Thecollegetocareertransition.pdf

Journal of Vocational Behavior 123 (2020) 103506

Available online 11 November 2020
0001-8791/© 2020 Elsevier Inc. All rights reserved.

The college-to-career transition in STEM: An eleven-year
longitudinal study of perceived and objective vocational
interest fit☆

Michelle H. Kim 1, *, Margaret E. Beier
Rice University, Department of Psychological Sciences, 6100 Main St, Houston, TX 77005, United States

A R T I C L E I N F O

Keywords:
Vocational interest fit
Career trajectory
Major retention
Career choice
Career attitudes

A B S T R A C T

This 11-year longitudinal study investigates the effects of vocational interest fit measured in
college on college-to-career trajectories. Vocational interest theories suggest that vocational in-
terest fit will affect choices and attitudes about major and job. The expectancy value model is
used in the study to describe how values assigned to an outcome (e.g., engaging in a major or a
job) ultimately influence major and career choices. The current study tracks a cohort of students
(N = 159) from 2007 to 2018 examining their vocational interest, matriculation major interest,
degree, first job after graduation, and job seven years post-graduation. Results showed that
vocational interest fit with major had a significant effect on major retention and first job choice,
and the relationships were mediated by subjective task values (attainment value, intrinsic value,
utility value, and relative cost). Vocational interest fit with job had a significant effect on career
attitudes, and these relationships were mediated by perceived career fit. Additionally, job fit
increased over time from first job after graduation to 2018 job. Results suggest that vocational
interest measured in college is useful in predicting future career trajectories.

1. Introduction

Career choice is one of the most critical decisions people make in their lives, largely because ending up in a job with a bad fit can be
a miserable experience. College students strive to find a major that is a good match with their interests, hoping that their major choice
will lead to a job with a good fit. Previous studies suggest that genuine interest in a major as well as interest in the work relevant to that
major are the main factors that influence students’ decisions to choose and change their majors in post-secondary education (Adams
et al., 1994; Beggs et al., 2008; Collins & Giordani, 2003; Stanislaw, 2014). Furthermore, theories of career choice support the idea that
vocational interest is an important determinant of career choice (Lent et al., 1994; Lent & Lopez, 1996).

☆ This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
* Corresponding author.

E-mail addresses: [email protected] (M.H. Kim), [email protected] (M.E. Beier).
1 This research was part of the lead author’s master’s thesis conducted at Rice University. Parts of this study were presented at the 34th Annual

Conference of the Society for Industrial and Organizational in Washington, D.C. in April 2019. The authors would like to thank the Rice
University Social Science Research Institute for funding this research. We would also like to thank Dr. Ashley Hanks, the research assistants in the
Adult Skills and Knowledge Lab, and Dissertation Committee members, Dr. Fred Oswald and Dr. Mikki Hebl, at Rice University for their assistance.
Michelle H. Kim is now at the Army Research Institute in Washington, D.C.

Contents lists available at ScienceDirect

Journal of Vocational Behavior

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

https://doi.org/10.1016/j.jvb.2020.103506
Received 1 August 2019; Received in revised form 13 March 2020; Accepted 19 October 2020

mailto:[email protected]

mailto:[email protected]

www.sciencedirect.com/science/journal/00018791

https://www.elsevier.com/locate/jvb

https://doi.org/10.1016/j.jvb.2020.103506

https://doi.org/10.1016/j.jvb.2020.103506

http://crossmark.crossref.org/dialog/?doi=10.1016/j.jvb.2020.103506&domain=pdf

https://doi.org/10.1016/j.jvb.2020.103506

Journal of Vocational Behavior 123 (2020) 103506

2

Holland’s theory of vocational interest suggests that people strive to find work environments that fulfill their values and interests,
and that vocational interest, or vocational personality, consists of work related interest, values, and abilities (Holland, 1997). Hence,
the fit, or the congruence, between one’s vocational interest and the environment (e.g., work or college) should predict attitudes about
and behaviors in those environments (Holland, 1997). Many studies have shown the effect of vocational interest on major decisions
(Allen & Robbins, 2008; Le et al., 2014; Wessel et al., 2008). Studies also have found that fit between one’s vocational interest and
corresponding work characteristics influences job-related outcomes such as satisfaction (Dik et al., 2010; Tsabari et al., 2005), per-
formance, and turnover (Van Iddekinge, Putka, & Campbell, 2011; Van Iddekinge, Roth, et al., 2011). Most research on vocational
interest fit is cross sectional, meaning that interest scales and job outcome measures are collected at the same time from a group of
students or employees (Dik et al., 2010; Van Iddekinge, Putka, & Campbell, 2011).

