Content uploaded by Hsin-Ya Liao
Author content
All content in this area was uploaded by Hsin-Ya Liao on Oct 12, 2018
Content may be subject to copyright.
Original Article
The Relation Between Interest
Congruence and College Major
Satisfaction: Evidence From
the Basic Interest Measures
Lili Bai
1,2
and Hsin-Ya Liao
3
Abstract
The relation between the degree of interest congruence (i.e., person–environment fit in interest
domain) and career satisfaction has been inconsistent and generally low across studies. Interest con-
gruence is typically measured at the broadband general interest level, bound within Holland’s Realistic,
Investigative, Artistic, Social, Enterprising, and Conventional (RIASEC) framework, and largely based on
the match of the high-point interest codes between persons and environments. Using two cross-cultural
college samples, we reexamined the congruence–satisfaction relation with a refined congruence index
by using narrowband basic interest measures and considering the entire basic interest profiles. As a
comparison, we used three additional congruence indices based on the entire general interest RIASEC
profiles or the high-point RIASEC codes. Findings showed stronger congruence–satisfaction relations
when the basic interest measure and/or complete interest profiles were used to generate interest
congruence indices. Implications for research and career practice are discussed.
Keywords
basic interest, interest congruence, satisfaction, academic majors, college students, career
assessment
The cornerstone of the person–environment fit (P-E fit) career theories (e.g., Holland’s [1973, 1997]
theory of career interests and personalities; theory of work adjustment, Dawis & Lofquist, 1984) is
that an individual’s career satisfaction is positively related to the degree of the fit between his or her
personality attributes (e.g., interests, values) and the characteristics of work or academic environ-
ment. Holland’s theory, for example, suggests that the greater the interest congruence (i.e., P-E fit in
1
Department of Applied Psychology, Fujian Normal University, Fuzhou, China
2
The Chinese University of Hong Kong, Hong Kong, China
3
Department of Educational Leadership, Sports Studies, and Educational/Counseling Psychology, Washington State
University, Pullman, WA, USA
Corresponding Author:
Lili Bai, Department of Applied Psychology, Fujian Normal University, Renwen Building, Qishan Campus, Fuzhou, Fujian
350117, China.
Email: phypsy@qq.com
Journal of Career Assessment
1-17
ªThe Author(s) 2018
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/1069072718793966
journals.sagepub.com/home/jca
interest domain), the better the satisfaction. However, empirical evidence on the relation
between interest congruence and satisfaction has been inconsistent (e.g., Assouline & Meir,
1987; Morris, 2003; Spokane, 1985) and generally low across studies, especially in studies
conducted in academic settings (Assouline & Meir, 1987; Tranberg, Slane, & Ekeberg, 1993;
Tsabari, Tziner, & Meir, 2005).
A possible explanation accounting for the low congruence–satisfaction relation is that interest
congruence is usually measured using Holland’s (1973, 1997) Realistic, Investigative, Artistic,
Social, Enterprising, and Conventional (RIASEC) hexagon taxonomy of interests, which is at the
broadband general interest level. Broadband general interests, typically represented by several broad
and general interest types based on a particular theoretical model, are organized to signify general
areas or themes of interests. Although general interests can provide a parsimonious yet inclusive
classification scheme for understanding the organization of human interests, they lack a certain level
of specificity to represent the full range of individual interests (e.g., Rounds, 1995; Tinsley, 2000).
Therefore, there is an increasing need to jump out of the RIASEC hexagon framework and look for a
refined interest taxonomy with greater specificity (Swanson & Gore, 2000). The narrower band
basic interest taxonomy, on the other hand, has received some attention in recent years for its
superior utility in predicting individuals’ career choices and outcomes as compared to the general
interest taxonomy (e.g., Liao, Armstrong, & Rounds, 2008; Ralston, Borgen, Rottinghaus, & Don-
nay, 2004; Rottinghaus, Hees, & Conrath, 2009). Basic interests are specific and homogeneous units
of interests that group together work activities that share similar properties and represent the same
abstract objects of interest, such as mathematics, finance, or teaching (Campbell, Borgen, Eastes,
Johansson, & Peterson, 1968; Clark, 1961). Until now, the basic interest taxonomy has not been used
to measure interest congruence. As the concept of interest congruence and the positive congruence–
satisfaction relation were originally proposed by Holland based on his RIASEC hexagon taxonomy
model, almost all the prior studies on the interest congruence–satisfaction relations were also based
on the general RIASEC interests. Therefore, several important questions remain: (a) When we jump
out of the RIASEC hexagon framework, how can interest congruence be assessed and how would the
congruence–satisfaction relation be like? (b) Would the congruence–satisfaction relation be
strengthened when the congruence index is based on the narrowband basic interest taxonomy? In
response to these questions, we conducted two studies using two cross-cultural college samples to
reexamine the relation between interest congruence and satisfaction with a refined interest congru-
ence index using the basic interest taxonomy.
P-E Fit Career Theories
There are two popular P-E fit theories in the field of vocational psychology: the theory of work
adjustment (TWA; Dawis & Lofquist, 1984) and Holland’s theory of career interests and person-
alities (Holland, 1973, 1997). According to the TWA (Dawis & Lofquist, 1984), satisfaction is
influenced by the fit between an individual’s needs and the reinforcers provided by work environ-
ments. Dawis (2005) also regarded interests as “a robust predictor of satisfaction” (p. 14), even
though he did not directly specify the role of interests on work adjustment in his TWA postulates. In
Holland’s (1973, 1997) theory, the positive relation between interest congruence and vocational
satisfaction makes up Holland’s major proposition. Holland proposed that vocational interests can
be summarized as six general types, collectively called as RIASEC, and work environments can also
be classified into the corresponding RIASEC types. The relations among the six types can be
illustrated as a hexagon where types that appear closer to one another on the hexagon are more
consistent (Holland, 1973, 1997), and the congruence between individual interests and work envir-
onments is expected to lead to positive career-related outcomes (e.g., satisfaction and achievement).
Based on a large amount of findings regarding interests congruence and career-related outcomes
2Journal of Career Assessment XX(X)
(e.g., Assouline & Meir, 1987; Kulik, Oldham, & Hackman, 1987), Holland (1987) concluded that
interest congruence tends to have a closer relation with satisfaction than with other career-related
outcomes (e.g., achievement).
Relation Between Interest Congruence and Career Satisfaction
Despite strong theoretical propositions, findings on the relation between interest congruence and
career-related satisfaction have been mixed. In Spokane’s (1985) review on interest congruence, for
instance, the aggregate findings suggest that interest congruence is positively related to job satisfac-
tion. In one meta-analysis conducted by Morris (2003), a significant mean correlation (mean of
rs¼.24) between congruence and employed adults’ job satisfaction was also found. Other meta-
analyses, on the contrary, reported nonsignificant mean congruence–satisfaction correlations with
means of rs ranging from .17 to .21, and these correlations obtained in academic settings were even
lower, with means of rs ranging from .03 to .10 (Assouline & Meir, 1987; Tranberg et al., 1993;
Tsabari et al., 2005). Recent empirical research even showed interest congruence as unrelated to
satisfaction with academic majors (Pozzebon, Ashton, & Visser, 2014). The inconsistent congru-
ence–satisfaction relations may be due to a variety of methods used to operationalize interest
congruence (e.g., Assouline & Meir, 1987; Tsabari et al., 2005). In particular, the magnitude of
congruence may depend on the specificity of interest measures and the methods used to generate
congruence indices.
Interest Measures at Different Taxonomy Levels
Interest measures can be generally divided into three levels based on the degree of specificity
when representing a full range of individual interests (Hansen, 1984): broadband general
interests at the top level, specific occupational interests at the bottom level, and basic interests
in the middle.
