Social-cognitive career theory variables as mediators for the relationship between deep learning and goal persistence in African American college students with disabilities

Abstract

BACKGROUND:

Education is one of the best pathways to middle class earnings. The Workforce Innovation and Opportunity Act requires state vocational rehabilitation agencies to reserve and expend at least 15% of its State allotment for pre-employment transition services to students with disabilities, including enrollment in comprehensive transition or postsecondary educational programs.

OBJECTIVE:

To evaluate the social-cognitive career theory (SCCT) variables of academic barriers coping self-efficacy, academic milestone self-efficacy, and academic outcome expectancy as mediators for the relationship between deep learning and goal persistence in a sample of African American college students with disabilities.

METHOD:

Quantitative descriptive research design using serial multiple mediation analysis (SMMA).

RESULTS:

SMMA results indicated that deep learning was significantly linked to goal persistence (total effect). Also, he direct effect between deep learning and goal persistence was still significant after controlling for the effects of the mediators, indicating deep learning is a strong predictor of goal persistence, and SCCT variables only partially mediated the relationship between deep learning and goal persistence. The mediation effects were estimates of the indirect effects for deep learning on goal persistence through (a) academic barriers coping self-efficacy, (b) academic milestone self-efficacy, (c) academic outcome expectancy, and (d) academic barriers coping self-efficacy, academic milestone self-efficacy and academic outcome expectancy.

CONCLUSION:

Findings of this study indicated that higher levels of deep learning had the benefit of helping African American college students with disabilities develop academic barriers coping self-efficacy, academic milestone self-efficacy and academic outcome expectancy, leading to goal persistence.

Keywords

Deep learning postsecondary education minority students social-cognitive career theory serial multiple mediation

1 Introduction

Poverty, unemployment and income inequality are major causes of human suffering among people with or without disabilities (Chan et al., 2017; Compton, Gfroerer, Conway, & Finger, 2014). There is strong research evidence to support that unemployment increases the risk of poverty and contributes to income inequality (Ryscavage, 1982). Psychosocial consequences of poverty and unemployment include depression, anxiety, domestic violence, drug use, and social relationship problems (Belle & Bullock, 2017; Compton et al., 2014; Weich & Lewis, 1998). In addition, people with lower incomes report poorer health and have a higher risk of illness and injury (Woolf et al., 2015), and income is a major reason for health disparities among racial and ethnic minority groups (Centers for Disease Control and Prevention [CDC], 2017). Unfortunately, the employment-to-population ratio of 17.9% for people with disabilities is significantly lower than the 65.3% for people without disabilities (U.S. Bureau of Labor Statistics, 2017a), and the poverty rate of 21.2% for people with disabilities is significantly higher than the 13.8% for people without disabilities (Kraus, 2017). The high prevalence of mental illness and secondary health problems due to unemployment and poverty is a major public health concern (Compton et al., 2014). Employment is being increasingly considered a rehabilitation and public health intervention for people with disabilities (Ditchman et al., 2013; Muller et al., 2017).

Customized training/apprenticeship, technical education, and college education have been established as one of the best pathways to middle-class earnings for people with disabilities (O’Neill et al., 2014; Petrilli, 2016). According to the U.S. Census Bureau, the impact of education level on work-life earnings surpasses that of all other demographic variables (Julian & Kominski, 2011). In 2017, the unemployment rate for people over 25 with a bachelor’s degree was 2.5%, compared to 5.3% for people who graduated high school but did not attend college and 7.7% for high school dropouts (U.S. Bureau of Statistics, 2017b). Although the high school graduation rate for students with disabilities has improved to 64.6% for the 2014–2015 school year, it is still roughly 20% lower than the national average (Danilova, 2017). Furthermore, while the national graduation rate for the year 2015 was 83.2%, it was only 77.8% for Hispanic students and 74.6% for black students (Danilova, 2017). Clearly, helping students with disabilities, especially students from racial and ethnic minority backgrounds, graduate from high school and ease their transition to postsecondary education or employment should be a high priority for special educators, transition specialists, accommodation specialists, and rehabilitation counselors (O’Neill et al., 2014). There are several learning approaches (e.g., deep approach vs. surface approach of learning) and career development theories (e.g., social-cognitive career theory) that are potentially useful for educators and counselors to gain better insights about factors influencing goal persistence of college students with disabilities from racial and ethnic minority backgrounds.

1.1 Deep approach to learning

The deep approach to learning is a prominent model of learning frequently described in the pedagogical literatures of higher education (Biggs, 1987; Entwistle, 2010; Haggis, 2003; Marton & Säljö, 1976; Stanger-Hall, 2012). According to Bain (2012), students who are engaging and learning the material motivated by deep interest and a sincere desire to truly know are more successful in college and in life. These intrinsic motivating factors for learning are more conducive to college success and lifelong learning than extrinsic motivation factors (such as high grades, academic awards or accolades; Bain, 2012).