Although informative, cross-sectional research cannot provide insight into the stability of vocational interests over time or whether
interest measures taken in school predict longer-term career-related outcomes. Some longitudinal studies have investigated within-
person change in interest profiles over time (Low et al., 2005; Tracey & Robbins, 2005), the effect of vocational interest fit on aca-
demic outcomes in academic settings (Allen & Robbins, 2008; Nye et al., 2012; Tracey et al., 2012), and the effect of interest fit on job
outcomes in job settings (Feij et al., 1999). However, we know of no studies that have examined the effect of interest fit on both
academic and job outcomes spanning academic and career trajectories.

In addition to vocational interest fit theory, we used Eccles and Wigfield’s (2002) expectancy value model to frame part of our
investigation. The expectancy value model describes how one’s motivation will be influenced by the value one assigns to an outcome
(i.e., obtaining a degree or career in a particular field) and expectations of achieving that outcome. The expectancy value model is
based on social-cognitive theory and describes how the value that students assign to engaging in their major or major-related activities
will ultimately influence retention in the major (Bandura, 1991). We assessed four subjective task values identified by Eccles and
Wigfield (2002) (attainment value, intrinsic value, utility value, and relative cost) to explain how fit impacts major retention and
career choice.

The current study contributes to the existing literature by investigating the long-term effect of interest measured during college on
college-to-career trajectories. Specifically, we examine the effect of vocational interest fit on retention in a major, retention in a career
related to the major, and career attitudes over the college-to-career trajectory. The sample comprises students who matriculated into
college with an interest in STEM in 2007 and who participated in a larger study aimed at examining student attrition in STEM
(Rittmayer & Beier, 2008). We followed up with participants in 2018 and assessed career choices and career attitudes. We investigated
the effect of vocational interest captured in 2007 on major choice and career trajectories. The study examined two different models of
the relationship between 1) interest fit with intended college major at matriculation into college on major retention and career choice
and 2) interest fit with current job (henceforth referred to as 2018 job) and career attitudes. The study also explored changes in job fit
over a seven-year period after college (i.e., from first job after college, henceforth referred to as first job, to 2018 job).

1.1. Vocational interest fit

Fundamentally, P-E fit theory predicts that when there is a match between the characteristics of the person and the resources
provided by the environment, positive outcomes such as psychological well-being will result. Equally important, when there is a
mismatch between the characteristics of the person and the characteristics of the environment, mental and physical strain will result
(Edwards et al., 1998). According to this theory, fit is important because people feel more comfortable and competent when they
experience fit (Chatman, 1989) and P-E fit fulfills people’s fundamental need for belonging, identity, and self-actualization (Cable &
Edwards, 2004). For instance, person-organization fit assesses a person’s perception of fit with organization, and person-job fit assesses
a person’s perception of fit with job (Lauver & Kristof-Brown, 2001). Person-organization fit is positively related to job satisfaction,
and negatively related to turnover intentions. Person-job fit is positively related to contextual performance (extra-role behaviors) and
job satisfaction, and is also negatively related to turnover intentions.

Vocational interest fit is a type of a Person-Environment (P-E) fit that refers to the congruence between the vocational interests of a
person and the attributes of a work or college environment. People desire work environments that allow them to “exercise their skills
and abilities, express their attitudes and values, and take on agreeable problems and roles” (Holland, 1997, p. 4). Accordingly,
vocational interest theory suggests that individuals are attracted to certain environmental properties that match their vocational
interest type, and that differences in the degree of fit can result in different attitudinal and decisional outcomes in the job context.
Hence, fit, which represents the degree of congruence between vocational interest type and environment type, can predict important
job-related outcomes.

