Journal of Vocational Behavior 83 (2013) 346–355

Contents lists available at ScienceDirect

Journal of Vocational Behavior

journal homepage: www.elsevier.com/locate/jvb

Gender differences in applying social cognitive career theory in

engineering students☆
Mercedes Inda 1, Carmen Rodríguez ⁎, José Vicente Peña 2
University of Oviedo, Spain

a r t i c l e i n f o a b s t r a c t

Article history: This study tested Social Cognitive Career Theory (SCCT) in predicting engineering interest and
Received 26 March 2013 major choice goals among male and female college students. Participants were 579 sophomore
Available online 22 June 2013 engineering students who completed measures of self-efficacy beliefs, outcome expectations,
interests, goals and social supports and barriers. Findings confirmed that the SCCT model
Keywords: produced a good fit to the data across gender. The findings also corroborated that women have
Social cognitive career theory less self-efficacy beliefs and interest than men. However, there are no statistically significant
Self-efficacy beliefs differences in measures relating to outcome expectations and goals. Moreover, women are
Outcome expectations
more likely to perceive support, especially from peers and family, while men are more likely to
Interest
perceive family barriers than women. For other supports and barriers there are no gender
Goals
Gender differences. Implications for future research and practice are discussed.
© 2013 Elsevier Inc. All rights reserved.

1. Introduction

Social Cognitive Career Theory (SCCT) is one of the theories used for explaining the formation of academic and vocational
interest, the selection and pursuit of career-relevant choices, and people's performance and persistence in educational and
occupational fields. The theory was developed by Lent and colleagues, and focuses on several cognitive-person variables and how
these interact with other aspects of the person and his or her environment to help shape the course of career development (Lent &
Brown, 2006; Lent, Brown, & Hackett, 1994, 2000). The theory proposes four cognitive-person variables: self-efficacy beliefs,
outcome expectations, interest and goals. Person variables such as gender, race/ethnicity or age and contextual factors (barriers
and support) are outside of the core model. Such variables can affect self-efficacy, interest, goals, and outcome expectations.
Contextual barriers and support are conceived within SCCT as environmental variables that can facilitate the formation and
pursuit of individuals' career choices (Lent & Brown, 2006; Lent et al., 2000).
Previous results in cross-sectional studies with engineering students have indicated that SCCT variables (self-efficacy beliefs,
outcome expectations) are strongly predictive of academic interest and goals related to engineering and, as also hypothesized by
SCCT, interest produced significant direct paths to goals (Lent, Brown, Sheu, Schmidt, Brenner, Gloster, et al., 2005). These findings
indicated that participants who perceived themselves to be efficacious in engineering and anticipated positive rewards from it
also expressed the engineering-related interest and reported the goals required to complete an engineering degree. Similar
conclusions were obtained in Lent, Brown, Schmidt, Brenner, Lyons and Treistman (2003) with participants on an introductory
engineering course, insofar as the study found that self-efficacy was predictive of outcome expectations, interest and goals;

☆ This article is based on work supported by the Ministry of Economy and Competitiveness (Spain) (EDU-2010-17233). We gratefully acknowledge the assistance of
R. W. Lent for sending the scales used in this study.
⁎ Corresponding author at: Facultad de Ciencias de la Educación, Despacho 322, C/Aniceto Sela, s.n. 33005 Oviedo, Spain. Fax: +34 985103226.
E-mail addresses: indamaria@uniovi.es (M. Inda), carmenrm@uniovi.es (C. Rodríguez), vipe@uniovi.es (J.V. Peña).
1
Address: Facultad de Ciencias de la Educación, Despacho 315 C/Aniceto Sela, s.n. CP:33005 Oviedo, Spain. Fax: +34 985103226.
2
Address: Facultad de Ciencias de la Educación, C/Aniceto Sela, s.n. CP:33005 Oviedo, Spain. Fax: +34 985103226.

0001-8791/$ – see front matter © 2013 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.jvb.2013.06.010
 M. Inda et al. / Journal of Vocational Behavior 83 (2013) 346–355 347