Marton and Säljö (1997) conducted a qualitative research study with college students and identified six major themes that underlie how students conceptualize the idea of learning. These six conceptualizations of learning were: (a) a quantitative increase in knowledge, (b) memorization, (c) acquisition of facts for subsequent use, (d) abstraction of meaning, (e) a process aimed at understanding reality, and (f) developing as a person. This hierarchy of learning can be classified into two basic approaches, with quantitative, memorizing and acquisition conceptions describing a “surface” learning approach, and abstraction, understanding reality and development as a person defining a “deep” learning approach. Deep learning involves placing meaning on things by exploring how they fit together. There is emerging evidence to support that the use of predominantly surface motives and strategies to learning leads to poor learning outcomes, while the use of deep approaches result in good college completion, employment, and lifelong learning outcomes (Biggs & Tang, 2007; Floyd, Harrington, & Santiago, 2009; Haggis, 2003).

1.2 Social-cognitive career theory

The social-cognitive career theory (SCCT) is one of the most central career development theories developed to explain how people form vocational interests, set career goals, develop self-efficacy and intention to persist in the educational and work environments (Betz, 2007; Lent, Brown, & Hackett, 1994, 2000; Lent, 2005; Lent et al., 2008). It is an extension of Albert Bandura’s (1986) general social cognitive theory adapted by vocational psychology researchers to elucidate academic performance and career development. Bandura (1986) postulated that four types of background learning experiences (performance accomplishments, vicarious learning, social persuasion, and physiological arousal) would lead to the development of self-efficacy for a given behavior or domain of behavior (e.g., academic milestone self-efficacy). Self-efficacy, in turn, influenced one’s career choice, level of academic performance, outcome expectancy and goal persistence (Cardoso et al., 2013). Social-cognitive career theory is composed of three constructs (a) interest development, (b) choice-making, and (c) performance, and three contextual factors (a) person (e.g., self-efficacy), (b) environmental (e.g., social support), and (c) behavioral (e.g., goal implementation). The SCCT has been used to successfully study women and underrepresented minority students’ decision to pursue science and technology careers (Byars-Winston, Estrada, Howard, Davis, & Zalapa, 2011; Lent et al., 2005, 2008), and it was found to predict goal persistence in a sample of college students from racial and ethnic minority backgrounds (Cardoso et al., 2013).

1.3 Purpose of the study

As mentioned, education is one of the best pathways to middle class earnings. It is for this reason that the Rehabilitation Act, as amended by the Workforce Innovation and Opportunity Act (2014), requires state vocational rehabilitation agencies to reserve and expend at least 15% of its State allotment for pre-employment transition services to students with disabilities, including enrollment in comprehensive transition or postsecondary educational programs. In order for transition-age youth with disabilities make a smooth transition to postsecondary education, it is necessary to help them develop an appreciation for the deep approach to learning and to successfully persist in college is necessary. The purpose of the present study was to identify specific factors that influence the development of the deep approach to learning and goal persistence by evaluating three key SCCT constructs (academic barriers coping self-efficacy, academic milestone self-efficacy, and academic outcome expectancy) as serial multiple mediators for the relationship between deep learning and goal persistence. The following research questions guided the theoretical examination:

What is the relationship between deep learning and goal persistence?

Can the relationship between deep learning and goal persistence be mediated by academic barriers coping self-efficacy, academic milestone self-efficacy, and academic outcome expectancy?

2 Methods

2.1 Participants

Participants consisted of 62 African American students with disabilities enrolling in a historically black university (HBCU). The sample included 22 (35.5%) men and 40 (64.5%) women, with a mean age of 22.19 years (SD = 6.60). There were 53 freshmen (85.5%), 8 sophomores (12.9%), and one junior (1.6%). In terms of academic majors, there were 32 students (53.2%) in Science, Technology, Engineering, Mathematic and Medicine (STEMM); 12 students (19.4%) in Business; 6 students (9.7%) in Social Services; 5 students (8.1%) in Education; 4 students (6.5%) in Criminal Justice; and 2 students (3.2%) in English.

2.2 Measures

2.2.1 Deep approach to learning

Deep approach was assessed by the Deep Factor Subscale of the Revised Study Process Questionnaire (R-SPQ; Biggs, Kember, & Leung, 2001), which is a modified version of the Study Process Questionnaire (SPQ; Biggs, 1987). The Deep Factor Subscale comprises 10 items and assesses two factors deep strategy and deep motive, each with five items (e.g., deep strategy, “I find that I have to do enough work on a topic so that I can form my own conclusions before I am satisfied;” deep motive, “I find that at times studying gives me a feeling of deep personal satisfaction.”) Each item is rated on a five point Likert-type rating scale, ranging from 1 (never or only rarely true of me) to 5(always or almost true of me). Scores are calculated by summing up the items with the higher score indicating a higher level of deep learning. The internal consistency reliability coefficient (Cronbach’s alpha) was 0.63 for deep strategy and 0.62 for deep motive (Biggs et al., 2001). The Cronbach’s alpha for the Deep Factor subscale scores was computed to be 0.83 in the present study.

