KURAM VE UYGULAMADA EĞİTİM BİLİMLERİ EDUCATIONAL SCIENCES: THEORY & PRACTICE

Received: April 11, 2016

Revision received: August 12, 2016 Copyright © 2017 EDAM
Accepted: December 22, 2016 www.estp.com.tr
OnlineFirst: February 15, 2017 DOI 10.12738/estp.2017.2.0264  April 2017  17(2)  529–548

Research Article

Korean Students’ Attitudes toward STEM Project-Based

Learning and Major Selection*
Sunyoung Han1
Sungkyunkwan University

Abstract
The trend of avoiding science, technology, engineering, and mathematics (STEM) majors has persisted
resulting in a lack of professionals in STEM fields. Further, the current STEM education system in Korea
does not meet domestic demands for STEM labor. To discover an educational approach encouraging students
to choose STEM majors at the post-secondary level, the current study employed a survey instrument
investigating students’ attitudes toward STEM project based learning (PBL), and examined the relationships
between these attitudes and intent to pursue a STEM major. A model depicting the relationships was
proposed, and the structure of the model was verified using structural equation modeling. Additionally, a
mediation relationship on the model was investigated and tested. Finally, the gender difference on the paths
of the structural model was analyzed using multiple-group analysis. The results indicated that students
who were positive toward PBL components (i.e., technology based learning, self-regulated learning, and
hands-on activities) except “collaborative learning” were more likely to have the intent to pursue a STEM
PBL. In addition, belief in STEM majors’ benefit played a role as a mediator. The findings of this study help
secondary teachers guide students through major and job selection by integrating PBL into mathematics
and science classrooms.

Keywords
Career choice • PBL • Structural equation modeling • Multiple-group analysis • Korean student

*
This paper was supported by Samsung Research Fund, Sungkyunkwan University, 2015.
1 Correspondence to: Sunyoung Han (PhD), 25-2 Sungkyunkwan-ro, Jongno-gu, Seoul, Korea. Email: [email protected]
Citation: Han, S. (2017). Korean students’ attitudes toward STEM project-based learning and major selection. Educational
Sciences: Theory & Practice, 17, 529–548. http://dx.doi.org/10.12738/estp.2017.2.0264
EDUCATIONAL SCIENCES: THEORY & PRACTICE

Korean society is currently facing challenges in science, technology, engineering,

and mathematics (STEM) fields. First, a trend toward avoiding STEM majors, which
indicates the majors relating to science, technology, engineering, and mathematics,
has persisted for the last 15 years; this has caused workforce deficits in STEM fields.
Due to the lack of STEM professionals, many small- and medium-sized businesses
requiring STEM knowledge and skills in Korea have been unable to fill open positions
(Shim, 2012). Another difficulty that Korean society has confronted is that promising
high school students have been more likely to select majors in medicine and medical
science rather than pure mathematics or science majors (Choi & Lee, 2008). As
STEM fields are pivotal for vitalizing the national economy, it is necessary to ensure
that high-quality students with great leadership potential select STEM professions.
At present, however, many students with an aptitude for STEM majors are choosing
to pursue majors in other fields. This is likely to cause severe problems not only in
the present but also in the future (Choi & Lee, 2008).

The current STEM education system in Korea does not meet domestic demands
for STEM labor. International test results (e.g., from the Programme for International
Student Assessment [PISA] and the Trends in International Mathematics and Science
Study [TIMSS]) have indicated that Korean students have less positive attitudes
toward learning mathematics and science compared with students in other countries,
although Korean students have ranked within the top five participating countries
in terms of academic achievement in these fields. Studies pointed out that Korean
students’ negative attitudes toward learning mathematics and science derive from the
traditional instructions commonly implemented in Korean classrooms and focusing
on an algorithmic approach, which indicates an instructional approach leading
students to simply memorize mathematics algorithms and not enough to make sense
of the background processes (Cho & Kim, 2013; Kim, 2012). With the decline of
students’ interest in mathematics and science and the increase in negative attitudes
toward learning mathematics and science, fewer students are choosing STEM majors
at the college level.