however, outcome expectations did not explain additional significant variation in interest beyond self-efficacy. Also, self-efficacy
produced a significant indirect path to goals through interest.
Lent, Lopez, Lopez, and Sheu (2008; see also Lent, Lopez, Sheu, & Lopez, 2011), have cited the utility of the SCCT choice model
in accounting for the interests and persistence goals of a sample of students majoring in the computing disciplines. They
demonstrated that self-efficacy was strongly predictive of outcome expectations, interest and goals, but outcome expectations did
not yield significant paths either to interest or goals.
Lent, Singley, Sheu, Schmidt, and Schmidt (2007) concluded that social-cognitive factors correlated positively with academic
satisfaction in engineering students. The results of the structural equation modeling analyses indicated that academic goal
progress, self-efficacy and contextual support were individually and collectively predictive of engineering students' academic
satisfaction. Also, self-efficacy and contextual supports combined to strongly predict perceived goal progress. Only outcome
expectations failed to explain significant variation in either goal progress or academic satisfaction.
With a sample of students in engineering and biological science majors, Byars-Winston, Estrada, Howard, Davis, and
Zalapa (2010) found that maths/science-related academic self-efficacy and outcome expectations were associated with
academic goals and interest. However, for engineering students, the contribution of academic self-efficacy to goals is only
indirect, mediated through outcome expectations and interest. In this group there was also a significant path from interest to
goals.
Other cross-sectional studies with college students reached similar conclusions. Thus, Lent, Brown, Brenner, Chopra, Davis,
Talleyrand, et al. (2001) found that maths/science-related course interest related to both maths/science-related course
self-efficacy and outcome expectations, producing bivariate correlations. Interest, self-efficacy and outcome expectations were
each found to correlate significantly with goals. Moreover, path analysis showed that the effect of outcome expectations on goals
was both direct and indirect, via interest; the effect of self-efficacy on goals was only indirect, via interest and outcome
expectations. Gainor and Lent (1998) extended SCCT to maths-related interest and academic choice in black college students.
They found that black students' maths self-efficacy and outcome expectations were predictive of their maths-related interest.
Moreover, self-efficacy appeared to affect interest indirectly, through outcome expectations. Only self-efficacy and outcome
expectations produced significant paths to maths goals.
Byars-Winston and Fouad (2008) tested a path model to investigate the relationship between maths- and science-related
self-efficacy beliefs and perceived career barriers to maths/science interest and goals in a sample of undergraduates. They found
direct and indirect relationships of outcome expectations to goals, and a direct relationship of interest to goals. Maths/science
self-efficacy directly predicted outcome expectations, and there was an indirect relationship of self-efficacy to goals through
interest.
In the study by Waller (2006) with a sample of non-traditional African-American college students, maths self-efficacy and
outcome expectation variables correlated positively, and provided significant and strong paths to maths interest. Also, maths
self-efficacy yielded a significant indirect path to maths interest through outcome expectations. In Waller's study, the direct effect
of maths-self efficacy on goals was also significant and strong; however, outcome expectations did not yield a significant direct
effect on goals. Maths self-efficacy and outcome expectations produced a significant indirect effect on goals through maths
interest. The direct effect of maths interest on goals was also significant.
A study carried out in Spain (Blanco, 2011) explored the usefulness of SCCT for predicting interests and goals in relation to
statistics among psychology students. The study found that self-efficacy and outcome expectations directly predicted interest, and
self-efficacy also appeared to influence interests through outcome expectations. Moreover, self-efficacy produced a significant
indirect path to goals through interest and outcome expectations. In addition, outcome expectations produced both a significant
direct path to goals and an indirect path to goals via interest. Interest also directly affected goals.
Relatively few SCCT studies have employed longitudinal designs. Lent, Sheu, et al.'s (2008) path analyses indicated support for
a model in which self-efficacy served as a temporal precursor of outcome expectations, interest and goals, but the paths from
self-efficacy to the other variables were modest in magnitude. Moreover, there was less support for a model in which the latter
variables produced reciprocal paths to self-efficacy. That is, the predominant temporal flow appeared to be from self-efficacy to
the other variables, rather than vice versa. In that study the findings do not support SCCT's hypotheses about the unique role of
outcome expectations in fostering interest and goals, or about interest as a unique antecedent of goals (see also Lent, Sheu,
Gloster, & Wilkins, 2010). Additionally, in other longitudinal study, self-efficacy beliefs served as a temporal precursor of the rest
of the cognitive variables and of the contextual support, but in contrast to the findings of Lent, Lopez, et al. (2008), Lent, Sheu, et
al. (2008), interest and self-efficacy were found to be reciprocally related (Lent et al., 2010).
Sheu, Lent, Brown, Miller, Hennessy and Duffy (2010) used meta-analytic path analysis to synthesize data (from 1981 to 2008)
relevant to SCCT's interest and choice hypotheses, organizing the literature in line with Holland's broad occupational themes. For
the realistic, investigative and enterprising themes, they found that contextual support and barriers produce both direct paths to
choice goals and indirect paths through both self-efficacy and outcome expectations. Moreover, self-efficacy beliefs yielded a
negative path to goals in the enterprising theme. For the artistic, social and conventional themes, they found that self-efficacy,
outcome expectations and interest each produced direct paths to goals. Moreover, self-efficacy was linked to goals indirectly
through outcome expectations and interest, while outcome expectations were linked to goals indirectly through interest. The
meta-analytic path analysis also revealed that outcome expectations and self-efficacy beliefs each contributed usefully to the
prediction of interest, and helped to explain variation in choice goals across Holland's themes. Furthermore, “self-efficacy is
assumed to function both as an antecedent of outcome expectations (and interest) and as a direct contributor to goals” (Sheu et
al., 2010, 262).
348 M. Inda et al. / Journal of Vocational Behavior 83 (2013) 346–355

Regarding contextual variables, in cross-sectional studies support and barriers were significantly related to self-efficacy, and
indirectly to goals through the mediation of self-efficacy (Lent, Lopez, et al., 2008; Lent, Sheu, et al., 2008; Lent et al., 2001; Lent et
al., 2003; Lent et al., 2005; Lent et al., 2007; Lent et al., 2011). In the meta-analytic path analysis (Sheu et al., 2010) and in the
longitudinal study by Lent, Sheu, et al. (2008), they also found that contextual variables produced a direct path to self-efficacy
beliefs. This is consistent with a view of environmental variables as serving primarily as proximal sources of efficacy information.
Lent et al. (2005) also “found partial support for SCCT's direct path hypothesis in that barriers, but not supports, produced a
significant, though small, direct path to goals” (Lent et al., 2005, 91). However, in a longitudinal study by Lent et al. (2010), and
contrary to the findings of several cross-sectional studies, contextual supports and barriers were not exclusively predictive of
self-efficacy (see also Rivera, Chen, Flores, Blumberg, & Ponterotto, 2007). That is, supports and barriers did not explain a
substantial variance in self-efficacy over time. However, contextual support did increase self-efficacy in predicting goals. Other
cross-sectional studies found that contextual variables produced a direct path to outcome expectations (Lent et al., 2007; Sheu
et al., 2010) or interest (Lent et al., 2001).

1.1. Gender and social cognitive career theory

In this context, our study set out to analyze the influence of the gender variable in the SCCT model, with a sample of
engineering students. Few studies have demonstrated the influence of gender on the social cognitive variables of the SCCT model.
Lent et al. (2005) found that women engineering students did not differ significantly from men in most of the social cognitive
variables. However, women did perceive more contextual support and fewer contextual barriers than men. Other studies found a
relation between gender and self-efficacy. Thus, Gainor and Lent (1998), in a study of maths choice intentions in black college
students, found that black men tended to report higher self-efficacy expectations than black women. In addition, Byars-Winston
and Fouad (2008) confirmed that men undergraduates reported higher maths/science self-efficacy and more positive outcome
expectations than did women. Moreover, women perceived more career barriers than men. In Hackett, Betz, Casas, and
Rocha-Singh (1992), women engineering students reported significantly lower positive outcome expectations than men.
In other cases, research has demonstrated the indirect influence of gender on self-efficacy beliefs and/or outcome expectations
through the mediation of previous experience of success in learning (Ferry, Fouad, & Smith, 2000; Hackett, 1985; Lapan,
Shaughnessy, & Boggs, 1996; Lent, Lopez, & Bieschke, 1991, 1993).
Also, the predictive utility of the social cognitive variables is not moderated by student gender. That is, SCCT variables may
help to explain engineering/computing-related interest and major-choice goals in women as well as in men (Lent et al., 2005;
Lent et al., 2011).