2.2.2 Self-efficacy

Self-efficacy was assessed with two scales: the 4-item Academic Milestone Self-Efficacy Scale (Lent, Brown, & Larkin, 1986) and the 7-item Academic Barriers Coping Self-Efficacy Scale (Lent, 2001; Lent et al., 2003). The academic milestone self-efficacy items asked students to indicate their confidence in their ability to complete academic requirements in their declared majors (e.g., “How much confidence do you have in your ability to excel in your major area of study over the next semester?”). On the academic barriers coping self-efficacy items, participants rated their confidence in their ability to cope with a variety of barriers that university students might experience (e.g., “cope with a lack of support from professors or your advisor”). All self-efficacy ratings were obtained on a 10-point scale, from 0 (no confidence) to 9 (complete confidence). Lent et al. (2005) found that the combined self-efficacy scale yielded a coefficient alpha estimate of 0.91 and correlated in theoretically expected directions with measures of outcome expectations, interests, goals, and social supports and barriers relative to pursuit of engineering majors. Cronbach’s alpha was 0.90 for the academic milestone self-efficacy subscale and 0.84 for the academic barriers coping self-efficacy scale in the present study.

2.2.3 Outcome expectancy

Outcome expectations were measured by the 6-item Academic Outcome Expectations Scale adapted from Lent et al.’s (2003) STEM Outcome Expectations Scale. Instead of science and technology, students were asked to indicate how strongly they agreed that their college degree would lead to each of the six positive outcomes, such as “a job with good pay and benefits.” Ratings were made on a 5-point agreement scale, from 1 (strongly disagree) to 5 (strongly agree). Lent et al. (2003, 2005) reported the academic outcome expectations scale yielded adequate internal consistency reliability estimates (coefficient alpha of 0.89 to 0.91) and related to measures of task and coping efficacy, interests, and major choice goals (Lent et al., 2003, 2005). The Cronbach’s alpha coefficient for the academic outcome expectations scale in the present study was estimated at 0.92.

2.2.4 Goal persistence

Goal persistence was adapted from Lent et al.’s (2003) Goal Persistence Scale for Science and Technology majors. Participants rated their level of agreement from 1 (strongly disagree) to 5 (strongly agree) with 8 statements about their academic plans (e.g., “I will complete my college degree.”). Lent et al. (2003, 2005) reported that the goal persistence measure yielded satisfactory estimates of internal consistency reliability (Cronbach’s alpha of 0.93 to 0.95) and correlated with self-efficacy in the theoretically expected direction. In addition, Lent et al. (2003) found this measure to be useful in predicting college completion in engineering majors. The Cronbach’s alpha coefficient for the goal persistence scale in the present study was estimated at 0.74.

2.3 Procedures

This study was conducted at a HBCU in the southern part of the United States after receiving Institutional Review Board approval. Career assessment data were collected from students with disabilities registered in any of the nine English 110 classes who self-identified as having a disability and volunteered to participate in the study.

2.4 Data analysis

The Statistical Package for the Social Sciences (SPSS 24.0) for Windows was used for descriptive statistics and correlation coefficients. The SPSS PROCESS v2.16 macro for SPSS written by Andrew Hayes (2013) was used to estimate the total, direct, and indirect effects recommended by Hayes (2012, 2013) for testing serial multiple mediation hypotheses. Mediation analyses were performed using ordinary least squares (OLS) regression to investigate academic barriers coping self-efficacy, academic milestone self-efficacy, and academic outcome expectancy as mediators of the relationship between deep learning and goal persistence. An a priori power analysis was conducted for the total R² value for the OLS regression analysis with three predictor variables, power of 0.80, and an alpha level of 0.05. Results from this analysis using GPOWER 3.1.4 (Faul, Erdfelder, Lang, & Buchner, 2007), a software tool for a general power analysis, suggested that a sample size of 53 would be needed to detect a medium to large effect size (f²= 0.25; Cohen, 1988). In the present study, a medium to large effect size was expected based on findings observed in SCCT studies with women, minorities, and college students with disabilities (Cardoso et al., 2013; Hackett & Betz, 1981; Kerr & Kurpius, 2004; Lent, 2005; Lent et al., 2008; Mullikin, Bakken, & Betz, 2007; Turner & Lapan, 2005).

3 Results

3.1 Preliminary analysis

Findings from the bivariate analysis revealed that variables in the present study were significantly correlated with each other, ranging from r = 0.32 (moderate effect size) to r = 0.64 (large effect size). The correlation matrix for the predictor, mediators, and outcome variable are presented in Table 1.