Holland’s (1973, 1997) RIASEC model is built at the general interest level and has been the
dominant theory to guide the development of interest assessments within the last 50 years. Most
popular interest inventories, such as the Self-Directed Search (Holland, 1985), the General Occu-
pational Themes (GOTs) of the Strong Interest Inventory (SII; Donnay, Morris, Schaubhut, &
Thompson, 2005; Harmon, Hansen, Borgen, & Hammer, 1994), the American College Testing
(ACT) Interest Inventory (2009), and the Personal Globe Inventory (PGI; Tracey, 2002), are
designed to measure Holland’s six general interest types. Although the RIASEC model and associ-
ated measures have established great utility in career guidance and other practices, the wide endor-
sement of Holland’s model is more for practicality rather than theoretical reasons. In fact, RIASEC
types were initially developed through a restricted range of occupational titles, and many factorial
analytic studies on interest structure identified more than six general interest factors (Rounds, 1995).
Therefore, researchers have started to notice that Holland’s six RIASEC general interest types may
be too broad to reflect the full range of career interests of individuals across communities and
cultures (Armstrong, Smith, Donnay, & Rounds, 2004; Deng, Armstrong, & Rounds, 2007; Donnay
& Borgen, 1996). Similarly, the RIASEC-based interest inventories would suffer from the same
limitation. In fact, a majority of previous studies on interest congruence adopted the RIASEC-based
interest inventories and resulted in generally low congruence–satisfaction relations. This may imply
that RIASEC measures are not a reliable predictor of vocational outcomes (Tinsley, 2000).
Recently, researchers have started to use finer interest taxonomies in developing interest inven-
tories instead of relying on Holland’s RIASEC model (Donnay et al., 2005; Harmon et al., 1994;
Liao et al., 2008). Their work mainly focuses on basic interests, which are more specific than
Holland’s six general interests, yet not as narrowly defined as occupational titles. Compared to
Bai and Liao 3
general interests, basic interests have the advantages of adequately capturing the complexity of
interest domains and providing a more fine-grained understanding of individual interests; therefore,
they are suggested to have greater utility in predicting career outcomes (e.g., Armstrong et al., 2004;
Day & Rounds, 1997; Liao et al., 2008). Several basic interest measures have demonstrated superior
capability in predicting career and academic choices as compared to the general interest measures.
For instance, the Basic Interests Scales (BISs; Donnay et al., 2005; Harmon et al., 1994) were proved
to be a more accurate predictor than the RIASEC-based GOTs with regard to occupational group
membership and academic majors (Donnay & Borgen, 1996; Gasser, Larson, & Borgen, 2007;
Ralston et al., 2004). The BISs were also found better at distinguishing between satisfied and
dissatisfied workers (Rottinghaus et al., 2009) than RIASEC measures. Another recently developed
public-domain basic interest measure, the basic interests markers (BIMs; Liao et al., 2008), was also
found to be superior to the RIASEC-based GOTs in differentiating college students’ academic
majors (Liao et al., 2008).
However, the existing validity evidence associated with the basic interest measures is on the basic
interest scores, not the interest congruence indices. To date, no study has directly investigated the
relations between interest congruence and career outcomes (e.g., satisfaction, achievement) based on
the basic interest taxonomy. The present study sought to fill the gap by exploring the congruence–
satisfaction relation at the basic interest level. As a comparison, the congruence–satisfaction relation
was also explored using a RIASEC general interest measure. We hypothesized that the congruence–
satisfaction relation would remain significantly positive and stronger when the basic interest mea-
sure is used than when the general interest measure is used.
Interest Congruence Indices
Interest congruence indices are usually calculated by comparing an individual’s RIASEC profiles (or
codes) and RIASEC profiles (or codes) of the environment. A myriad of methods using RIASEC
profiles or codes has been used to calculate congruence indices (see Brown & Gore [1994], Camp &
Chartrand [1992], for an overview of the congruence indices). For examples, in the first letter
hexagon distance (FLHD) index, Holland (1973) compared the first-letter individual and environ-
mental RIASEC codes and assigned an integer value ranging from 1 to 4 to present the degree of
congruence according to their relative positions on the RIASEC hexagon (1 ¼opposite on hexagon,
2¼alternate,3¼adjacent, and 4 ¼identical). As for a more sophisticated index, Brown and Gore
(1994) compared the first three letters of the RIASEC codes between individuals and environments
and used a formula to calculate an index score (also called the C index). In general, most prior
researchers adopted the matching of the high-point interest codes to generate interest congruence
indices. However, omitting other interest profiles or codes may affect the accuracy of congruence
index (De Fruyt, 2002).
Two interest congruence indices considering the complete RIASEC profiles have been
recommended and commonly used in recent studies (Allen & Robbins, 2010; Tracey, Allen,
& Robbins, 2012). The first index uses Euclidean distance between two RIASEC interest
points from the individual and the environment (e.g., occupation) in a two-dimensional space.
The second index uses profile correlation (McCloy, Campbell, Oswald, Lewis, & Rivkin,
1999) to estimate the match between individuals’ complete interests profiles and those of
environments. Unlike the Euclidean distance index, the profile correlation index (herein
shortened as profile index) is not bound with the RIASEC two-dimensional hexagonal struc-
ture. As basic interests are not strictly bound with Holland’s RIASEC model, we argued that
profile correlation can be a useful method to calculate interest congruence at the basic
interest level. Therefore, in the present study, we chose to use the profile index to represent
the basic interest congruence index.
4Journal of Career Assessment XX(X)
Overview of the Study
In two studies, we used two cross-cultural data sets to explore the relation between how congruent
college students’ interests match with their academic majors and their satisfaction with academic
majors. In Study 1, we recruited a large sample of college students in the United States and used a
RIASEC-based general interest measure (i.e., GOTs of the SII; Donnay et al., 2005; Harmon et al.,
1994) and a basic interest measure (i.e., BIMs; Liao et al., 2008) to generate congruence indices. In
Study 2, we recruited a sample of college students in Hong Kong and used another RIASEC-based
general interest measure (i.e., Chinese version of Personal Globe Inventory—Short [PGI-SC];
Zhang, Kube, Wang, & Tracey, 2013) and a basic interest measure (i.e., Hong Kong version of
basic interest markers, BIMs-HK; Davey, Bai, & Liao, 2016). As prior evidence indicates a superior
validity of basic interests in predicting career-related outcomes, we hypothesized that the relations
between interest congruence and satisfaction using basic interest measures would be significant and
stronger than the relations using general interest measures. In both studies, we used three interest
congruence indices: profile index (McCloy et al., 1999), FLHD (Holland, 1973), and C index
(Brown & Gore, 1994). As prior evidence indicates an advantage of using profile correlation to
generate congruence indices based on the complete interest profiles, we hypothesized that the
congruence–satisfaction relations would be stronger when the profile index is used than when the
FLHD and C index based on the high RIASEC codes are used.
Study 1
Participants and Procedures
The sample consisted of 1,044 college students from a variety of majors at a large Midwestern
university in the United States. The sample was recruited from the career development and advanced
psychology courses at six different time points from 2004 to 2008: Spring 2004 (n¼216), Fall 2004
(n¼169), Fall 2006 (n¼160), Spring 2007 (n¼143), Fall 2007 (n¼191), and Spring 2008 (n¼
165). Part of the sample collected between 2004 and 2006 was used to develop the BIMs (Liao et al.,
2008).
