, 63-84.

24.

Jonson-Reid

Davis

Saunders

Williams

(2005). Academic self-efficacy among African American youths: Implications for school social work practice. Children & Schools, 27, 5-14.

25.

Jinks

Morgan

(1999). Children’s perceived academic self-efficacy: An inventory scale. The Clearing House: A Journal of Educational Research, Controversy, and Practices, 72, 224-230.

26.

Karabenik

S. A.

Maehr

M. L.

(2007). Final report to the National Science Foundation: MSP—Motivation Assessment Program. Retrieved from: http://ma.mspnet.org/media/data/MSP-MAP_Final_Report.pdf?media_000000006004.pdf

27.

Lewis

Ream

R. K.

Bocian

K. M.

Cardullo

R. A.

Hammond

K. A.

Fast

L. A.

(2012). Con cariño: Teacher caring, math self-efficacy, and math achievement among Hispanic English learners. Teachers College Record, 114, 1-42.

28.

Lee

Y. S.

Jonson-Reid

(2016). The role of self-efficacy in reading achievement of young children in urban schools. Child and Adolescent Social Work Journal, 33, 79-89.

29.

*Lewis

J. L.

Ream

R. K.

Bocian

K. M.

Cardullo

R. A.

Hammond

K. A.

Fast

L. A.

(2012). Con carino: Teacher caring, math self-efficacy, and math achievement among Hispanic English learners. Teachers College Record, 114(7).

30.

Liew

McTigue

E. M.

Barrois

Hughes

J. N.

(2008). Adaptive and effortful control and academic self-efficacy beliefs on achievement: A longitudinal study of 1st through 3rd graders. Early Childhood Research Quarterly, 23, 515-526.

31.

Marat

(2005). Assessing mathematics self-efficacy of diverse students form secondary schools in Auckland: Implications for academic achievement. Issues in Educational Research, 15, 37-68.

32.

*McNamara

M. W.

(2012). Self-efficacy, stress, and acculturation as predictors of first year science success among Latina/os at a South Texas university (Doctoral dissertation). Texas A&M University–Corpus Christi.

33.

McWhirter

E. H.

Luginbuhl

P. J.

Brown

(2014). ¡Apóyenos! Latina/o student recommendations for high school supports. Journal of Career Development, 41, 3-23.

34.

McWhirter

E. H.

Ramos

Medina

(2013). ¿Y ahora qué? Anticipated immigration status barriers and Latina/o high school students’ future expectations. Cultural Diversity & Ethnic Minority Psychology, 19, 288-297.

35.

*Mejia Arias

E. E

. (2007). The relationship of perceived parent and family support, support of university environment, and academic self-efficacy on the academic achievement of Latina/o college students. Dissertation Abstracts International: Section A. Humanities and Social Sciences, 67(9-A), 3300.

36.

Metheny

McWhirter

E. H.

(2013). Contributions of social status and family support to college students’ career decision self-efficacy and outcome expectations. Journal of Career Assessment, 21, 378-394.

37.

Midgley

Maehr

M. L.

Hruda

L. Z.

Anderman

Freeman

K. E.

Urdan

(2000). Manual for the patterns of adaptive learning scales. Ann Arbor, 1001, 48109-1259.

38.

*Migray

(2003). The relationships among math self-efficacy, academic self-concept and math achievement (Unpublished doctoral dissertation). Arizona State University, USA.

39.

Moher

Liberati

Tetzlaff

Altman

D. G.

& Prisma Group. (2009). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. PLoS Medicine, 6(7), e1000097.

40.

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 of Psychology, 38, 30-38.

41.

Murillo

L. A.

Schall

J. M.

(2016). “They didn’t teach us well”: Mexican-origin students speak out about their readiness for college literacy. Journal of Adolescent & Adult Literacy, 60, 315-323.

42.

Muris

(2001). A brief questionnaire for measuring self-efficacy in youths. Journal of Psychopathology and Behavioral Assessment, 23, 145-149.

43.

National Center for Education Statistics. (2015). Public High School Graduation Rates. Washington, DC. Retrieved from http://nces.ed.gov/programs/coe/indicator_coi.asp

44.

*Niehaus

Rudasill

K. M.

Adelson

J. L.

(2012). Self-efficacy, intrinsic motivation, and academic outcomes among Latino middle school students participating in an after-school program. Hispanic Journal of Behavioral Sciences, 34, 118-136.

45.

Nora

Crisp

(2012). Future research on Hispanic students: What have we yet to learn? and what new and diverse perspectives are needed to examine Latina/o success in higher education? (White paper prepared for the Hispanic Association of Colleges and Universities).

46.

Ojeda

Flores

L. Y.

Navarro

R. L.

(2011). Social cognitive predictors of Mexican American college students’ academic and life satisfaction. Journal of Counseling Psychology, 58, 61-71.

47.

Owen

S. V.

Froman

R. D.

(1988). Development of a College Academic Self-Efficacy Scale. Paper presented at the Annual Meeting of the National Council on Measurement in Education. New Orleans LA.

48.

Padilla

A. M.

(2006). Bicultural social development. Hispanic Journal of Behavioral Sciences, 28, 467-497.