Table 1
Correlations of the predictor, serial mediators, and outcome variable (N = 62)

1 2 3 4 5

1. Deep learning –

2. Goal Persistence 0.53^** –

3. Academic Barriers Coping Self-Efficacy 0.31^* 0.32^* –

4. Academic Milestone Self-Efficacy 0.39^ 0.52^ 0.64^ –

5. Academic Outcome Expectancy 0.37^ 0.44^ 0.43^ 0.41^** –

Mean 6.34 3.83 6.42 6.30 4.30

SD 1.42 0.63 1.56 1.84 0.63

	1	2	3	4	5
1. Deep learning	–
2. Goal Persistence	0.53^**	–
3. Academic Barriers Coping Self-Efficacy	0.31^*	0.32^*	–
4. Academic Milestone Self-Efficacy	0.39^**	0.52^**	0.64^**	–
5. Academic Outcome Expectancy	0.37^**	0.44^**	0.43^**	0.41^**	–
Mean	6.34	3.83	6.42	6.30	4.30
SD	1.42	0.63	1.56	1.84	0.63

^*p≤0.05, ^**p < 0.01.

Specifically, deep learning was positively related to academic barriers coping self-efficacy (r = 0.31, p < 0.05), academic milestone self-efficacy (r = 0.39, p < 0.01), academic outcome expectancy (r = 0.37, p < 0.01), and goal persistence (r = 0.53, p < 0.01). Goal persistence was positively associated with academic barriers coping self-efficacy, academic milestone self-efficacy, and academic outcome expectancy (r = 0.32, p < 0.05, r = 0.52, p < 0.01, and r = 0.44, p < 0.01, respectively). Academic barriers coping self-efficacy and academic milestone self-efficacy were positively associated with academic outcome expectancy (r = 0.43, p < 0.01 and r = 0.41, p < 0.01, respectively). Academic barriers coping self-efficacy was positively associated with academic milestone self-efficacy (r = 0.64, p < 0.01).

3.2 Serial multiple mediation analysis

A serial multiple mediation (SMM) analysis was performed to test academic barriers coping self-efficacy, academic milestone self-efficacy, and academic outcome expectancy as mediators of the relationship between deep learning and goal persistence in a sample of African American college students with disabilities. The estimates of the indirect effects were for: (a) deep approach on goal persistence through academic barriers coping self-efficacy, through academic milestone self-efficacy, and through academic outcome expectancy; (b) through both academic barriers coping self-efficacy and academic milestone self-efficacy, through both academic milestone self-efficacy and academic outcome expectancy; and (c) through academic barriers coping self-efficacy, academic milestone self-efficacy, and academic outcome expectancy. The following are key terms for the path coefficients used to describe the direct, indirect, and total effects:

Direct effect of deep learning: c’

Specific indirect effect of deep learning through academic barriers coping self-efficacy: a₁b₁

Specific indirect effect of deep learning through academic milestone self-efficacy: a₂b₂

Specific indirect effect of deep learning through academic outcome expectancy: a₃b₃

Specific indirect effect of deep learning through academic barriers coping self-efficacy and academic milestone self-efficacy: a₁d₂₁b₂

Specific indirect effect of deep learning through academic milestone self-efficacy and academic outcome expectancy: a₂d₃₂b₃

Specific indirect effect of deep learning through academic barriers coping self-efficacy and academic outcome expectancy: a₁d₃₁b₃

Specific indirect effect of deep learning through academic barriers coping self-efficacy, academic milestone self-efficacy, and academic outcome expectancy: a₁d₂₁d₃₂b₃

Total indirect effect of deep learning: a₁b₁ + a₂b₂ + a₃b₃ + a₁d₂₁b₂ + ad₃₁b₃ + a₂d₃₂b₃

Total effect of deep learning:

The R² for the SMM analysis model was computed to be 0.28 (f²= 0.39), indicating a large effect size. A graphical representation of this model along with associated standardized path coefficients are presented in Fig. 1.

Fig.1

Path coefficients for the serial multiple mediation analysis on goal persistence. Note: Dotted line denotes the total effect of deep learning on goal persistence when academic barrier coping self-efficacy, academic milestone self-efficacy, and academic outcome expectancy are not included as serial multiple mediators. ^*p < 0.05; ^**p < 0.01; ⁺p > 0.05, n.s.

3.2.1 Total effect

As can be observed in Fig. 1, high levels of deep learning was associated with higher levels of goal persistence (c = 0.53, standard error [SE] = 0.11, p < 0.01).

3.2.2 Direct effects

There are four significant direct effects and five non-significant direct effects. The significant direct effects include the following paths:

Deep learning was directly linked to academic barriers coping self-efficacy (path a₁= 0.31).

Deep learning was directly linked to academic milestone self-efficacy (after controlling for academic barriers coping self-efficacy; path a₂= 0.21).

Academic barriers coping self-efficacy was positively linked to academic milestone self-efficacy (after controlling for the effect of deep learning; d₂₁= 0.57).

Academic milestone self-efficacy was directly linked to goal persistence (after controlling for the effect of deep learning, academic barriers coping self-efficacy, and academic outcome expectancy; path b₂= 0.37).

The non-significant direct paths are described below:

Deep learning was not linked to academic outcome expectancy (after controlling for the effect of academic barriers coping self-efficacy and academic milestone self-efficacy; path a₃ = 0.23).