1.2. Types of contextual support and barriers

SCCT suggests that people are less willing to move their interests to goals and their goals to actions when they perceive
difficulties in overcoming the barriers that exist in their context. Several theories have described the influence of students'
perceptions of parental support on self-efficacy beliefs and career outcomes. In the study by Byars-Winston and Fouad (2008),
parental involvement directly predicted maths/science self-efficacy and outcome expectations, and indirectly predicted maths/
science interest and goals.
As regards teacher support, Fouad, Hackett, Smith, Kantamneni, Fitzpatrick, Haag, et al.'s (2010, 369) study with high school
and college students showed that teachers were the top sources of maths support for females at all levels. Women's perception
that maths teachers expected them to do well constituted a consistent source of support for females. For secondary students, lack
of role models and teachers' lack of inspiration were the main maths barriers. In science, lack of inspiration and lack of advice
from teachers were seen as barriers for females. Conversely, the perception that teachers wanted them to do well was viewed as a

Table 1
Distribution of sample by gender and degrees.

Women Men Total

(N = 163) (N = 416)

Electrical Engineering 7 (29.2%) 17 (70.8%) 24 (4.1%)

Electronic Engineering 9 (25%) 27 (75%) 36 (6.2%)
Chemical Industrial Engineering 13 (68.4%) 6 (31.6%) 19 (3.3%)
Mechanical Engineering 23 (24.5%) 71 (75.5%) 94 (16.2%)
Industrial Technologies Engineering 22 (25%) 66 (75%) 88 (15.2%)
Telecommunications Engineering 21 (32.3%) 44 (67.4%) 65 (11.2%)
Computer Science (Hardware) 2 (6.5%) 29 (93.5%) 31 (5.4%)
Mining and Geological Engineering 19 (38.8%) 30 (61.2%) 49 (8.5%)
Computer Science (Software) 10 (13%) 67 (87%) 77 (13.3%)
Energy Engineering 15 (38.5%) 24 (61.5%) 39 (6.7%)
Topographic Engineering 9 (45%) 11 (55%) 20 (3.5%)
Forest Engineering 2 (14.3%) 12 (85.7%) 14 (2.4%)
Chemical Engineering 11 (47.8%) 12 (52.2%) 23 (4%)

Note. N = 579. Contingency Coefficient = .27, p = .00.

 M. Inda et al. / Journal of Vocational Behavior 83 (2013) 346–355 349

strong resource for middle school and high school females. The top support cited by middle school and high school males was also
that teachers expected them to do well. Hackett et al. (1992, 536) concluded, in a study with engineering students, that there was
a statistically significant ethnicity ∗ gender interaction influencing the ratings of faculty encouragement, with Euro-American and
Mexican-American men reporting significantly more encouragement than Euro-American or Mexican-American women.
The few studies to date on the influence of peer support and barriers have been carried out in high schools. Fouad, Hackett,
Smith, Kantamneni, Fitzpatrick, Haag, et al.'s (2010) reported that in science, friends were important barriers for both boys and
girls at middle-school level, but were an important source of support in relation to maths for girls at the high-school level.
As far as financial support and barriers are concerned, Perrone, Sedlacek, and Alexander (2001), confirmed that women, in
college studies, reported more financial barriers than men to pursuing their career goals.
The present study was designed to investigate the influence of gender on the SCCT model in engineering students. Specifically,
the research focused on testing six principal hypotheses:

H1. The research confirms the basic relationships between the core variables of the SCCT model for the men's sample and for the
women's sample.

H2. The SCCT model adjusts for the men's sample and for the women's sample.

H3. Women score lower than men on measures of core variables of the SCCT model: self-efficacy beliefs, outcome expectations,
interests and goals.

H4. Women perceive less contextual support and more contextual barriers than men.

H5. The contextual barriers that women perceive are better predictors than the contextual supports of self-efficacy beliefs,
outcome expectations, interests and goals.

H6. The contextual supports that men perceive are better predictors than perceived barriers of self-efficacy beliefs, outcome
expectations, interests and goals.

2. Method

2.1. Participants

Participants were 579 second-year engineering students enrolled at the University of Oviedo (Spain) (163 women and 416
men). Age range was 18 to 37 years, (M = 20, SD = 1.9), though in 55 cases participants did not indicate their age. Age
distribution was non-normal, with measures of skewness and kurtosis higher than 1 (absolute value), and mean age for 80% of the
participants was 20 or less. Analysis of the age by gender yielded statistically significant differences (Mann–Whitney U =
24603.50, p = .01). In women, the mean age was 19 (SD = 1.58), and in men 20 (SD = 2). There were no differences in terms of
race-ethnicity, all the students being white and Spanish.
The gender distribution by engineering degrees (Table 1) shows a statistically significant relationship (χ2 = 46.33, p = .00;
Contingency Coefficient = .27, p = .00). All degrees are male-dominated, except for Industrial Chemistry. Males are most
predominant in Computer Science and Forestry Engineering.