Among the 1,044 participants, 644 (61.7%) were women, 399 (38.2%) were men, and 1 (0.1%)
did not specify his or her gender. Their ages ranged from 17 to 39 years, with a mean age of 19.48
years (standard deviation [SD]¼1.69). The sample was distributed across college classes: 460
(44.1%) freshmen, 286 (27.4%) sophomores, 75 (7.2%) juniors, 214 (20.5%) seniors, 6 (0.6%)
graduate students, and 3 (0.3%) unknown. The racial breakdown of the sample was as follows:
595 Whites/Caucasians (57.0%), 246 Blacks/African Americans (23.6%), 82 Latinos/Hispanics
(7.9%), 71 Asians/Asian Americans (6.8%), 45 multiracial/biracial and Others (4.3%), and 5
unknown (0.5%).
The study project was approved by the institutional review board of the participating university.
Among 1,044 students, all completed the BIMs, 956 completed the GOTs of the Strong Interest
Inventory (Donnay et al., 2005; Harmon et al., 1994), and 819 reported their satisfaction with their
major fields of study.
Measures
BIMs. The original BIMs (Liao et al., 2008) contain 31 basic interest scales (e.g., business, mathe-
matics, teaching, law, and finance) with 343 items consisting of short and contextualized verbal
activity phrases. Sample items include “plan the expansion of a company” (business), “solve an
algebraic equation” (mathematics), and “develop a lecture” (teaching). For the purpose of the study,
we chose 30 basic interest scales (323 items total) from the original BIMs, excluding the Family
Bai and Liao 5
Activity Scale (20 items). Students were asked to rate to what degree they like the activities on a
Likert-type scale ranging from 1 (strongly dislike)to5(strongly like). Cronbach’s as of the 30 BIM
scales for Study 1 ranged from .85 to .95.
We first obtained the 30 BIM raw scores by averaging scale items and then standardized these
raw scores into standard z-scores based on the mean BIM raw scores from the entire sample of 1,044
students. We chose to use standard scores because they are more comparable among different scales
(Cohen & Swerdlik, 2010, pp. 108–109). As z-scores represent the numbers of SDs from the means,
a positive z-score indicates higher interest than the average of the whole sample, whereas a negative
z-score indicates lower interest than the average of the whole sample.
GOTs. Comprised of six scales, the GOTs, one of the content scales from the SII (Donnay et al., 2005;
Harmon et al., 1994), is a wildly used commercial interest measure which assesses Holland’s
RIASEC interest types. In the SII, students were asked to indicate their preferences for a variety
of people, environments, and leisure activities using a 5-point Likert-type scale (1 ¼strongly dislike,
5¼strongly like). Students recruited at earlier time points (Spring 2004 and Fall 2004) completed
the 1994 SII (Harmon et al., 1994; n¼316) and those recruited at later time points (Fall 2006 to
Spring 2008) completed the 2005 SII (Donnay et al., 2005; n¼640). The six GOT scores used in this
study are standard scores (M¼50, SD ¼10). Both 1994 and 2005 SIIs yielded good internal
consistencies, with Cronbach’s as over .90 for the GOTs (Donnay et al., 2005; Harmon et al.,
1994). Although the 2005 GOTs contain more items (153 in total) than the 1994 GOTs (140 items),
both structures are essentially the same (Donnay et al., 2005). Additionally, the 1994 GOT and 2005
GOT scales are highly correlated (rs ranged from .90 to .98), suggesting that the six GOTs remain
consistent between two editions and can be combined for analyses.
Satisfaction with academic majors. In the demographic questionnaire of the online SII, students rated
their satisfactions with their current major or course of study in a single item on a 6-point Likert-type
scale ranging from 1 (very satisfied)to6(very dissatisfied). Studies have indicated that the use of
single item in assessing satisfaction is acceptable (e.g., Lounsbury, Saudargas, Gibson, & Leong,
2005; Robinson, Shaver, & Wrightsman, 1991). We reversed the item scores in the present study,
with higher scores indicating higher levels of satisfaction.
Analyses and Results
Students’ interest profiles. We first generated students’ interest profiles. As indicated earlier, we
transformed the 30 BIM raw scores into standard z-scores (M¼0, SD ¼1). Since the 6 GOT
scores are standard scores (M¼50, SD ¼10), there was no need for further transformation. We used
the 30 standard BIM scores to represent students’ basic interest profiles, and the 6 GOT scores
(RIASEC scores) to represent students’ general interest profiles.
Major interest profiles. Next, we generated interest profiles for academic majors. Based on Holland’s
(1997) assumption that environments are made up of people, we used mean interest scores of
participants within each major to represent the interest profiles of each academic major. This
strategy has been adopted in prior research (e.g., Allen & Robbins, 2010). In this study, the 30
mean BIM scores of participants within a major represent a major’s basic interest profile. Similarly,
the 6 mean GOT scores of participants within a major represent a major’s general interest profile.
It is important to note that a certain number of participants are required to get a reliable major
interest profile. Following Tracey, Allen, and Robbins (2012), we chose to analyze the major interest
profiles with at least 20 students per major. As a result, 322 students from nine majors (i.e.,
psychology, biology, marketing, advertising, political science, sociology, economics, finance, and
6Journal of Career Assessment XX(X)
communication) were selected. Among 322 selected students, all completed the BIMs, 271 com-
pleted the SII with GOT scores, and 219 reported satisfaction with their majors.
1
Table 1 presents the
general interest profiles and basic interest profiles of these nine majors.
Table 1. Interest Profiles and Interest Codes of Nine Majors (Study 1).
Interest Profiles Psy. Bio. Mar. Adv. Pol. Soc. Eco. Fin. Com.
General interest profiles
1. Realistic (R) 43.49 46.67 40.31 41.63 44.49 44.27 49.10 43.12 43.94
2. Investigative (I) 43.26 52.94 37.07 37.00 43.02 46.71 46.56 41.59 41.81
3. Artistic (A) 50.34 44.59 45.13 48.88 42.14 50.03 44.48 41.69 46.16
4. Social (S) 57.69 50.68 49.59 47.18 47.97 59.71 48.35 47.46 52.05
5. Enterprising (E) 52.58 48.76 60.38 60.00 49.60 54.35 58.79 60.00 60.50
6. Conventional (C) 47.07 47.06 46.62 48.15 50.14 49.20 59.64 56.08 50.94
General interest three-letter RIASEC codes
SEA ISE ESC EAC CES SEA CER ECS ESC
Basic interest profiles
1. Athletic coaching 0.22 0.18 0.19 0.26 0.03 0.14 0.37 0.15 0.06
2. Business 0.41 0.41 0.90 0.49 0.19 0.02 0.87 0.81 0.43
3. Creative arts 0.24 0.01 0.08 0.08 0.32 0.21 0.26 0.66 0.13
4. Creative writing 0.28 0.08 0.15 0.03 0.13 0.37 0.21 0.55 0.06
5. Engineering 0.36 0.36 0.48 0.25 0.17 0.15 0.47 0.03 0.18
6. Finance 0.46 0.24 0.51 0.00 0.14 0.06 1.19 1.25 0.07
7. Human relations management 0.28 0.23 0.33 0.08 0.18 0.49 0.52 0.31 0.46
8. Information technology 0.30 0.06 0.06 0.28 0.18 0.07 0.78 0.23 0.00
9. Law 0.04 0.14 0.31 0.17 1.51 0.10 0.19 0.06 0.01
10. Life science 0.17 1.12 0.54 0.45 0.29 0.15 0.03 0.34 0.33
11. Management 0.16 0.04 0.61 0.07 0.40 0.32 0.64 0.61 0.34
12. Manual labor 0.32 0.04 0.45 0.26 0.14 0.11 0.10 0.11 0.14
13. Mathematics 0.25 0.18 0.18 0.37 0.19 0.02 0.59 0.51 0.35
14. Medical service 0.26 1.21 0.69 0.74 0.37 0.25 0.10 0.49 0.51
15. Outdoor agriculture 0.16 0.39 0.36 0.27 0.40 0.00 0.25 0.21 0.26
16. Office work 0.20 0.45 0.34 0.14 0.12 0.20 0.51 0.48 0.26
17. Performing arts 0.39 0.14 0.13 0.01 0.36 0.37 0.17 0.45 0.03
18. Personal service 0.22 0.05 0.42 0.09 0.30 0.46 0.06 0.23 0.26
19. Physical risk taking 0.18 0.29 0.10 0.18 0.47 0.30 0.45 0.04 0.02
20. Physical science 0.27 0.91 0.62 0.38 0.25 0.09 0.32 0.26 0.27
21. Politics 0.09 0.05 0.20 0.08 1.55 0.24 0.26 0.17 0.14
22. Professional advising 0.25 0.08 0.50 0.06 0.09 0.52 0.41 0.36 0.38
23. Protective 0.08 0.39 0.46 0.57 0.63 0.35 0.10 0.14 0.00
24. Religious activities 0.15 0.19 0.64 -0.09 0.01 0.03 0.00 0.19 0.14
25. Sales 0.38 0.22 0.87 0.53 0.21 0.03 0.51 0.67 0.64
26. Skilled trades 0.32 0.19 0.40 0.27 0.06 0.02 0.27 0.12 0.05
27. Social science 0.72 0.10 0.26 0.29 0.07 0.85 0.09 0.58 0.00
28. Social service 0.70 0.04 0.26 0.36 0.08 0.88 0.44 0.46 0.08
29. Teaching 0.43 0.05 0.02 0.29 0.15 0.79 0.09 0.34 0.03
30. Technical writing 0.29 0.14 0.27 0.14 0.08 0.26 0.43 0.02 0.06
Note. Psy ¼psychology; bio. ¼biology; mar. ¼marketing; adv. ¼advertising; pol. ¼political science; soc. ¼sociology; eco. ¼
economics; fin. ¼finance; com. ¼communication. The general interest profiles were derived from the General Occupational
Themes of the Strong Interest Inventory (Donnay et al., 2005; Harmon et al., 1994), and the basic interest profiles were
derived from the Basic Interest Markers (Liao et al., 2008).