Academic barriers coping self-efficacy was not associated with academic outcome expectancy (after controlling for the effects of academic milestone self-efficacy and deep learning; path d₃₁= 0.26)

Academic barriers coping self-efficacy was not associated with goal persistence (after controlling for the effects of deep learning, academic milestone self-efficacy, and academic outcome expectancy; path b₁= –0.12).

Academic milestone self-efficacy was not associated with academic outcome expectancy (after controlling for the effect of deep learning and academic coping self-efficacy; path d₃₂= 0.16).

Academic outcome expectancy was not linked to goal persistence (after controlling for the effect of deep learning, academic barriers coping self-efficacy, and academic milestone self-efficacy; path b₃= 0.21).

The total effect between deep learning and goal persistence (path c = 0.53) was appreciably reduced after controlling for the effect of the mediators (path c’ = 0.35), indicating that the serial multiple SCCT mediators partially mediated the effect of deep learning on goal persistence. The effects of these mediators were formally tested and the results are reported below.

3.2.3 Indirect effects

The mediation effects were estimates of the indirect effects for deep learning on goal persistence through (a) academic barriers coping self-efficacy; (b) academic milestone self-efficacy; (c) academic outcome expectancy; and (d) academic barriers coping self-efficacy, academic milestone self-efficacy, and academic outcome expectancy together. If the bias-corrected bootstrap confidence intervals (CI) for the products of these paths do not include zero, the specific indirect effects would be considered statistically significant (Hayes, 2013). The SPSS PROCESS procedure with 10,000 bootstrap samples revealed three significant indirect effects for the relationship between deep learning and goal persistence.

Through academic barriers coping and academic milestone self-efficacy (deep learning → [academic barriers coping self-efficacy → academic milestone self-efficacy] → goal persistence; point estimate = 0.07, 95% CI: 0.01 to 0.19).

Through academic barriers coping and academic outcome expectancy (deep learning → [academic barriers coping self-efficacy → academic outcome expectancy] → goal persistence; point estimate = 0.02, 95% CI: 0.00 to 0.08).

Through academic milestone self-efficacy (deep learning → [academic milestone self-efficacy] → goal persistence; point estimate = 0.08, 95% CI: 0.00 to 0.25).

4 Discussion

There is strong empirical evidence to support that apprenticeship, vocational/technical education, and college education provide one of the best career pathways to the middle class for people with disabilities (O’Neill et al., 2014). Transition-age youth with disabilities are graduating with standard diplomas and academic qualifications to attend higher education at an increasing rate (Wolanin & Steele, 2004). Despite these qualifications, students with disabilities are considered the most recently marginalized group to move toward equal opportunity in higher education (Wolanin & Steele, 2004). Helping students with disabilities, especially students with disabilities from racial and ethnic minority backgrounds, make a smooth transition to postsecondary education, persist in completing their educational programs, and successfully finding a job with good pay and benefits is a high priority of special educators, transition specialists, accommodation specialists, and rehabilitation counselors (O’Neill et al., 2014).

There are several learning approaches and career development theories that are potentially useful for educators and counselors to gain better insights about factors influencing goal persistence of minority college students with disabilities. In the present study, three SCCT variables (academic barriers coping self-efficacy, academic milestone self-efficacy, and academic outcome expectancy) were evaluated as serial multiple mediators for the relationship between deep learning and goal persistence in a sample of African American college students with disabilities. The results indicated a strong relationship between deep learning and goal persistence. Deep learning was positively associated with academic barriers coping self-efficacy and academic milestone self-efficacy, but not with academic outcome expectancy. The relationship between deep learning and academic outcome expectancy was mediated by academic barriers coping self-efficacy and academic milestone self-efficacy. Academic barriers coping self-efficacy was positively associated with academic milestone self-efficacy (after controlling for the effect of deep learning) but not with academic outcome expectancy. The total effect between deep learning and goal persistence was significantly reduced by the serial multiple mediators of academic barriers coping and academic milestone self-efficacy; by the serial multiple mediators of academic barriers coping self-efficacy and academic outcome expectancy; and by the single mediator of academic milestone self-efficacy alone. However, the direct effect (the relationship between deep learning and goal persistence after controlling for the significant mediators) remain significant indicating the serial multiple mediators only partially mediated the effect of deep learning on goal persistence. The R² for the SCCT mediation model was high indicating a large effect size.