Holland’s (1997) theory of vocational personalities and work environments suggests that most people resemble a combination of
six vocational personality types, or interest types, that comprise the acronym RIASEC: Realistic, Investigative, Artistic, Social,
Enterprising, and Conventional. Realistic interest involves practical and hands-on activities; investigative interest involves working
with ideas and critical thinking; artistic interest involves working with forms, designs, and patterns; social interest involves working
with and providing services to other people; enterprising interest involves starting up and carrying out projects as leaders; and con-
ventional interest involves following a set of rules and working with data (Holland, 1997).

1.2. Fit with major and major retention and career choice

Holland’s (1997) theory of vocational interest suggests that people desire work environments that are compatible with their in-
terests, values, and abilities. Further, misfit between a person’s work-related interests and work environments can motivate people to

M.H. Kim and M.E. Beier

Journal of Vocational Behavior 123 (2020) 103506

3

change their environment, which can explain why people choose (i.e., choice behavior) and stay (i.e., persistence behavior) in an
environment (Le et al., 2014). Indeed, research shows that vocational interest fit measured using RIASEC dimensions is predictive of
academic outcomes (Allen & Robbins, 2008; Le et al., 2014). Allen and Robbins (2008) examined 48,232 students across 25 different
four-year colleges and universities. The authors examined the relationship between fit with major and retention in the major and found
that interest fit with college major and first year GPA predicted retention in the major at the end of third year (estimated regression
coefficients were 0.383 for fit with college major and 0.360 for first year GPA). Le et al. (2014) examined 207,093 students at 51
postsecondary institutions on the relationship between vocational interest fit and persistence in STEM and found that people with
higher fit with STEM major were less likely to leave STEM at the end of third year (β = 0.126). In summary, these studies showed that
interest fit with major predicts important retention outcomes in the major.

H1. Objective major fit is positively correlated with a) obtaining a STEM degree at graduation and b) the likelihood of having a STEM
first job, and c) the likelihood of having a STEM 2018 job.

Although previous research suggests that there is a relationship between objective vocational interest fit and college major
retention (Allen & Robbins, 2008; Tracey et al., 2012; Wessel et al., 2008), much of this research examines major choice during the
second or third year of college (Allen & Robbins, 2008; Tracey et al., 2012) and commitment to one’s major (Wessel et al., 2008) as
outcome variables, rather than retention in the major through graduation. There are studies that examine the congruence between the
vocational interest profile of the person measured in college and interest profile of future career aspirations (Borgen, 1972) or career
choice (Bartling & Hood, 1981). However, we know of no research that has examined how vocational interest fit with college major
predict major retention or career choice after college. The current study fills these research gaps by examining the relationship between
objective vocational interest measured at college matriculation to predict both major retention at graduation and career trajectories
after graduation.

Assessment of the value of STEM activities for participants while enrolled in college was based on Eccles and Wigfield’s (2002)
expectancy value model. We assessed four facets of subjective task value as identified by Eccles and Wigfield (2002): attainment value,
intrinsic value, utility value, and relative cost. Attainment value assesses the extent to which the activities align with one’s self concept.
Intrinsic value assesses the extent to which personal pleasure is derived from engaging in the activities. Utility value assesses the extent
to which the activity is perceived to be useful for attaining a specific outcome (e.g., career goal). And relative cost assesses the extent to
which the activities detract from other areas of life or development.

We posited that the four subjective task value facets would mediate the relationship between objective fit and major retention and
STEM job choice (Fig. 1). The expectancy value model suggests that the value that students assign to academic activities can influence
motivation to engage in a major and persist in the major or in a job relevant to the major. For instance, if a student’s intrinsic value for
STEM activities is high, he or she will derive personal satisfaction from engaging in STEM activities and he or she will be more likely to
persist in the major and to choose a job related to STEM. There has also been support for the idea that subjective task value mediates the
relationship between educational interventions (active learning) and career aspirations in STEM (Beier et al., 2019).

The current study examined the mediating role of subjective task value facets on the relationship between objective fit with major
and STEM major retention and first-job and 2018 job choice. We expected that the relationship between objective major fit and STEM
major retention and job choice would be explained by subjective task value variables (i.e., attainment, intrinsic, utility, and cost;
Fig. 1).