Bai and Liao 7
Interest congruence indices. In the current study, we used three congruence indices: profile index
(McCloy et al., 1999), FLHD (Holland, 1973), and C index (Brown & Gore, 1994). As mentioned
earlier, the profile index relies on the complete interest profiles, while the FLHD index and C index
use the high RIASEC codes.
For the profile index, we obtained the correlations between students’ basic interest profiles and
the basic interest profiles of their majors. Similarly, we obtained the correlations between students’
general interest profiles and the general interest profiles of their majors. Higher profile correlations
indicate greater congruence while lower or negative profile correlations indicate less or poor con-
gruence. To avoid the dependency between a student’s profile and the profile of his or her major
when generating the profile index, each student’s major profile was captured by the mean interest
scores of all students within the same major without that student (Su, 2012).
As the FLHD and C index are based on the high RIASEC codes, we obtained these two indices
only at the general interest level. The computations of the FLHD and C index require exact one-letter
and three-letter RIASEC codes, so we used students’ highest one and three RIASEC profiles to form
their one-letter and three-letter RIASEC codes. Similarly, we used the highest one-letter and three-
letter RIASEC profiles of the selected nine majors to form one-letter and three-letter RIASEC codes
of majors. There were 19 ties when generating students’ RIASEC codes. We followed Bowles’s
(2008) procedure to resolve the ties where the tied codes were ranked in a way that the arrangement
can represent the highest degree of consistency for the three-letter codes (p. 161).
2
Table 1 presents
the three-letter RIASEC codes of nine majors. For the FLHD (Holland, 1973), we compared each
student’s first-letter RIASEC code with the first-letterRIASECcodeofhisorhermajorand
assigned an integer value according to the relative positions of these two codes on the RIASEC
hexagon (1 ¼opposite on hexagon,2¼alternate,3¼adjacent, and 4 ¼identical). The C index
(Brown & Gore, 1994) was determined by comparing each student’s three-letter RIASEC code and
the three-letter RIASEC code of his or her major. Particularly, we compared the first, second, and
third letters of each student and his or her major sequentially and assigned an integer value from 0 to
3 per comparison (0 ¼opposite on hexagon,1¼alternate,2¼adjacent, and 3 ¼identical). Then,
the C index score was calculated as the sum of weighted values of three comparisons using the
formula of C¼3(X
1
)þ2(X
2
)þ(X
3
), where X
1
,X
2
, and X
3
are the assigned values of the first,
second, and third letter comparisons, respectively. The C index scores range from 0 to 18, with
higher scores reflecting greater congruence.
Relation between interest congruence and major satisfaction. We used the Pearson correlation to measure
the relations between four interest congruence indices (i.e., basic interest profile index [BIPI],
general interest profile index [GIPI], FLHD, and C index) and students’ satisfaction with their
academic major. Since 103 of 322 (32%) selected students did not report satisfaction with majors,
we compared the four congruence indices between students who reported satisfaction with majors
(n¼219) and those who did not (n¼103). Results of independent sample ttest showed no
significant differences between two groups on any congruence index, BIPI: t(320) ¼0.92, GIPI:
t(101) ¼1.43, FLHD: t(269) ¼0.74; and C index: t(269) ¼0.57, ns. Therefore, we used pairwise
deletion to generate congruence–satisfaction correlations. Table 2 presents means, SDs, and inter-
correlations of four congruence indices and students’ satisfaction with academic majors. Results
showed that all four congruence indices were significantly correlated with students’ satisfaction with
majors (ps < .05), with the BIPI–satisfaction correlation (r¼.26) being the highest one, followed by
the correlations generated from the GIPI (r¼.23), FLHD (r¼.18), and C index (r¼.15).
To further examine the incremental validity of the BIPI over the GIPI, we performed the hier-
archical regression analysis as suggested by Cohen and Swerdlik (2010, p. 196), with the GIPI
entering as the predictor in Step 1 and the BIPI entering in Step 2 to predict students’ major
satisfaction. Results showed that the basic interest-based BIPI significantly improved the prediction
8Journal of Career Assessment XX(X)
of students’ major satisfaction beyond that provided by the general interest-based GIPI, DR
2
¼.03,
DF(1, 216) ¼7.01, p< .01. When the BIPI was added in Step 2 as a predictor, the BIPI significantly
predicted students’ major satisfaction (b¼.19, p< .01), yet the GIPI marginally predicted students’
major satisfaction (b¼.13, p¼.05).
Study 2
Participants and Procedures
The sample for Study 2 consisted of 455 college students from a variety of majors at a comprehen-
sive university in Hong Kong. The student sample was recruited through the university mass mail
system. All participants completed a paper–pencil questionnaire and were reimbursed with small
incentives for their participation. Among the 455 participants, 324 (71.2%) were women and 131
(28.8%) were men. Their ages ranged from 18 to 40 years with a mean age of 20.45 years
(SD ¼2.30). The sample was distributed across college classes: 159 (34.9%) freshmen, 88
(19.3%) sophomores, 159 (34.9%) juniors, 15 (3.3%)seniors,and34(7.5%) graduate students.
Study 2 was approved by the survey and behavioral research ethics committee of the partici-
pating university.
Measures
BIM-HK. Based on the BIMs, the BIM-HK (Davey et al., 2016) was developed for the Hong Kong
context using an imposed–indigenous approach. It contains 32 basic interest scales with 465 items.
Similar to Study 1, we chose 31 basic interest scales (451 items total) from the BIM-HK, excluding
the Family Activity Scale (14 items). Students were asked to rate to what degree they like the
activities described by each item on a Likert-type scale ranging from 1 (strongly dislike)to5
(strongly like). The Cronbach’s as of the 31 BIM-HK scales in Study 2 ranged from .88 to .96.