The deep and surface approaches to learning is an important topic in the pedagogical literatures of higher education. Deep learning emphasizes the importance of examining new facts and ideas critically, integrating them into existing thinking on the topic, and identifying multiple associations between ideas (Houghton, 2004). Characteristics of deep learning include searching for meaning, bringing together the central argument or concepts needed to solve a problem, differentiating between argument and evidence, relating ideas to different learning modules, connecting new and previous knowledge, and applying course content to real life (Biggs, 1999; Entwistle, 1988; Houghton, 2004). Surface learning can be useful initially to help students develop foundational knowledge about a learning topic before they advance to deep approach to learning. Surface learning can be defined as accepting new facts and ideas uncritically and treating these ideas as unconnected items (Biggs, 1999; Entwistle, 1988; Houghton, 2004). Characteristics of surface learning include depending on rote learning, receiving information passively, failing to recognize new material as building on previous work, and seeing course content simply as materials to be learned for passing the examination (Biggs, 1999; Entwistle, 1988; Houghton, 2004). In the present study, SCCT variables partially explain the deep learning and goal persistence relationship in a sample of African American college students with disabilities, suggesting helping students develop a strong motivation for deep learning, along with the development of academic barriers coping self-efficacy, academic milestone self-efficacy, and academic outcome expectancy, can increase students’ goal persistence.

4.1 Implications for professional practice

Findings of the present study indicated that helping students to adopt the deep approach to learning could significantly increase their odds of completing postsecondary education, which will increase opportunities to find a good paying job with benefits and become a life-long learner in their career. Special educators, secondary transition specialists, accommodation specialists, and rehabilitation counselors can help students by providing support services related to deep learning, including study skills training, time management training, tutorial services, and lectures. Strategies including educational support groups should be provided to encourage students to be: (a) intrinsically curious about topics in their education; (b) determined to do well and mentally engaging when doing academic work; (c) interested in having the appropriate foundational knowledge; and (d) able to pursue diverse interests through good time management (Houghton, 2004). Rehabilitation counselors should also create opportunity for students to have positive experiences with education in order to increase students’ confidence in their ability to understand and succeed. Positive experience may include hands-on activities in the classroom or community, interviewing experts in a topic area to foster curiosity and mental engagement, or extending learning outside of the classroom to develop a stronger knowledge base.

The findings that SCCT variables partially mediated the relationship between deep learning and goal persistence may be particularly relevant for designing retention and career development interventions for underrepresented minorities with disabilities in pursuit of advanced education. Specifically, academic barriers coping self-efficacy, academic milestone self-efficacy, and academic outcome expectancy are strong predictors of goal persistence. Students’ self-efficacy and self-efficacy beliefs can be increased by counseling and academic support interventions. Therefore, early interventions designed to increase teacher and parent expectations, develop strong academic skills in minority children with disabilities, and promote interest in attending college at the primary and secondary school levels are crucial to the academic success of underrepresented minority students with disabilities. For college students with low academic barriers coping self-efficacy and academic milestone self-efficacy, providing social support services to encourage students to persevere and overcome barriers may be as important as providing effective study skills, time management, and remedial and tutoring services. Social support can be fostered by helping students develop empathetic and social self-efficacy and advocacy skills, and by educating academic staff and faculty about diversity and disability inclusion issues. Increasing academic self-efficacy through efforts such as these will increase outcome expectancy. Since outcome expectancy is significantly associated with career interest and since interest is related to goal persistence, increased outcome expectancy will indirectly increase goal persistence (Cardoso et al., 2013). Therefore, summer learning institutes, field trips, motivational speakers/role models, and internships should be incorporated in academic support programming for underrepresented minority students with disabilities to increase outcome expectations and goal persistence.

5 Conclusion

Poverty, race/ethnicity, and disability intersect to negatively affect representation of people with disabilities in higher education. Minority students with disabilities are the most recently marginalized group to move toward equal opportunity in higher education. The deep learning-SCCT frameworks provide invaluable information and practical guidelines for designing best practice educational and career development services to students with disabilities from racial and ethnic minority backgrounds to increase goal persistence, career development, and job placement.

Conflict of interest

None to report.

Footnotes

Acknowledgments

The contents of this article were developed with support from a Field-Initiated Research Grant funded by the National Institute on Disability, Independent Living and Rehabilitation Research Grant 90IF0103-02-00 to Southern University at Baton Rouge, Louisiana. The ideas, opinions, and conclusions expressed, however, are those of the authors and do not represent recommendations, endorsements, or policies of the U.S. Department of Health and Human Services.

It is to our great sorrow to report that Professor Alo Dutta passed away unexpectedly after the submission of the manuscript. She was the principal investigator of this research project and the lead writer of the manuscript. All co-authors unanimously agree that she deserves to be first author of this article.

References

Bain,

(2012) What the best college students do? Cambridge, MA: Harvard University Press/Belknap Press.

Bandura,

(1986), Social foundations of thought and action: A social cognitive theoryEnglewood Cliffs, NJ: Prentice-Hall.

Belle,

, & Bullock,

H. E.

(2017). The psychological consequences of unemployment. Society for the Psychological Study of Social Issues. Retrieved from https://www.spssi.org/index.cfm?fuseaction=page.viewpage&pageid=1457.

Betz,

N. E.

(2007) Career self-efficacy: Exemplary recent research and emerging directions. Journal of Career Assessment, 15, 403–422.

Biggs,

(1987) Student Approaches to Learning and Studying. Research Monograph. Australian Council for Educational Research Ltd., Radford House, Frederick St., Hawthorn 3122, Australia.