2.2. Procedure and instruments

All participants completed the measures in the autumn of 2011. Data were collected during lectures by members of research
team. Previously, we had told the students that they could leave the room if they did not want to participate in the study, and a
few did so. After a brief presentation in which the researcher described the purpose of the study, the students were asked to fill
out the questionnaire. The questionnaire took about 30 min to complete. All participated voluntarily in the data collection, and
there was no remuneration or course credits for participation.
The questionnaire included socio-demographic questions (age, sex, course year) and various sections with items for
measuring the different SCCT variables. The questionnaire had 77 items, and was a Spanish translation and adaptation of the
Engineering Fields Questionnaire developed by Lent and colleagues. In the translation process, American linguistic expressions
were adapted to Spanish. Professor Robert Lent kindly permitted us to translate the measures that he had designed for testing the
Social Cognitive Career Theory.
To measure self-efficacy beliefs we included a scale (1 = no confidence to 9 = complete confidence) asking the students
about their level of confidence for successfully pursuing engineering studies. For example: “How much confidence do you have in
your ability to complete all of the basic formation (i.e., algebra, calculus, statistics, management) requirements for your
engineering major with grades of 6 or better?”. For the full sample, the scale yielded an α coefficient of .88; in the original version,
Lent, Brown, and Larkin (1986) found an internal consistency reliability of α = .89. Later, Lent et al. (2005) obtained an alpha
coefficient of .91.
The outcome expectations measure asked participants if they agreed that an engineering degree would allow them to obtain
different results in their professional career (1 = strongly disagree to 9 = strongly agree) (e.g.: “Graduating with a BSc degree in
350 M. Inda et al. / Journal of Vocational Behavior 83 (2013) 346–355

engineering will likely allow me to receive a good job offer”). Alpha coefficient in the present sample was .88, while Lent et al.'s
(2003) study obtained adequate internal consistency reliability (α = .91) and that of Lent et al. (2005) obtained an internal
consistency reliability of alpha .89.
The interest measure asked participants to indicate their degree of interest in performing engineering-related activities (1 =
very low interest to 5 = very high interest), (e.g. “How much interest do you have in solving mathematical problems?”). For the
full sample, the scale had a coefficient α of .82. In Lent et al. (2003) this scale produced an alpha coefficient of .83, and in Lent et al.
(2005) the alpha coefficient was .80.
The goal measure required students to indicate their level of agreement with statements about their academic intentions using
a 5-point scale (1 = strongly disagree to 5 = strongly agree) (e.g., Agreement or disagreement with the following statement:
“I plan to remain enrolled on an engineering major for the next semester”). The alpha coefficient value of this factor in the present
study was .88. Lent et al. (2003) reported an alpha coefficient value for this scale of .95, while the alpha coefficient value in Lent et
al. (2005) was .93.
The contextual support and barrier measure asked students to indicate how they perceived the different types of contextual
support and barriers (e.g., “Feel support for this decision from important people in your life, such as teachers”). For the full
sample, the scale had an α coefficient of .82. This scale had a possible score range of 1–5, with higher support and barrier scores
reflecting, respectively, stronger positive and negative expectations relative to pursuing an engineering major. Lent et al. (2005)
reported that this scale yielded an alpha coefficient of .86.
In our study we excluded gender/ethnicity, teacher, peer and family barriers. These items were excluded from the factor
analysis because they showed a correlation of under .15 with the total test score. Previous factor analyses have shown that these
items measured the same thing as other items, making them redundant (e.g., “Feel a lack of support from professors or your
advisor”). The number of items removed was 29. Readers can consult the results of the factor analysis in Rodríguez, Inda, and Peña
(in press).
The eight factors of contextual support/barriers are not included in the model fit, since they were not obtained from factor
analysis, but rather constructed directly from the original instrument. This study constitutes the first phase of a research project
aimed at exploring in depth the “contextual supports and barriers” construct. Four support factors (teachers, parents, peers and
financial) and four barrier factors (teachers, parents, peers and financial) were identified, following Lent and Brown's instrument
(2006). Their alpha coefficients were: teacher support, .66 (e.g., “Feel support for this decision from important people in your life,
such as teachers”); peer support, .70 (e.g., “Get encouragement from your friends for pursuing this major”); parental support, .62;
(e.g., “Feel that your family members support this decision”) financial support, .61 (e.g., “Have enough money saved up to be able
to complete your education in this field”); teacher barriers, .75 (e.g., “Have lecturers or teaching assistants who are difficult to
understand”); peer barriers, .56 (e.g., “Receive negative comments or discouragement about your major from friends”), parental
barriers, .76 (e.g., “Feel pressure from parents or other important people to change your major to some other field”), and financial
barriers, .62 (e.g., “Financial concerns, such as having too little money to afford things like computer software or tutoring, which
could help you to do well in your coursework”). To further explore the latent structure of support and barriers, we performed a
factor analysis using ULS (Unweighted Least Squares) factoring and promin rotation (teacher support, .53; parental support, .79;
teacher barriers, .81 and parental barriers, .57), (TLI. = .95; C.F.I. = .98, RMSR. = .03). However, the other factors – peer
supports, financial support, peer barriers and financial barriers – were outside of the structure. Given this, the authors of the
present study are fully aware that these results need to be revised on the basis of further and more in-depth studies.

2.3. Data analyses

The descriptive statistics related to the mean, standard deviation, skewness and kurtosis were calculated. All SCCT model
factors presented values lower than 1, in absolute value, so that the normality principle was fulfilled.
We tested the fit of the measurement model for women and men using the Pearson correlations and standard deviation, for
women (Table 2) and men (Table 3). The analyses were carried out using the following computer programs: M-Plus (Muthén &
Muthén, 2010) and SPSS 19. We decided to apply ML (Maximum Likelihood) as the estimation method.
We carried out a multivariate analysis of variance (MANOVA) with the criterion variables: self-efficacy beliefs, outcome
expectations, interests, goals and contextual supports/barriers, and factor: gender. The analysis applied to the sample indicated
adequacy of the data for conducting a MANOVA, as skewness and kurtosis values were within the range − 1 to 1, the distribution

Table 2
Correlations and descriptive statistics for women.