PGI-SC. The PGI-SC (Zhang et al., 2013) was translated and revised from the PGI-S (Tracey, 2010),
an abbreviated version of the PGI (Tracey, 2002). The PGI-SC has been validated in China as an
alternative measure to the full version of the PGI (Zhang et al., 2013). The PGI-SC consists of 10
scales (4 items for each scale), including eight interest scales and two prestige scales. The scores of
the eight interest scales can be transformed into 6 RIASEC scores. For Study 2, we used the
transformed RIASEC scores to represent the general RIASEC interests. The PGI-SC has been
validated in several diverse samples of Chinese high school and college students, with Cronbach’s
as ranging from .73 to .83 (Zhang et al., 2013). Students were asked to rate to what degree they like
the activities described by each item on a Likert-type Scale ranging from 1 (strongly dislike)to7
(strongly like). Cronbach’s as of the eight interest scales in Study 2 ranged from .69 to .89.
Table 2. Descriptive Statistics and Intercorrelations of Four Interest Congruence Indices and Satisfaction With
Academic Majors (Study 1).
Variables nRange MSD12345
1. Satisfaction with academic majors 219 1–6 4.62 1.05 —
2. Basic interest profile index 322 1 to 1 0.33 0.34 .26** —
3. General interest profile index 271 1 to 1 0.61 0.37 .23** .44** —
4. First letter hexagonal distance 271 1 to 4 3.38 0.82 .19** .36** .63** —
5. C index 271 0 to 18 12.56 3.49 .16* .31** .66** .76** —
*p< .05. **p< .01.
Bai and Liao 9
Satisfaction with academic major. Participants’ satisfactions with their academic majors were measured
by the 6-item Academic Major Satisfaction Scale (AMSS; Nauta, 2007). Respondents rated their
agreement with the items using a 5-point Likert-type scale from 1 (strongly disagree)to5(strongly
agree). Examples of items are “I often wish I hadn’t gotten into this major” (reverse scored) and
“Overall, I am happy with the major I’ve chosen.” Cronbach’s afor the 6 AMSS items was .90
(Nauta, 2007). Similarly, Cronbach’s afor the AMSS in Study 2 was .90.
Analyses and Results
We first transformed both basic interest scores and general interest scores (i.e., RIASEC scores)
of each participant into standard z-scores (M¼0, SD ¼1) based on the entire sample of 455
students. We then adopted the methods used in Study 1 to analyze the data. Accordingly, 193
students from six majors (i.e., business administration, accountancy, engineering, nursing,
biology, and social work) were selected and their interest congruence indices were generated.
Table 3 presents the general interest profiles and basic interest profiles of these six majors.
Table 4 presents means, standard deviations, and intercorrelations of four congruence indices
and students’ major satisfaction. All four congruence indices were significantly correlated with
students’ satisfaction with majors (ps < .05), with the BIPI–satisfaction correlation (r¼.34)
being the highest one, followed by the correlations generated from the GIPI (r¼.21), FLHD
(r¼.20), and C index (r¼.15). Results of the hierarchical regression analysis also showed
that BIPI significantly improved the prediction of students’ major satisfaction beyond that
provided by GIPI, DR
2
¼.08, DF(1, 190) ¼16.06, p<.01.WhentheBIPIwasaddedinStep
2 as a predictor, only the BIPI significantly predicted students’ major satisfaction (b¼.34,
p< .01), not the GIPI (b¼.00, p¼.98).
Discussion
The current study provides empirical evidence on the utility of basic interest measures in measuring
P–E interest congruence and predicting career-related outcomes in a cross-cultural context. In the
United States and Hong Kong samples, the results demonstrated significant and positive congru-
ence–satisfaction relations using basic interest measures. Moreover, major satisfaction was most
strongly related to the congruence index built off the basic interest taxonomy (BIPI; r
Study 1
¼.26;
r
Study 2
¼.34), followed by the index generated from the complete general interest profiles (GIPI;
r
Study 1
¼.23; r
Study 2
¼.21) and the two traditional RIASEC indices—FIHD (r
Study 1
¼.19;
r
Study 2
¼.20) and C index (r
Study 1
¼.16; r
Study 2
¼.15). These results advocate the use of the basic
interest taxonomy and complete interest profiles.
Relation Between Interest Congruence and Satisfaction in Academic Settings
The relation between interest congruence and career-related satisfaction has been inconsistent and
generally low across prior studies, especially in academic settings. As reported from prior meta-
analyses, the mean interest congruence–satisfaction correlations ranged from .17 to .24 across all
studies and were much lower in academic settings, ranging from .03 to .10 (Assouline & Meir,
1987; Morris, 2003; Tranberg et al., 1993; Tsabari et al., 2005). Recent empirical research even
showed interest congruence unrelated to satisfaction with academic majors (Pozzebon et al., 2014).
Some researchers have suggested that such varied congruence–satisfaction relations may be due to
how congruence indices are generated. In essence, the use of general interest measures and simple
congruence indices may have resulted in lower congruence–satisfaction relations (e.g., Edwards,
2007; Tinsley, 2000). In the current study, we reexamined the congruence–satisfaction relations with
10 Journal of Career Assessment XX(X)
a refined interest congruence index using narrowband basic interest measures and considering the
entire interest profiles. Results showed that congruence–satisfaction relations do vary according to
congruence indices. As expected, when basic interest measures and complete interest profiles were
Table 3. Interest Profiles and Interest Codes of Six Majors (Study 2).
Interest profiles Bus. Acc. Eng. Nur. Bio. Soc.
General interest profiles
1. Realistic (R) 0.01 0.17 1.04 0.28 0.16 0.44
2. Investigative (I) 0.27 0.52 0.17 0.11 0.65 0.18
3. Artistic (A) 0.28 0.39 0.13 0.06 0.17 0.29
4. Social (S) 0.07 0.14 0.31 0.10 0.10 0.58
5. Enterprising (E) 0.25 0.70 0.18 0.27 0.30 0.03
6. Conventional (C) 0.22 0.55 0.50 0.34 0.11 0.61
General interest three-letter RIASEC codes
ECR ECS RCI ISA IRS SAE
Basic interest profiles
1. Agriculture 0.53 0.41 0.02 0.17 0.53 0.03
2. Athletic 0.06 0.14 0.32 0.23 0.63 0.22
3. Business 0.31 0.49 0.08 0.52 0.25 0.52
4. Communication and media 0.10 0.11 0.44 0.34 0.46 0.06
5. Creative arts 0.35 0.28 0.01 0.05 0.19 0.08
6. Creative writing 0.15 0.33 0.42 0.25 0.34 0.20
7. Engineering 0.08 0.27 0.94 0.16 0.20 0.62
8. Finance 0.43 0.84 0.05 0.30 0.08 0.65
9. Humanity social science 0.46 0.42 0.20 0.25 0.27 0.22
10. Human resources 0.15 0.41 0.15 0.26 0.23 0.15
11. Information technology 0.15 0.13 1.03 0.36 0.16 0.52
12. Law 0.21 0.08 0.08 0.17 0.27 0.03
13. Life science 0.33 0.53 0.19 0.37 0.94 0.44
14. Management 0.19 0.51 0.02 0.39 0.18 0.34
15. Manual labor 0.01 0.02 0.38 0.19 0.16 0.08
16. Mathematics 0.11 0.05 0.73 0.25 0.08 0.57
17. Medical service 0.31 0.46 0.07 1.12 0.85 0.40
18. Office work 0.12 0.64 0.07 0.00 0.18 0.22
19. Performing arts 0.36 0.04 0.24 0.22 0.09 0.06
20. Personal service 0.23 0.30 0.25 0.04 0.04 0.20
21. Physical risk taking 0.06 0.40 0.20 0.14 0.48 0.08
22. Physical science 0.19 0.47 0.59 0.13 0.37 0.54
23. Politics 0.01 0.12 0.05 0.35 0.26 0.39
24. Professional advising 0.02 0.30 0.23 0.15 0.23 0.40
25. Protective 0.15 0.03 0.39 0.11 0.20 0.08
26. Religious activities 0.15 0.33 0.13 0.07 0.16 0.24
27. Sales 0.29 0.46 0.03 0.40 0.09 0.27
28. Skilled trades 0.06 0.19 0.77 0.11 0.29 0.39
29. Social service 0.43 0.34 0.25 0.28 0.05 0.99
30. Teaching 0.19 0.13 0.26 0.23 0.35 0.37
31. Technical writing 0.01 0.15 0.62 0.38 0.10 0.35
Note. Bus. ¼business administration; acc. ¼accountancy; eng. ¼engineering; nur. ¼nursing; bio. ¼biology; soc. ¼social
work. The general interest profiles were derived from the Chinese version of Personal Globe Inventory—Short (Zhang et al.,
2013), and the basic interest profiles were derived from the Hong Kong version of Basic Interest Markers (Davey et al., 2016).