Biggs,

, & Tang,

(2007), Teaching for quality learning at university (3rd ed.). Buckingham: Open University Press.

Biggs,

, Kember,

, & Leung,

D. Y. P.

(2001) The revised two-factor Study Process Questionnaire: R-SPQ-F. British Journal of Educational Psychology, 71(1), 133–147.

Byars-Winston,

, Estrada,

, Howard,

, Davis,

, & Zalapa,

(2011) 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, 205–218.

Cardoso,

, Dutta,

, Chiu,

C. Y.

, Johnson,

E. T.

, Kundu,

, & Chan,

(2013) Social-cognitive predictors of STEM career interests and goal persistence in college students with disabilities from racial and ethnic minority backgrounds. Rehabilitation Research, Policy, & Education, 27, 271–284.

10.

Centers for Disease Control and Prevention (2017). Minority health determines the health of the nation. Retrieved from https://www.cdc.gov/aging/disparities/index.htm.

11.

Chan,

, Iwanaga,

, Umucu,

, Yaghmaian,

, Wu,

J.R.

, Bengtson,

, & Chen,

(2017) Evidence-based practice and research utilization. In Tarvydas

V. M.

& Hartley

M. T.

(Eds.), The professional practice of rehabilitation counseling (pp. 359–380). New York: Springer Publishing Company.

12.

Cohen,

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.

13.

Compton,

W. M.

, Gfroerer,

, Conway,

K. P.

, & Finger,

M. S.

(2014) Unemployment and substance outcomes in the United States 2002–2010. Drug and Alcohol Dependence, 142, 350–353.

14.

Danilova,

(2017). Low-income, minority students lag in high school graduation rates. The Morning Calls. Retrieved from http://www.mcall.com/news/nationworld/mchigh-school-graduation-rates-20170503-story.html.

15.

Ditchman,

, Wu,

, Chan,

, Fitzgerald,

, Lin,

C. P.

, & Tu,

W. M.

(2013) Vocational rehabilitation. In D.

Strauser.

(Ed.), Career development, employment, and disability in rehabilitation: From theory to practice, (pp. 343–360). New York: Springer Publishing Company.

16.

Entwistle,

(2010) Taking stock: An overview of key research findings. In Hughes

J. C.

and Mighty

(Eds.), Taking stock: Research on teaching and learning in higher education. Kingston, Ontario: School of Policy Studies, Queens University.

17.

Entwistle,

(1988), Styles of learning and teachingLondon, United Kingdom: David Fulton Publishers.

18.

Faul,

, Erdfelder,

, Lang,

A. -G.

, & Buchner,

(2007) G*Power: A flexible statistical power analysis for the social, behavioral, and biomedical sciences. Behavior Research Methods, 39, 175–191.

19.

Floyd,

, Harrington,

Santiago,

(2009) The effect of engagement and perceived course value on deep and surface learning strategies. Informing Science: The International Journal of an Emerging Transdiscipline, 39, 181–190.

20.

Hackett,

, & Betz,

N. E.

(1981) A self-efficacy approach to the career development of women. Journal of Vocational Behavior, 18, 326–339.

21.

Haggis,

(2003) Constructing images of ourselves? A critical investigation into ‘Approaches to Learning’ research in higher education. British Educational Research Journal, 29(1), 89–104.

22.

Hayes,

A. F.

(2013), Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford Press.

23.

Hayes,

A. F.

(2012). PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling. Retrieved from http://www.afhayes.com/public/process2012.pdf.

24.

Houghton,

(2004) Engineering subject centre guide: Learning and teaching theory for engineering academics. Loughborough: HEA Engineering Subject Centre.

25.

Julian,

, & Kominski,

(2011). Education and synthetic work-life earnings estimates. American Community Survey Reports, ACS-14. U.S. Census Bureau, Washington, DC.

26.

Kerr,

, & Kurpius,

S. E.

(2004). Encouraging talented girls in math and science: Effects of a guidance intervention. High Ability Studies, 15, 85–102.

27.

Kraus,

(2017). Disability statistics annual report 2016. Durham, NH: University of New Hampshire. Retrieved from https://disabilitycompendium.org/sites/default/files/user uploads/2016 AnnualReport.pdf..

28.

Lent,

R. W.

(2001) Vocational psychology and career counseling: Inventing the future. Journal of Vocational Behavior, 59, 213–225.

29.

Lent,

R. W.

(2005). A social cognitive view of career development and counseling. In Brown

S. D. .

& Lent

R. W. .

(Eds.), Career development and counseling: Putting theory and research to work, (pp. 101–127). New York: Wiley.

30.

Lent,

R. W.

, Brown,

S. D.

, & Hackett,

(1994) Toward a unifying social cognitive theory of career and academic interest, choice, and performance [Monograph]. Journal of Vocational Behavior, 45, 79–122.

31.