49.

Pajares

Graham

(1999). Self-efficacy, motivation constructs, and mathematics performance of entering middle school students. Contemporary educational psychology, 24(2), 124-139.

50.

Pintrich

P. R.

Smith

D. A. F.

Garcia

McKeachie

W. J.

(1991). A Manual for the Use of the Motivated Strategies for Learning Questionnaire (MSLQ). Ann Arbor, MI: The Regents of the University of Michigan.

51.

Reyes

N. A. S.

Nora

(2012). Lost among the data: A review of Latina/o first generation college students. Hispanic Association of Colleges and Universities (nd): n. pag. Latina/o first generation college students. White paper prepared for the Hispanic Association of Colleges and Universities, July 2012. San Antonio, TX: The University of Texas at San Antonio.

52.

Richardson

Abraham

Bond

(2012). Psychological correlates of university students’ academic performance: A systematic review and meta-analysis. Psychological Bulletin, 138, 353-387.

53.

*Riconscente

M. M.

(2014). Effects of perceived teacher practices on Latino high school students’ interest, self-efficacy, and achievement in mathematics. The Journal of Experimental Education, 82, 51-73.

54.

*Rivera

V. Y.

(2012). The impact of teachers’ expectations, parents’ expectations, and academic self-efficacy on the achievement of English language learners (Doctoral dissertation). Texas A&M University, College Station.

55.

Robbins

S. B.

Lauver

Davis

Langley

Carlstrom

(2004). Do psychosocial and study skill factors predict college outcomes? A meta-analysis. Psychological Bulletin, 130, 261-288.

56.

*Roberts

W. S.

III . (2014). Academic achievement and self-efficacy among diverse populations (Doctoral dissertation). University of West Florida, Pensacola.

57.

Ross

Kena

Rathbun

KewalRamani

Zhang

Kristapovich

Manning

(2012). Higher education: Gaps in Access and Persistence Study (Statistical Analysis Report, NCES 2012-046). National Center for Education Statistics. Washington, DC.

58.

Sáenz

V. B.

Ponjuan

(2012). Latino males: Improving college access and degree completion—A new national imperative (Perspectivas: Issues in Higher Education Policy and Practice, No. 1) Tempe, AZ.

59.

*Salas

(2001). Mexican American students’ perspectives: School success as a function of family support, caring teachers, rigorous curricula and self-efficacy (Doctoral dissertation). University of Texas at Austin.

60.

Schneider

Martinez

Owens

(2006). Barriers to educational opportunities for Hispanics in the United States. In Tienda

Mitchell

(Eds.), Hispanics and the future of America: Panel of Hispanics in the United States (pp. 179-227). Washington, DC: The National Academies Press.

61.

Sheu

H. B.

Lent

R. W.

Brown

S. D.

Miller

M. J.

Hennessy

K. D.

Duffy

R. D.

(2010). Testing the choice model of social cognitive career theory across Holland themes: A meta-analytic path analysis. Journal of Vocational Behavior, 76, 252-264.

62.

Smith

J. K.

Smith

L. F.

Gilmore

Jameson

(2012). Students’ self-perception of reading ability, enjoyment of reading and reading achievement. Learning and Individual Differences, 22, 202-206.

63.

Solberg

O’Brien

Villareal

Kennel

Davis

(1993). Self-efficacy and Hispanic college students: Validation of the college self-efficacy instrument. HispanicJournal of Behavioral Sciences, 15, 80795.

64.

Song

Hooper

Loke

(2013). Publication bias: What is it? How do we measure it? How do we avoid it? Open Access Journal of Clinical Trials, 2013(5), 71-81.

65.

*Stevens

Olivárez

Jr. Lan

W. Y.

Tallent-Runnels

M. K.

(2004). Role of mathematics self-efficacy and motivation in mathematics performance across ethnicity. The Journal of Educational Research, 97, 208-222.

66.

*Stevens

Olivárez

Jr. Hamman

(2006). The role of cognition, motivation, and emotion in explaining the mathematics achievement gap between Hispanic and White students. Hispanic Journal of Behavioral Sciences, 28, 161-186.

67.

Wigfield

Eccless

J. S.

Yoon

K. S.

Harold

R. D.

Arbreton

A. J. A.

Freedman-Doan

et al . (1997). Change in children’s competence beliefs and subjective task values across the elementary school years: A 3-year study. Journal of Educational Psychology, 89, 451-469.

68.

Woodcock

R.W.

Johnson

M.B.

(1977). Woodcock-Johnson Tests of Achievement. The Riverside Publishing Company: CA.

69.

Woodcock

R. W.

McGrew

K.S.

Mather

(2001). Woodcock-Johnson III Tests of Achievement. Itasca, IL: Riverside Publishing.

70.

*Zell

M. C.

(2010). Achieving a college education: The psychological experiences of Latina/o community college students. Journal of Hispanic Higher Education, 9, 167-186.

71.

*Zychinski

K. E.

Polo

A. J.

(2012). Academic achievement and depressive symptoms in low-income Latino youth. Journal of Child and Family Studies, 21,565-577.