Bai and Liao 11
used to generate the congruence indices, the congruence–satisfaction correlations (r
Study 1
¼.26;
r
Study 2
¼.34) were the highest.
Generally, a correlation coefficient around .30 is usually considered a weak to moderate correla-
tion (Coolican, 2014, p. 524). However, a weak congruence–satisfaction correlation does not imply
that the concurrent validity of a particular interest measure is poor. First, interest measures cannot
directly measure the true interest congruence between persons and environments. Rather, interest
measures indirectly assess individuals’ interests by using selected interest categories to generate
their interest profiles. The environmental interest profiles cannot be measured directly either, as they
are often inferred from the individual interest profiles of the same environment (e.g., occupation,
academic majors). Second, to generate interest congruence indices, additional mathematical oper-
ations and transformations are needed in order to represent the match between persons and environ-
ments. These mathematical operations oftentimes obscure other important information provided by
interest measures (Edwards, 2007). Third, one’s satisfaction can be affected by other factors besides
interest congruence, such as social context, job stressors, and personal disposition (Fritzsche &
Parrish, 2005). Students’ academic satisfaction may be also affected by variables such as career
self-efficacy, career optimism, and conscientiousness (Komarraju, Swanson, & Nadler, 2013;
McIlveen, Beccaria, & Burton, 2013).
Taken together, due to the indirect nature of congruence indices and various factors toward
satisfaction, it is understandable that the interest congruence–satisfaction correlation is not as high
as most concurrent validity evidence found in psychological assessments. Some scholars argued that
small correlations are crucial from the theoretical and practical standpoints (Aron, Coups, & Aron,
2012). It is also suggested that even a small amount of contribution, like 5–10%of satisfaction
variance explained by congruence indices, is meaningful for one’s career development (e.g., Tsabari
et al., 2005). Therefore, in the current study, the congruence–satisfaction correlations based on the
basic interest taxonomy (r
Study 1
¼.26; r
Study 2
¼.34) are considered acceptable and meaningful.
Basic Interest Measures Versus General Interest Measures
In the present study, we compared the utility of two types of interest measures: basic interest
measures and general interest measures. As indicated earlier, basic interest measures have been
reported to have superior concurrent validity to general interest measures in differentiating college
students’ major fields of study (Liao et al., 2008), occupational membership (Donnay & Borgen,
1996), and workers’ satisfaction (Ralston et al., 2004; Rottinghaus et al., 2009). In the present study,
we provide another evidence of concurrent validity by connecting interest congruence to satisfac-
tion. When considering the complete interest profiles, the congruence indices based on the basic
interest measures generated stronger congruence–satisfaction correlations (r
Study 1
¼.26; r
Study 2
¼
.34) than the indices based on the general interest measures (r
Study 1
¼. 23; r
Study 2
¼.21). In both
Table 4. Descriptive Statistics and Intercorrelations of Four Interest Congruence Indices and Satisfaction With
Academic Majors (Study 2).
Variables Range MSD12345
1. Satisfaction with academic majors 1–5 3.46 0.83 —
2. Basic interest profile index 1 to 1 0.40 0.25 .34** —
3. General interest profile index 1 to 1 0.39 0.49 .21** .61** —
4. First letter hexagonal distance 1 to 4 3.07 0.92 .20** .46** .62** —
5. C index 0 to 18 11.26 3.77 .15* .42** .66** .80** —
Note.n¼193.
*p< .05. **p< .01.
12 Journal of Career Assessment XX(X)
Study 1 and Study 2, the congruence indices basedonthebasicinterestmeasuresimprovedthe
prediction of students’ major satisfaction beyond that provided by the congruence indices based on
the general interest measures. The increment was significant in the U.S. sample (Study 1; R
2
improved from .05 to .08) and much more substantial in the Hong Kong sample (Study 2; R
2
from
.04 to .12). Notably, in both samples, when general and basic interest congruence indices were
included to predict major satisfaction, the general interest congruence indices did less or almost
none contribution (in the Hong Kong sample) as compared to the basic interest congruence
indices. The results suggest the basic interest measure alone can provide abundant information
on individual and environmental vocational interests in generating congruence indices and pre-
dicting career outcomes.
Although there is ample evidence supporting the superior validity of the basic interest measures,
most basic interest measures are still perceived as a subdivision or supplementary to the RIASEC
general interest measures (Liao et al., 2008). In the current study, we demonstrated that basic interest
measures alone can predict career outcomes better than the widely used RIASEC measures. In
addition to the superior validity in predicting career outcomes, basic interest measures can be a
useful tool to assess jobs or academic majors which are hard to define using general interest profiles.
For example, some fields, such as technology and communication (e.g., managing social media
online), have been overlooked under traditional general interest models since these fields involve
multiple general interest types (e.g., Artistic [A], Social [S], Enterprising [E], Investigative [I] for
social media-related work). In this case, basic interests appear better in covering a wide variety of
occupational and educational environments by offering a comprehensive set of finer and specific
profiles. Therefore, we call for future endeavors to promote the research and applications of basic
interest measures.
It is interesting to note that the concept of interest congruence was originally proposed by Hol-
land, and most prior studies on interest congruence were conducted within the RIASEC hexagon
framework. Since Holland’s model has been challenged and new interest assessments and frame-
works have been developed as alternatives to Holland’s RIASEC hexagon (e.g., the basic interests,
Campbell et al., 1968; the hierarchical model, Gati, 1991), it is valuable to jump out of the RIASEC
hexagon framework in assessing interest congruence and the congruence–satisfaction relation. In
this study, we adopted the basic interest taxonomy to estimate interest congruence and demonstrated
stronger congruence–satisfaction relations than the relations generated from the RIASEC congru-
ence indices. The current finding infers that congruence is a generic concept that does not need to be
bound with Holland’s RIASEC hexagon model.
Utility of Profile Correlation for Interest Congruence Index
In the present study, we found that the congruence–satisfaction correlations were strengthened
when we used the profile index, a profile correlation method, to estimate the match between
individuals’ complete interest profiles and those of environments. The use of profile correlation
as an interest congruence index has received increasing popularity within the past decades
(McCloy et al., 1999). The profile index is considered a more comprehensive and flexible
method than traditional indices since it captures the entire interest profiles and is not bound
with the RIASEC circumplex structure (Tracey et al., 2012). However, empirical research that
compares the utility of the profile index and traditional indices is scant. Recently, Phan and
Rounds (2014) compared the profile index with three traditional indices and found that the
profile index was the strongest predictor of job satisfaction. Consistent with Phan and Rounds
(2014), our study also found that the profile index can predict college students’ academic
satisfaction better than traditional indices.