Lent,

R. W.

, Brown,

S. D.

, & Hackett,

(2000). Contextual supports and barriers to career choice: A social cognitive analysis. Journal of Counseling Psychology, 47, 36–49.

32.

Lent,

R. W.

, Brown,

S. D.

, & Larkin,

K. C.

(1986). Self-efficacy in the prediction of academic performance and perceived career options. Journal of Counseling Psychology, 47, 265–269.

33.

Lent,

R. W.

, Brown,

S. D.

, Schmidt,

, Brenner,

, Lyons,

, & Treistman,

(2003). Relation of contextual supports and barriers to choice behavior in engineering majors: Test of alternative social cognitive models. Journal of Counseling Psychology, 50, 458–465.

34.

Lent,

R. W.

, Brown,

S. D.

, Sheu,

, Schmidt,

, Brenner,

B. R.

, Gloster,

C. S.

, & Treistman,

(2005) Social cognitive predictors of academic interests and goals in engineering: Utility for women and students at historically Black universities. Journal of Counseling Psychology, 52, 84–92.

35.

Lent,

R. W.

, Sheu,

H. B.

, Singley,

, Schmidt,

J. A.

, Schmidt,

L. C.

, & Gloster,

C. S.

(2008) Longitudinal relations of self-efficacy to outcome expectations, interests, and major choice goals in engineering students. Journal of Vocational Behavior, 73, 328–335. 10.1016/j.jvb.2008.07.005.

36.

Marton,

, & Säljö,

(1976) On qualitative differences in learning I: Outcome and process. British Journal of Educational Psychology, 46, 4–11.

37.

Marton,

Säljö,

(1997). Approaches to learning. In Marton,

Hounsell,

& Entwistle

(Eds.), The experience of learning, Edinburgh: Scottish Academic Press.

38.

Muller,

, Peebles,

M. C.

, Chiu,

C. Y.

, Iwanaga,

, Tang,

, Brooks,

, Eagle,

, & Chan,

(2017) Association of employment and health and well-being in fibromyalgia. Journal of Rehabilitation, 83(2), 37–43.

39.

Mullikin,

E. A.

, Bakken,

L. L.

, & Betz,

N. E.

(2007) Assessing research self-efficacy in physician-scientists: The Clinical Research Appraisal Inventory. Journal of Career Assessment, 15, 367–387.

40.

O’Neill,

, Kang,

H. -J.

, Sánchez,

, Muller,

, Aldrich,

, Pfaller,

, & Chan,

(2015) Effect of college or university training on earnings of people with disabilities: A case control study. Journal of Vocational Rehabilitation, 43, 93–102.

41.

Petrilli,

M. J.

(2016) Not just college: Technical education as a pathway to the middle class. Washington, DC: Brookings Institution. Retrieved from https://www.brookings.edu/blog/social-mobility-memos/2016/04/01/not-just-college-technical-education-as-a-pathway-to-the-middle-class/.

42.

Ryscavage,

P. M.

(1982) Employment problems and poverty: Examining the linkages. Monthly Labor Review, 55–59. Retrieved from https://www.bls.gov/opub/mlr/1982/06/rpt1full.pdf.

43.

Stanger-Hall,

K. F.

(2012) Multiple-choice exams: An obstacle for higher-level thinking in introductory science classes. Cell Biology Education—Life Sciences Education, 11(3), 294–306.

44.

Turner,

S. L.

, & Lapan,

R. T.

(2005) Evaluation of an intervention to increase adolescents’ efficacy and interests in non-traditional careers. Journal of Vocational Behavior, 66(3), 516–531.

45.

U.S. Bureau of Labor Statistics (2017a). Persons with a disability: Labor force characteristics summary. Retrieved from https://www.bls.gov/news.release/disabl.nr0.htm.

46.

U.S. Bureau of Labor Statistics (2017b). Unemployment rate 2.5 percent for college grads, 7.7 percent for high school dropouts. Retrieved from https://www.bls.gov/opub/ted/2017/unemployment-rate-2-point-5-percent-for-college-grads-7-point-7-percent-for-high-school-dropouts-january-2017.htm.

47.

Weich,

, & Lewis,

(1998) Poverty, unemployment, and common mental disorders: Population based cohort study. British Medical Journal, 66(3), 115–119.

48.

Wolanin,

T. R.

, & Steele,

P. E.

(2004). Higher education opportunities for students with disabilities: A primer for policymakers. Washington, D. C.: The Institute for Higher Education Policy. Retrieved from https://www.ahead.org/uploads/docs/resources/ada/Opportunities%20for%20Students%20with%20Disabilities.pdf.

49.

Woolf,

S. H.

, Aron,

, Dubay,

, Simon,

S. M.

, Zimmerman,

, & Luk,

K. X.

(2015). How are income and wealth link to health and longevity?Richmond, VA: Virginia Commonwealth University, Center on Society and Health. Retrieved from https://www.urban.org/sites/default/files/publication/49116/2000178-How-are-Income-and-Wealth-Linked-to-Health-and-Longevity.pdf.