H2. Subjective task value facets will mediate the relationship between objective major fit and STEM major retention (measured by
obtaining a STEM degree) and STEM first job and STEM 2018 job. Specifically, (a) mediation for attainment, utility, and intrinsic
values will be positive (i.e., these mediators will be positively related to objective major fit and STEM major retention and job choices),
and (b) mediation for relative cost will be negative (i.e., relative cost will have a negative relationship with objective major fit, STEM
major retention, and job choice).

Person

profile

College

major

profile

Subjective task values

– Attainment value

– Intrinsic value

– Utility value

– Relative cost

– STEM retention

– Career choice (first

job, 2018 job)

Objective

fit

Fig. 1. Objective Major Fit – Subjective Task Value – Major Retention Model. Subjective task values mediate the relationship between objective fit
and retention outcomes.

M.H. Kim and M.E. Beier

Journal of Vocational Behavior 123 (2020) 103506

4

1.3. Fit and career attitudes

Many studies have examined the relationship between objective interest fit and job attitudes, particularly, turnover intentions and
job satisfaction. Indeed, researchers have found that vocational interest fit is related to turnover intentions (Van Iddekinge, Putka, &
Campbell, 2011; Van Iddekinge, Roth, et al., 2011). In a study of soldiers in the U.S. Army, Van Iddekinge, Putka, and Campbell (2011)
found that vocational interests showed significant incremental validity for predicting intentions to continue in the Army (R2 = 0.08)
above cognitive ability. Van Iddekinge, Roth, et al. (2011) conducted a meta-analysis comprised of 74 studies with 141 independent
samples and found estimated validities for single interest scales were − 0.19 for turnover intention and − 0.15 for actual turnover. In
summary, these studies support vocational interest theory, which posits that interest fit with one’s job will have a significant impact on
job attitudes and behaviors.

The magnitude of the relationship between job satisfaction and interest fit is often moderate to small. Assouline and Meir (1987)
conducted a meta-analysis and found a significant mean correlation of 0.21 between fit and satisfaction using 53 correlations. In
addition, Tranberg et al. (1993) and Tsabari et al. (2005) found a mean correlation of 0.17 (not significantly different from zero) using
22 and 53 effect sizes respectively. Gottfredson and Holland (1990) showed a significant correlation between objective interest fit and
job satisfaction (r = 0.36) using a sample of bank tellers. The current study examines the relationship between interest fit and work
satisfaction. Unlike job satisfaction, which is influenced by external factors (i.e., salary, organizational climate) unrelated to interest
(Spector, 1997), work satisfaction is an intrinsic form of satisfaction that measures one’s satisfaction with the work itself and is thus
more directly related to interest (Prediger, 2000).

Career commitment concerns the extent to which people identify with and value their profession, and their willingness to exert time
and effort to pursue their career goals. Career commitment has been shown to be positively related to job involvement (i.e., psy-
chological involvement with one’s work) and negatively related to career withdrawal (i.e., intention to leave one’s career; Blau, 1985;
Goulet & Singh, 2002). Although we know of no research that directly examines the relationship between vocational interest fit and
career commitment, Holland’s (1997) theory would predict that having a good interest fit with one’s job would positively influence a
person’s identification with work and job commitment.

The current study examines work satisfaction, career commitment, and career withdrawal. Notably, career attitude assessments
used in the current study are relative to a profession – not an organization. This is important because a person might be committed and
satisfied with his or her profession and the work relevant to the profession, but be disenchanted with, and intending to leave, his or her
organization. The current study is focused on attitudes about professions, not organizations.

H3. Objective 2018 job fit will be positively correlated with work satisfaction, and career commitment, and negatively correlated
with career withdrawal.

Perceived fit measures individuals’ perceptions of their compatibility with their environments (Kristof-Brown et al., 2005). Because
perceived fit directly measures people’s perception of fit, it is more proximal to people’s attitudes than objective fit. We hypothesized
that perceived fit would mediate the relationship between objective fit and career attitudes. That is, perceived fit would explain the
relationship between objective fit and career attitudes (Fig. 2).

H4. Perceived career fit will mediate the relationship between objective 2018 job fit and career attitudes.