Bai and Liao 13
Limitation and Future Direction
There are some limitations in the present study. First, in both U.S. and Hong Kong samples, we
chose limited academic majors and participants for analyses, as it requires a certain number of
participants to generate interest profiles for academic majors. Although Tracey et al. (2012) rec-
ommend at least 20 student participants per major, the relative small sample sizes are still challen-
ging to produce reliable results. Small sample sizes also affect the representativeness of the majors
since we included all students of the same majors to generate major profiles regardless of students’
level of preparedness and persistence. Additionally, it is desirable to obtain a separate sample to
generate major profiles in order to avoid dependency when calculating profile correlation. There-
fore, future researchers may consider recruiting a large, diverse, and representative sample to
develop interest profiles for academic majors or occupations, and if possible, this sample is inde-
pendent from the sample used for individuals’ interest profiles. We also encourage future research
using community-based and working samples. With sufficient sizes for a variety of majors or
occupations, researchers can further compare the congruence–satisfaction correlations across majors
or occupations.
Second, in this study, we tested the concurrent validity by simultaneously connecting students’
interests to their satisfaction. Future researchers may consider conducting longitudinal studies to
explore how interest congruence predicts later satisfaction and other career outcomes over time.
Third, since the current study only investigated one career outcome (i.e., satisfaction with major),
future researchers may explore other career or academic outcomes (e.g., achievement, persistence)
to strengthen the validity of the congruence–satisfaction findings. Fourth, when comparing the
findings between the U.S. and Hong Kong data sets, it may appear that the validity of basic interest
measures is more prominent in Hong Kong than in the United States. However, the measures used in
two studies were not identical. Therefore, such cross-cultural comparison cannot be concluded.
Future researchers could consider using culturally equivalent measures for cross-cultural
comparisons.
Implications for Career Guidance and Counseling
The current findings offer several implications for career guidance and counseling. First, the results
inform career practitioners that college students whose interests are more congruent with their
academic majors tend to have higher satisfaction with their majors. Second, interest measures that
are not based on the RIASEC model can also be a useful tool to facilitate students’ career explo-
ration. Third, as our findings suggest that basic interest measures are valid and useful tools in
measuring interest congruence and predicting satisfaction, we advocate a greater application of
basic interest measures in career assessment and practice. When considering model parsimony,
on the other hand, basic interest measures may not seem as practical as typical RIASEC general
interest measures due to their lengthy scales and items. Therefore, we call for the development of
shorter forms of basic interest measures. Lastly, the current findings suggest the value of using the
complete interest profiles to generate meaningful and accurate information for career practice.
However, interpreting the whole interest profiles, especially the basic interest profiles, may appear
overly complex. Therefore, we advocate the development of computerized programs that can handle
the complexity of interest profiles and provide the best matching majors or occupations for career
guidance and counseling.
Authors’ Note
The earlier portion of this article was presented at the 27th Association for Psychological Science Annual
Convention in New York in May 2015.
14 Journal of Career Assessment XX(X)
Acknowledgments
The authors thank Dr. Alvin Leung; Dr. James Rounds; the editor, Dr. Ryan D. Duffy; and two anonymous
reviewers for their valuable comments on the previous versions of this article. The authors specially thank
Mingdan Lu and Jinlu Cao for their assistance with the data coding.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or pub-
lication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publica-
tion of this article: The preparation of this article was partially supported by the Fujian Social Science Planning
Project of China (grant no. FJ2017C030) to Lili Bai.
Notes
1. Among 322 selected students, 51 (16%) students who did not complete the Strong Interest Inventory (SII)
were recruited from the advanced psychology courses as SII was not administered in these courses. For the
rest of 271 students who completed the basic interest markers and SII, 52 (19%) did not report the online
demographic question regarding satisfaction with their majors.
2. For example, if a student obtains the same scores on Rand Sand his or her three-letter RIASEC (Realistic,
Investigative, Artistic, Social, Enterprising, and Conventional) code is either ESR or ERS, we choose ESR
over ERS as ESR is more consistent than ESR.
References
American College Testing. (2009). ACT interest inventory technical manual. Retrieved from http://www.act.
org/research/researchers/pdf/ACTInterestInventoryTechnicalManual.pdf.
Allen, J., & Robbins, S. (2010). Effects of interest–major congruence, motivation, and academic performance
on timely degree attainment. Journal of Counseling Psychology,57, 23–35. doi:10.1037/a0017267
Armstrong, P. I., Smith, T. J., Donnay, D. A. C., & Rounds, J. (2004). The strong ring: A basic interest model of
occupational structure. Journal of Counseling Psychology,51, 299–313. doi:10.1037/0022-0167.51.3.299
Aron, A., Coups, E. J., & Aron, E. N. (2012). Statistics for psychology (6th ed.). Boston, MA: Pearson.
Assouline, M., & Meir, E. I. (1987). Meta-analysis of the relationship between congruence and well-being
measures. Journal of Vocational Behavior,31, 319–332. doi:10.1016/0001-8791(87)90046-7
Bowles, S. M. (2008). Is congruence dead? An examination of the correlation between Holland’s congruence
and job satisfaction using improved methodology (Doctoral dissertation). West Virginia University,
Morgantown, WV.
Brown, S. D., & Gore, P. A. Jr. (1994). An evaluation of interest congruence indices: Distribution characteristics
and measurement properties. Journal of Vocational Behavior,45, 310–327. doi:10.1006/jvbe.1994.1038
Camp, C. C., & Chartrand, J. M. (1992). A comparison and evaluation of interest congruence indices. Journal of
Vocational Behavior,41, 162–182. doi:10.1016/0001-8791(92)90018-U
Campbell, D. P., Borgen, F. H., Eastes, S. H., Johansson, C. B., & Peterson, R. A. (1968). A set of basic interest
scales for the strong vocational interest blank for men. Journal of Applied Psychology,52, 1–54. doi:10.
1037/h0026495
Clark, K. E. (1961). Vocational interests of nonprofessional men. Minneapolis: University of Minnesota Press.
Cohen, R. J., & Swerdlik, M. E. (2010). Psychological testing and assessment: An introduction to tests and
measurement (7th ed.). New York, NY: McGraw-Hill.
Coolican, H. (2014). Research methods and statistics in psychology (6th ed.). Hove, England: Psychology
Press.
Bai and Liao 15
Davey, N. S., Bai, L., & Liao, H. Y. (2016). Back to basics: Towards an emic-etic approach in cross-cultural
vocational assessment. Paper presented at the Div. 17 Student Poster Session at the Annual Convention of
the American Psychological Association, Denver, CO.
Dawis, R. V. (2005). The Minnesota theory of work adjustment. In S. D. Brown & R. W. Lent (Eds.), Career
development and counseling: Putting theory and research to work (pp. 3–23). Hoboken, NJ: Wiley.
Dawis, R. V., & Lofquist, L. H. (1984). A psychological theory of work adjustment: An individual-differences
model and its applications. Minneapolis: University of Minnesota Press.
Day, S. X., & Rounds, J. (1997). “A little more than kin, and less than kind”: Basic interests in vocational
research and career counseling. The Career Development Quarterly,45, 207–220. doi:10.1002/j.2161-0045.
1997.tb00465.x
De Fruyt, F. (2002). A person-centered approach to P–E fit questions using a multiple-trait model. Journal of
Vocational Behavior,60, 73–90. doi:10.1006/jvbe.2001.1816
Deng, C. P., Armstrong, P. I., & Rounds, J. (2007). The fit of Holland’s RIASEC model to US occupations.
Journal of Vocational Behavior,71, 1–22. doi:10.1016/j.jvb.2007.04.002
Donnay, D. A. C., & Borgen, F. H. (1996). Validity, structure, and content of the 1994 Strong Interest
Inventory. Journal of Counseling Psychology,43, 275–291. doi:10.1037/0022-0167.43.3.275
Donnay, D. A. C., Morris, M. L., Schaubhut, N. A., & Thompson, R. C. (2005). Strong interest inventory
®
manual. Mountain View, CA: Consulting Psychologists Press.