Variable 1 2 3 4 M SD

Self-efficacy beliefs 5.37 1.41

Outcome expectations .20⁎ 6.64 1.21
Interests .21⁎⁎ .26⁎⁎ 3.48 .67
Goals .30⁎⁎ .33⁎⁎ .29⁎⁎ 5.95 .89
Contextual supports/barriers .24⁎⁎ .41⁎⁎ .21⁎ .39⁎⁎ 3.87 .62

Note. N = 163.
⁎ p ≤ .01.
⁎⁎ p ≤ .001.
 M. Inda et al. / Journal of Vocational Behavior 83 (2013) 346–355 351

Table 3
Correlations and descriptive statistics for men.

Variable 1 2 3 4 M SD

Self-efficacy beliefs 5.86 1.44

Outcome expectations .27⁎ 6.44 1.32
Interests .27⁎ .24⁎ 3.65 .73
Goals .42⁎ .37⁎ .25⁎ 5.86 .92
Contextual supports/barriers .24⁎ .43⁎ .18⁎ .32⁎ 3.87 .62

Note. N = 416. All correlations are significant.

⁎ p ≤ .001.

was normal, and the Levene test and Box's M test yielded values of over .05; equality of variances between men and women in the
criterion variables is confirmed. The MANOVA allows the type I error to be corrected and the small differences in the criterion
variables to be analyzed.
We also conducted a multiple regression analysis to analyze the influence and predictive capacity of the contextual support
and barriers in relation to the SCCT core model: self-efficacy beliefs, outcome expectations and interests (criterion variables). The
stepwise method was applied for the procedure of entering the variables, in order to avoid redundant predictor variables. The fact
of leaving out some of the regressor variables in this type of model is usually related to the possibility of multicollinearity among
the predictor variables; statistically speaking, many of them can be eliminated because they do not contribute any sufficiently
significant additional explanation to the variance.

3. Results

The results of the Pearson linear correlation coefficients between the variables, means and standard deviation are presented in
Table 2 (women) and Table 3 (men).

3.1. Model fit by gender

To explore the possibility that the structural paths in the theoretical model differ by gender, we carried out multiple-groups
analysis. It was also useful to assess the invariance of the model across gender. This analysis involved testing two models. In the
first model, the values of the structural paths were allowed to vary between men and women. In the second model, the values of
the structural paths were constrained to be equal in the women's and men's samples. Both models produced good indicators of fit
to the data. For the first model, CFI = 1, TLI-NNFI = .96, RMSEA = .05, RMSR = .01; χ2 (813,2) = 4.16, p = .12; and for the
second model CFI = .98, TLI-NNFI = .99, RMSEA = .02, RMSR = .03; χ2 (813,7) = 8.45, p = .30. The difference in fit between
the two models was not significant Δχ2 = 4.29, p = 0.10. Some coefficients are equal in women and men, but in others small
differences can be observed (e.g.: influence of outcome expectations on interest, interest on goals, self-efficacy on goals, and social
supports/barriers on goals).

Interest

F=.21** M= .10*
F=.17** F=.23***
M=.22*** M=.18***
Contextual Goals
Self-Efficacy F=.18***
supports/barriers F=.39*** M=.31***
M=.39***

F= .11*
F=.62*** M=.18*** F=.15**
M=.63*** M=.20**

Outcome Expectations

F=.41*** M=.23***

Fig. 1. Separate parameter estimates for males (M) and females (F). *p ≤ .05; **p ≤ .01; ***p ≤ .001. For females R2 self-efficacy = .07; R2 outcome
expectations = .18; R2 interest = .10, R2 goals = .25. For males R2 self-efficacy = .07; R2 outcome expectations = .22; R2 interest = .10, R2 goals = .27.
352 M. Inda et al. / Journal of Vocational Behavior 83 (2013) 346–355

The parameters are shown in Fig. 1 for men and women. Supports and barriers were exogenous variables, and self-efficacy
beliefs, outcome expectations, interests and goals were endogenous variables. All variables, except for supports and barriers, were
modeled as latent factors, the reason for this being found in Lent's main theory. Lent et al. (1994) considered as core person
variables, self-efficacy, outcome expectations, interests and goals, and as contextual variables, supports and barriers. In the
present study we set out to check Lent's first model (1994, 2006) in a Spanish sample. The core person variables are not observed
directly, but rather deduced from the participant's perception, and since the contextual supports and barriers are obvious in the
participants' environment, there is no need for indicators to evaluate them.
First, we can confirm almost all of H1. It is clear that the most basic relationships between variables from the SCCT model are
supported by men and women. Thus, women's and men's self-efficacy beliefs were predictive of interest in engineering and
technology activities and their attempts to persist in engineering studies (goals). The results show, in men and women, the
influence of outcome expectations on interest and academic goals, in turn, interest also influences goals. On the other hand,
perceived contextual supports and barriers exert direct effects on self-efficacy beliefs, outcome expectations and goals.
Furthermore, it should be noted that H2 is supported because the model fits for both samples (women and men).

3.2. MANOVA results

The results are shown in Table 4, and reveal statistically significant differences in self-efficacy beliefs. Men have more
confidence than women in successfully completing their engineering studies. In the variable interest, men are more interested in
academic and scientific activities related to the engineering fields. In this context, H3 is partially confirmed because women score
lower than men on measures of self-efficacy beliefs and interest. However, women do not differ significantly from men in other
variables: outcome expectations and goals.
Moreover, H4 is not corroborated mainly because there are no differences between men and women in perceived contextual
barriers and support. However, women are more likely to perceive support than barriers (in contradiction to H4), especially from
peers and parents, while men perceive more parental barriers than women. Summarizing, the eta value indicated low effect size
(Table 4).