1.4. Fit over time

P-E fit theory suggests that stress and strain will result from the discrepancy between environmental supplies and personal char-
acteristics (Cummings & Cooper, 1979). One interesting argument raised by some researchers is that when people face stressors, they
increase their effort at first, and thus their overall performance will not be damaged over the short-term (Robert & Hockey, 1997;
Tafalla & Evans, 1997). The negative impact of misfit appears when people are unable to sustain effort to compensate for misfit over

Person

profile

Job

profile

Perceived career fit

Career Attitudes

– Work satisfaction

– Career commitment

– Career withdrawal

Objective

fit

Fig. 2. Objective Job Fit – Perceived Job Fit – Career Attitudes Model. Subjective fit mediates the relationship between objective fit and job attitudes.

M.H. Kim and M.E. Beier

Journal of Vocational Behavior 123 (2020) 103506

5

time. In this scenario, people will likely quit jobs with bad fit and find jobs with better fit.
Research on personality and aging also supports the idea that P-E fit increases over time. Research shows that people become more

psychologically mature with age (i.e., more self-aware of their interests, values, and identity), and thus work to create environments
that fit for them (Caspi et al., 2005; Kooij et al., 2017). Kooij et al. (2017) suggest that people learn more about their weaknesses and
strengths with age (Bosma & Kunnen, 2001) and select experiences that help them to develop strong professional identities over time.
Caspi et al. (2005) proposed the maturity principle, which suggests that people become more dominant, agreeable, conscientious, and
emotionally stable with age (Roberts et al., 2006), and these changes lead to self-actualization and personal growth, as well as being
more effective in love, work, and health. These studies suggest that as people become more mature and knowledgeable about
themselves with age, they will find jobs that will be a better fit for them.

Additionally, researchers have suggested that aging is associated with achieving greater psychological well-being through adopting
intrinsic values over extrinsic ones, and greater self determination to achieve intrinsic goals (Sheldon & Kasser, 2001). This implies that
people will strive to find jobs that match better with their interests and intrinsic values in contrast to jobs that match with extrinsic
values such as salary expectations. The current study will examine changes in fit over time from first job after college to 2018 job.

H5a. Job fit will increase from objective first job fit (first job obtained after graduation) to objective 2018 job fit.

Further, based on the idea that people will select experiences that help them to develop professional identities (Kooij et al., 2017),
we predict that people who have a greater number of career changes will have made the changes to achieve a better fit with their job. In
other words, people with more career changes will have a greater increasing fit trend compared to people with a lower number of job
and career changes.

H5b. Number of career changes will moderate the relationship between job fit and time. The increase in fit over time will be greater
for people who have greater number of career changes.

1.5. Measuring vocational fit

P-E fit can be categorized into objective and perceived fit. Objective fit measures examine fit by independently measuring the
characteristics of the person and the environment, and modeling the difference as an indicator of fit. For example, a self-report measure
of vocational personality might be matched with the vocational interest profile of a job. The Occupational Information Network
(O*NET; Peterson et al., 1999) is an occupational information database that provides comprehensive job information for more than
900 jobs. Job information includes general job descriptions, an index of job requirements, job complexity ratings, job preparation
ratings, and the job’s correspondence with each of the six RIASEC codes for each job in the O*Net.

In contrast to objective fit, measures of perceived fit tend to be self-report assessments aimed at understanding a person’s
perception of fit with their environment. For example, a person might be asked to indicate their perception of how well they fit with
their job. As such, one distinction between objective and perceived assessments of fit is that assessments of perceived fit consider both
the person and environmental characteristics simultaneously and include any idiosyncratic weighting of person and environmental
characteristics that might be important to any one individual. Objective assessments of fit consider person and environment separately
and thus are not subject to the idiosyncratic judgments or biases that might influence perceptions of fit (Kristof-Brown et al., 2005).
Research supports that objective and perceived fit are distinct constructs in that correlations between objective and perceived fit tend
to be significant but small in magnitude (e.g., around r = 0.22; Dineen et al., 2002, and r = 0.33; Cable & Judge, 1996).

Because participants in the current study matriculated with an interest in STEM, we operationalized