Edwards, J. R. (2007). Polynomial regression and response surface methodology. In C. Ostroff & T. Judge (Eds.),
Perspectives on organizational fit (pp. 361–372). New York, NY: Lawrence Erlbaum.
Fritzsche, B. A., & Parrish, T. J. (2005). Theories and research on job satisfaction. In S. D. Brown & R. W. Lent
(Eds.), Career development and counseling: Putting theory and research to work (pp. 180–202). Hoboken,
NJ: John Wiley & Sons.
Gasser, C. E., Larson, L. M., & Borgen, F. H. (2007). Concurrent validity of the 2005 strong interest inventory:
An examination of gender and major field of study. Journal of Career Assessment,15, 23–43. doi:10.1177/
1069072706294516
Gati, I. (1991). The structure of vocational interests. Psychological Bulletin,109, 309–324. doi:10.1037/0033-
2909.109.2.309
Hansen, J. C. (1984). The measurement of vocational interests: Issues and future directions. In S. D. Brown &
R. W. Lent (Eds.), Handbook of counseling psychology (pp. 99–136). New York, NY: Wiley.
Harmon, L. W., Hansen, J. C., Borgen, F. H., & Hammer, A. L. (1994). Strong Interest Inventory: Applications
and technical guide. Stanford, CA: Stanford University Press (Distributed by Consulting Psychologists Press).
Holland, J. L. (1973). Making vocational choices: A theory of careers (Vol. 37). Englewood Cliffs, NJ:
Prentice-Hall.
Holland, J. L. (1985). The self-directed search manual. Odessa, FL: Psychological Assessment Resources.
Holland, J. L. (1987). Some speculation about the investigation of person–environment transactions. Journal of
Vocational Behavior,31, 337–340. doi:10.1016/0001-8791(87)90048-0
Holland, J. L. (1997). Making vocational choices: A theory of vocational personalities and work environments
(3rd ed.). Odessa, FL: Psychological Assessment Resources.
Komarraju, M., Swanson, J., & Nadler, D. (2013). Increased career self-efficacy predicts college students’
motivation, and course and major satisfaction. Journal of Career Assessment,22, 420–432. doi:10.1177/
1069072713498484
Kulik, C. T., Oldham, G. R., & Hackman, J. R. (1987). Work design as an approach to person–environment fit.
Journal of Vocational Behavior,31, 278–296. doi:10.1016/0001-8791(87)90044-3
Liao, H. Y., Armstrong, P. I., & Rounds, J. (2008). Development and initial validation of public domain Basic
Interest Markers. Journal of Vocational Behavior,73, 159–183. doi:10.1016/j.jvb.2007.12.002
Lounsbury, J., Saudargas, R., Gibson, L., & Leong, F. (2005). An investigation of broad and narrow personality
traits in relation to general and domain-specific life satisfaction of college students. Research in Higher
Education,46, 707–729. doi:10.1007/s11162-004-4140-6
16 Journal of Career Assessment XX(X)
McCloy, R., Campbell, J., Oswald, F. L., Lewis, P., & Rivkin, D. (1999). Linking client assessment profiles to
O* NET occupational profiles. Raleigh, NC: National Center for O*NET Development.
McIlveen, P., Beccaria, G., & Burton, L. J. (2013). Beyond conscientiousness: Career optimism and satisfaction
with academic major. Journal of Vocational Behavior,83, 229–236. doi:10.1016/j.jvb.2013.05.005
Morris, M. A. (2003). A meta-analytic investigation of vocational interest-based job fit, and its relationship to
job satisfaction, performance, and turnover. Dissertation Abstracts International: Section B: The Sciences
and Engineering,64, 2428.
Nauta, M. M. (2007). Assessing college students’ satisfaction with their academic majors. Journal of Career
Assessment,15, 446–462. doi:10.1177/1069072707305762
Phan, W. M. J., & Rounds, J. (2014). Interest congruence and job satisfaction: Methodological issues and
potential solutions. Paper presented at the 29th Annual Conference of the Society for Industrial and
Organizational Psychology, Honolulu, HI.
Pozzebon, J. A., Ashton, M. C., & Visser, B. A. (2014). Major changes personality, ability, and congruence in
the prediction of academic outcomes. Journal of Career Assessment,22, 75–88. doi:10.1177/
1069072713487858
Ralston, C. A., Borgen, F. H., Rottinghaus, P. J., & Donnay, D. A. C. (2004). Specificity in interest measure-
ment: Basic Interest Scales and major field of study. Journal of Vocational Behavior,65, 203–216. doi:10.
1016/S0001-8791(03)00096-4
Robinson, J. P., Shaver, P. R., & Wrightsman, L. S. (1991). Measures of personality and social psychological
attitudes. San Diego, CA: Academic Press.
Rottinghaus, P. J., Hees, C. K., & Conrath, J. A. (2009). Enhancing job satisfaction perspectives: Combining
Holland themes and basic interests. Journal of Vocational Behavior,75, 139–151. doi:10.1016/j.jvb.2009.
05.010
Rounds, J. (1995). Vocational interests: Evaluating structural hypotheses. In D. J. Lubinski & R. V. Dawis (Eds.),
Assessing individual differences in human behavior: New concepts, methods, and findings (pp. 177–232).
Palo Alto, CA: Davies-Black.
Spokane, A. R. (1985). A review of research on person–environment congruence in Holland’s theory of careers.
Journal of Vocational Behavior,26, 306–343. doi:10.1016/0001-8791(85)90009-0
Su,R.(2012).The power of vocational interests and interests congruence in predicting career success
(Doctoral dissertation). University of Illinois at Urbana-Champaign, Urbana, IL.
Swanson, J. L., & Gore, P. A. Jr. (2000). Advances in vocational psychology theory and research. In S. D.
Brown & R. W. Lent (Eds.), Handbook of counseling psychology (3rd ed., pp. 233–269). Hoboken, NJ: John
Wiley & Sons.
Tinsley, H. E. A. (2000). The congruence myth: An analysis of the efficacy of the person–environment fit
model. Journal of Vocational Behavior,56, 147–179. doi:10.1006/jvbe.1999.1727
Tracey, T. J. G. (2002). Personal globe inventory: Measurement of the spherical model of interests and
competence beliefs. Journal of Vocational Behavior,60, 113–172. doi:10.1006/jvbe.2001.1817
Tracey, T. J. G. (2010). Development of an abbreviated personal globe inventory using item response theory:
The PGI-short. Journal of Vocational Behavior,76, 1–15. doi:10.1016/j.jvb.2009.06.007
Tracey, T. J. G., Allen, J., & Robbins, S. B. (2012). Moderation of the relation between person–environment
congruence and academic success: Environmental constraint, personal flexibility and method. Journal of
Vocational Behavior,80, 38–49. doi:10.1016/j.jvb.2011.03.005
Tranberg, M., Slane, S., & Ekeberg, S. E. (1993). The relation between interest congruence and satisfaction: A
meta-analysis. Journal of Vocational Behavior,42, 253–264. doi:10.1006/jvbe.1993.1018
Tsabari, O., Tziner, A., & Meir, E. I. (2005). Updated meta-analysis on the relationship between congruence
and satisfaction. Journal of Career Assessment,13, 216–232. doi:10.1177/1069072704273165
Zhang, Y., Kube, E., Wang, Y., & Tracey, T. J. G. (2013). Vocational interests in China: An evaluation of the
personal globe inventory-short. Journal of Vocational Behavior,83, 99–105. doi:10.1016/j.jvb.2013.03.009
Bai and Liao 17