3.3. Multiple regression results

As shown in Tables 5 and 6, H5 is supported, but H6 is not confirmed. The different types of perceived contextual support had
greater influence than perceived barriers on the core variables of the SCCT model for men and women. The data analysis shows
that teacher support is an important variable that greatly influences the core variables of the SCCT model, especially in the case of
men.
Further analysis shows similarities and differences in the contextual support/barriers perceived by women and men. In this
regard, teacher support is the only variable that helps to increase women's self-efficacy beliefs, while for men, perceived peer
barriers must be added to this variable. Outcome expectations in women are predicted by teacher support and family barriers. For
men, peer barriers and family support, in addition to teacher support, have predictive value for outcome expectations. It is
interesting to note that in the case of men, when they have family support, teacher support reduces the predictive power for
outcome expectations. Interest is determined by peer support for females and by teacher support for males. Regarding academic
goals, in the case of women, these are predicted by peer support and barriers, peer support being the most important predictor.
For men, family barriers are the best predictors of academic goals, followed by teacher and financial support. However, the latter
two variables lose strength if there are peer barriers.

Table 4
Descriptive statistics in a five factor SCCT model by gender.

Women Men

Criterion variables M (SD) M (SD) F(163,416) η2

Self-efficacy 5.37 (1.40) 5.88 (1.44) 14.42⁎ .02

Outcome expectations 6.63 (1.20) 6.43 (1.32) 2.8 .00
Interest 3.48 (.67) 3.65 (.73) 7.34⁎⁎ .01
Goals 5.94 (.89) 5.86(.92) .94 .002
Teacher support 3.53 (.79) 3.40 (.72) 3.37 .006
Peer support 4.06 (.64) 3.85 (.60) 13.7⁎ .023
Family support 4.21 (.74) 3.96 (.65) 13.7⁎ .023
Financial support 3.68 (.80) 3.58 (.76) 2.05 .004
Teacher barriers 2.96 (.77) 2.99 (.80) .06 .00
Peer barriers 2.27 (.60) 2.27 (.65) .001 .00
Family barriers 1.65 (.70) 1.80 (.70) 5.27⁎⁎⁎ .01
Financial BARRIERS 2.30 (.67) 2.40 (.68) 2.96 .005
⁎ p ≤ .001.
⁎⁎ p ≤ .01.
⁎⁎⁎ p ≤ .05.
 M. Inda et al. / Journal of Vocational Behavior 83 (2013) 346–355 353

Table 5
Summary of stepwise regression analysis of core variables of the SCCT model for women.

Criterion Predictors β t R2

Self-efficacy beliefs Teacher support .26 3.39⁎ 6%

Outcome expectations Teacher support .36 5.08⁎ 19%
Family barriers −.21 −2.93⁎⁎
Interest Peer support .30 3.94⁎ 8%
Goals Peer support .31 4.34⁎ 20%
Peer barriers −.27 −3.73⁎
⁎ p ≤ .001.
⁎⁎ p ≤ .01.

4. Discussion

Taken as a whole, the results of this research almost entirely confirm the SCCT model for Spanish engineering students.
According to the SCCT model, self-efficacy beliefs of women and men determine their interest in engineering and technology
activities (Blanco, 2011; Byars-Winston & Fouad, 2008; Byars-Winston et al., 2010; Gainor & Lent, 1998; Lent, Lopez, et al., 2008;
Lent, Sheu, et al., 2008; Lent et al., 2001; Lent et al., 2003; Lent et al., 2005; Lent et al., 2010; Lent et al., 2011; Sheu et al., 2010;
Waller, 2006). Moreover, the outcome expectations of women and men were predicted by self-efficacy (Blanco, 2011;
Byars-Winston & Fouad, 2008; Gainor & Lent, 1998; Lent, Lopez, et al., 2008; Lent, Sheu, et al., 2008; Lent et al., 2001; Lent et al.,
2003; Lent et al., 2010; Lent et al., 2011; Sheu et al., 2010; Waller, 2006). In addition, their attempts to persist in engineering
studies (goals) were directly linked to self-efficacy (Gainor & Lent, 1998; Lent, Lopez, et al., 2008; Lent, Sheu, et al., 2008; Lent et
al., 2003; Lent et al., 2005; Lent et al., 2007; Lent et al., 2010; Lent et al., 2011; Sheu et al., 2010; Waller, 2006).
Also, we found support for the SCCT proposal that the assessments made by both women and men about the relevance of their
studies to their professional future (outcome expectations) influence their interest and goals (Blanco, 2011; Byars-Winston et al.,
2010; Gainor & Lent, 1998; Sheu et al., 2010).
Other SCCT proposals were also supported. Perceived contextual support and barriers influence self-efficacy beliefs (Lent,
Lopez, et al., 2008; Lent, Sheu, et al., 2008; Lent et al., 2001; Lent et al., 2003; Lent et al., 2005; Lent et al., 2007; Lent et al., 2011;
Sheu et al., 2010), outcome expectations (Lent et al., 2007; Sheu et al., 2010) and goals for women and men (Lent, Lopez, et al.,
2008; Lent, Sheu, et al., 2008; Lent et al., 2007; Lent et al., 2011).
Furthermore, the predictive utility of the social cognitive variables is not moderated by student's gender. That is, SCCT
variables can help explain engineering/computing-related interest and major-choice goals in women as well as in men (Lent et
al., 2005; Lent et al., 2011).
Regarding the differences between men and women in their scores on the SCCT model core variables, the results vary
depending on the variable analyzed. In confirmation of previous research findings, women have less confidence in their ability to
successfully complete engineering studies (Byars-Winston & Fouad, 2008; Gainor & Lent, 1998). In addition, women have less
personal interest in academic and scientific activities related to the field of engineering. However, there are no statistically
significant differences in measures relating to outcome expectations and goals — as Lent et al. (2005) also found; on the other
hand, and in contrast to our own findings, Lent et al. (2005) did not find significant differences for self-efficacy beliefs and
interest.
It was also found that women are more likely to perceive support, especially from peers and family, while men are more likely
than women to perceive family barriers. For other types of support and barriers there are no gender differences. In this regard,
other studies suggest the existence of small but significant gender differences, reporting that women perceive more contextual
support and fewer contextual barriers in their academic development (Lent et al., 2005), or support the contrary assertion that
women perceive more barriers in their academic development than men (Byars-Winston & Fouad, 2008).

Table 6
Summary of stepwise regression analysis of core variables of the SCCT model for men.

Criterion Predictors β t R2

Self-efficacy beliefs Teacher support .24 5.07⁎ 8%

Peer barriers −.16 −3.48⁎
Outcome expectations Teacher support .23 4.63⁎ 20%
Peer barriers −.16 −3.68⁎
Family support .25 5.05⁎
Interest Teacher support .23 4.77⁎ 5%
Goals Family barriers −.24 −3.65⁎ 24%
Teacher support .16 3.27⁎
Financial support .16 3.31⁎
Peer barriers −.19 −2.86⁎⁎
⁎ p ≤ .001.
⁎⁎ p b .01.
354 M. Inda et al. / Journal of Vocational Behavior 83 (2013) 346–355

Another finding is that, for men and women, different types of contextual support, and especially in relation to teacher
support, have more influence than contextual barriers. Thus, teacher support is the only variable that helps to increase women's
self-efficacy beliefs, while for men peer barriers must also be considered (see also Fouad, Hackett, Smith, Kantamneni,
Fitzpatrick, Haag, et al.'s, 2010; Hackett et al., 1992). Furthermore, teacher support also influences women's outcome
expectations and, in the case of men, also predicts their interest and their goals. The importance of teacher support can be
explained if we consider the academic context of Spanish universities. Engineering studies are very difficult and require great
intellectual effort on the part of the students. The dropout and failure rates in these studies are the highest in Spanish
universities, and students perceive high levels of requirements and difficulties (NIS, 2013). In this context, we can understand
how teacher support/barriers are more important than parental support/barriers. Having teachers involved, who know the
academic discipline and who encourage and help their students, is a powerful factor that influences students' self-efficacy
beliefs and outcome expectations.
Peer support and barriers also play an important role, though there are fewer studies examining this variable (see Fouad,
Hackett, Smith, Kantamneni, Fitzpatrick, Haag, et al.'s, 2010; Lent et al., 2002). However, in our research they significantly predict
some cognitive-person variables of the SCCT model. It is not easy to explain this influence, but it may be due to the academic
environment created in relation to this type of university course. Students must not only attend lectures, but also go to extra
classes which must be paid for, and this may promote competitiveness among them. In this context, it is easy to understand how
there may be feelings of distrust toward peers, but also good assessments of colleagues who can help.
Finally, we should mention that financial barriers and support have little significance in our sample, unlike the case of other
studies (Lent et al., 2002; McWhirter, 1997; McWhirter, Torres, Salgado, & Valdez, 2007). In general, neither women nor men
perceive that economic barriers can play an important role in their academic and vocational development. In Spain, funding for
education is guaranteed by the State, and the pressure to find resources is lower than in other countries. As costs gravitate more
and more towards the families, the evolution of this variable will become evident in one direction or the other.

4.1. Implications

The present research has practical implications that should be considered in future research and practice with college
students, with a view to gaining a more comprehensive understanding of vocational careers. The results also suggest the utility of
the SCCT model in the context of career development. According to the findings, professionals responsible for career counseling
should focus on increasing self-efficacy beliefs in the area of scientific and technological studies. If we are to increase graduation
rates in science and technology (a fundamental objective of the European Union), interventions, especially in high schools, should
promote the development of self-efficacy beliefs, especially among women. As low technology interest and low self-efficacy
beliefs are risk factors for early attrition in engineering fields, high school counselors should bear this in mind in initiatives to
foster optimal career development in adolescent girls.
The findings also suggest that certain barriers and supports may be fairly generic, whereas others may well differ as a function
of gender. Lent et al. (2002) highlighted the need to conceptualize and assess barriers and support in relation to particular groups.
Thus, practitioners should assess contextual variables (perceived contextual support and barriers) to gain an understanding of
their influence on self-efficacy beliefs, outcome expectations, interests and goals, especially in the case of women. Counselors
could encourage women to discuss barriers that they anticipate in their career paths, explore possible coping strategies relative to
these barriers and identify support systems (family, teacher, and peers) that can help promote their vocational career
(Constantine, Wallace, & Kindaichi, 2005).
Finally, teacher support was clearly meaningful for the students (both women and men). Hence, academic and vocational
counseling departments in universities should help teachers analyze their role in the development of their students' academic
careers.

4.2. Suggestions for future research

Our research, in response to the request made in Lent et al. (2002, 70), identifies specific categories of contextual support
and barriers. In the future, research needs to be carried out to further understand and explain the mechanisms of different
social support and barriers (family, peers, and teachers). One way to understand these experiences in rich detail would be to
conduct qualitative research that allow women and men to express how they believe contextual factors and their gender have
influenced their career decisions. Qualitative methods may be helpful in efforts to gain a more in-depth understanding about the
influence of contextual factors and background variables (gender) on self-efficacy beliefs, outcome expectations, interests and
goals.

References
Blanco, A. (2011). Applying social cognitive career theory to predict interest and choice goals in statistics among Spanish psychology students. Journal of
Vocational Behavior, 78, 49–58.
Byars-Winston, A., Estrada, Y., Howard, C., Davis, D., & Zalapa, J. (2010). Influence of social cognitive and ethnic variables on academic goals of underrepresented
students in science and engineering: A multiple groups analysis. Journal of Counselling Psychology, 57(2), 205–218.
Byars-Winston, A., & Fouad, N. A. (2008). Math and science social cognitive variables in college students. Contributions of contextual factors in predicting goals.
Journal of Career Assessment, 16(4), 425–440.
 M. Inda et al. / Journal of Vocational Behavior 83 (2013) 346–355 355

Constantine, M., Wallace, B., & Kindaichi, M. (2005). Examining contextual factors in the career decision status of African American adolescents. Journal of Career Assessment,
13(3), 307–319.
Ferry, T. R., Fouad, N. A., & Smith, P. L. (2000). The role of family context in a social cognitive model for career related choice behavior: A math and science
perspective. Journal of Vocational Behavior, 57, 348–364.
Fouad, N. A., Hackett, G., Smith, P. L., Kantamneni, N., Fitzpatrick, M., Haag, S., et al. (2010). Barriers and supports for continuing in mathematics and science:
Gender and educational level differences. Journal of Vocational Behavior, 77, 361–373.
Gainor, K. A., & Lent, R. W. (1998). Social cognitive expectations and racial identity attitudes in predicting the math choice intentions of black college students.
Journal of Counselling Psychology, 45(4), 403–413.
Hackett, G. (1985). Role of mathematics self-efficacy in the choice of math-related majors of college women and men: A path analysis. Journal of Counseling
Psychology, 32(1), 47–56.
Hackett, G., Betz, N., Casas, J., & Rocha-Singh, I. (1992). Gender, ethnicity and social cognitive factors predicting the academic achievement of students in
engineering. Journal of Counseling Psychology, 39(4), 527–538.
Lapan, R. T., Shaughnessy, P., & Boggs, K. (1996). Efficacy expectations and vocational interest as mediators between sex and choice of math/science college
majors: A longitudinal study. Journal of Vocational Behavior, 49, 277–291.
Lent, R. W., & Brown, S. (2006). On conceptualizing and assessing social cognitive constructs in careers research: A measurement guide. Journal of Career
Assessment, 14(1), 12–35.
Lent, R. W., Brown, S. D., Brenner, B., Chopra, S. B., Davis, T., Talleyrand, R., et al. (2001). The role of contextual supports and barriers in the choice of math/science
educational options: A test of social cognitive career hypotheses. Journal of Counselling Psychology, 48(4), 474–483.
Lent, R. W., Brown, S. D., & Hackett, G. (1994). Toward a unifying social cognitive of career and academic interest, choice and performance. Journal of Vocational
Behavior, 45, 79–122.
Lent, R. W., Brown, S. D., & Hackett, G. (2000). Contextual supports and barriers to career choice: A social cognitive analysis. Journal of Counseling Psychology, 47(1), 36–49.
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, 33(3), 265–269.
Lent, R. W., Brown, S. D., Schmidt, J., Brenner, B., Lyons, H., & Treistman, D. (2003). Relation of contextual supports and barriers to choice behavior in engineering
majors: Test of alternative social cognitive models. Journal of Counseling Psychology, 50(4), 458–465.
Lent, R. W., Brown, S. D., Sheu, H., Schmidt, J., Brenner, B. R., Gloster, C. S., et al. (2005). Social cognitive predictors of academic interests and goals in engineering:
Utility for women and students at historically black universities. Journal of Counselling Psychology, 52(1), 84–92.
Lent, R. W., Brown, S., Talleyrand, R., McPartland, E., Davis, T., Chopra, S., et al. (2002). Career choice barriers, supports and coping strategies: College students'
experiences. Journal of Vocational Behavior, 60, 61–72.
Lent, R. W., Lopez, F. G., & Bieschke, K. J. (1991). Mathematics self-efficacy: Sources and relation to science-based career choice. Journal of Counseling Psychology, 38(4),
424–430.
Lent, R. W., Lopez, F. G., & Bieschke, K. J. (1993). Predicting mathematics-related choice and success behaviors: Test of an expanded social cognitive model. Journal
of Vocational Behavior, 42, 223–236.
Lent, R. W., Lopez, A. M., Lopez, F. G., & Sheu, H. (2008). Social cognitive career theory and the prediction of interests and choice goals in the computing disciplines.
Journal of Vocational Behavior, 73, 52–62.
Lent, R. W., Lopez, A. M., Sheu, H., & Lopez, A. (2011). Social cognitive predictors of the interest and choices of computing majors: Applicability to
underrepresented students. Journal of Vocational Behavior, 78, 184–192.
Lent, R. W., Sheu, H., Gloster, C. S., & Wilkins, G. (2010). Longitudinal test of social cognitive model of choice in engineering students at historically black
universities. Journal of Vocational Behavior, 76, 387–394.
Lent, R. W., Sheu, H., Singley, D., Schmidt, J., Schmidt, L., & Gloster, C. (2008). Longitudinal relations of self-efficacy to outcome expectations, interest, and major
choice goals in engineering students. Journal of Vocational Behavior, 73, 328–335.
Lent, R. W., Singley, D., Sheu, H., Schmidt, J., & Schmidt, L. (2007). Relation of social-cognitive factors to academic satisfaction in engineering students. Journal of
Career Assessment, 15(1), 87–97.
McWhirter, E. H. (1997). Perceived barriers to education and career: Ethnic and gender differences. Journal of Vocational Behavior, 50, 124–140.
McWhirter, E. H., Torres, D., Salgado, S., & Valdez, M. (2007). Perceived barriers and postsecondary plans in Mexican American and White adolescents. Journal of
Career Assessment, 15(1), 119–138.
Muthén, L., & Muthén, B. (2010). Mplus. Statistical Analysis With Latent Variables. User's Guide. Los Angeles: Muthén & Muthén.
National Institute Of Statistics (NIS) (2013). Statistics of university education in Spain. Unpublished report.
Perrone, K., Sedlacek, W., & Alexander, C. (2001). Gender and ethnic differences in career goal attainment. The Career Development Quarterly, 50, 168–178.
Rivera, L., Chen, E., Flores, L., Blumberg, F., & Ponterotto, J. (2007). The effects of perceived barriers, role models and acculturation on the career self-efficacy and
career consideration of Hispanic women. The Career Development Quarterly, 56, 47–61.
Rodríguez, C., Inda, M., & Peña, J. (2013). Validación de la teoría cognitivo social de desarrollo de la carrera (SCCT) con una muestra de estudiantes de ingeniería.
Revista Educación, XX1 (in press).
Sheu, H., Lent, R. W., Brown, S., Miller, M., Hennessy, K., & 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.
Waller, B. (2006). Math interest and choice intentions of non-traditional African-American college students. Journal of Vocational Behavior, 68, 538–547.