To encourage students to choose STEM majors at the post-secondary level, it is

important to establish learning environments that stimulate student interest in STEM
majors (Lent, Sheu, Gloster, & Wilkins, 2010; Wang, 2013). According to the social
cognitive career theory (SCCT; Bandura, 1986; Lent et al., 2010), students’ interests
influence their academic and career choices, as do self-efficacy beliefs, outcome
expectations, environmental supports and barriers, and choice actions. In other words,
learning environments that boost student interest may affect students’ academic
intentions and, in turn, what they finally decide regarding a major. Therefore, it is
critical for educators and policymakers to realize that major selection can be guided
by school-based educational efforts.

530
Han / Korean Students’ Attitudes toward STEM Project-Based Learning and Major Selection

Project-based learning (PBL) is one instructional approach that has been commonly
used in STEM classrooms. Studies have examined the positive impacts of PBL on
student interest in STEM disciplines (Baran & Maskan, 2010; Domínguez & Jaime,
2010; Johnson, Johnson, & Holubec, 1998; Kaldi, Filippatou, & Govaris, 2011; van
Rooij, 2009; Veenman, Kenter, & Post 2000). However, no studies have investigated
how PBL stimulated students’ interest in STEM disciplines, which in turn, influenced
their major selection. In addition, studies have analyzed neither which PBL factors
affect students in terms of academic choice nor how PBL factors work in relation to each
other. Therefore, the current study examines whether and how instructional strategies
utilizing PBL approaches affect students in their intent to pursue a STEM major.

Intent to Pursue a STEM Major

Because of the interest among policymakers and educators in increasing the
number of students pursuing STEM majors, factors contributing to STEM major
selection and retention have been investigated. Studies (e.g., Byars-Winston,
Estrada, Howard, Davis, & Zalapa, 2010; Maltese & Tai, 2011; Wang, 2013) on
students’ selection of and persistence in STEM majors have been based on SCCT
(Lent, Brown, & Hackett, 1994; Lent et al., 2010), which was developed based on
Bandura’s (1986) general social cognitive theory. The SCCT proposes that students’
self-efficacy and outcome expectations in studying mathematics and science are
the primary predictors of their intent to pursue a STEM major. Maltese and Tai
(2011) conducted regression analyses to assess the school-based factors relevant to
students’ persistence in completing STEM majors. Congruent with the SCCT, the
results indicated that STEM-related enrollment and persistence were influenced by
students’ interest in and belief regarding the usefulness of STEM majors (Maltese
& Tai, 2011). Wang (2013) examined how students’ educational experiences at the
secondary level played a critical role in fostering their interest and in developing
predispositions toward STEM majors; in particular, Wang found that the exposure
of students to mathematics and science courses was a stronger factor predicting the
intention to pursue STEM majors than was mathematics achievement at the high
school level. Byars-Winston et al. (2010) added ethnic factors and tested relationships
among social cognitive, cultural, and contextual factors in relation to STEM degree
selection and completion. The results indicated that ethnic variables were related to
self-efficacy, which was mediated through interest in STEM and indirectly influenced
students’ selection of a STEM major.

These studies (i.e., Byars-Winston et al., 2010; Maltese & Tai, 2011; Wang, 2013)
have indicated in common that students’ interest in STEM fields is the critical factor in
whether they major in STEM disciplines at the college level. In high school, students’
own interests stimulate them to enroll in STEM courses; by increasing their exposure

531
EDUCATIONAL SCIENCES: THEORY & PRACTICE

to STEM disciplines, this influences their intent to major in a STEM field. Innate
factors such as personal attribution and talent might affect their interest in choosing a
specific major. However, external factors, including the learning environment, have
also been indicated as important influences on students’ interests (Akalin & Sucuoglu,
2015; Sengül & Dereli, 2013). It is thus critical to note that students’ interest in STEM
disciplines can be increased through altering the educational environment, such as by
adopting educational activities or instructional strategies.

Students are also influenced in their intent to pursue STEM majors by outcome
expectations and self-referenced beliefs pertaining to STEM. Students who believed
STEM would be useful in the future have been found to be more likely to complete
and earn STEM degrees (Lent et al., 1994; Maltese & Tai, 2011). Therefore,
demonstrating the utility of STEM subjects is necessary for teachers to provide
students with information regarding the significance and role of STEM fields in their
future careers.

Whether demographic factors (e.g., ethnicity, gender, and socioeconomic status

[SES]) influence students’ choice of a STEM major has been debated. Maltese and
Tai (2011) found that race/ethnicity was not a statistically significant indicator of the
selection of a STEM major or persistence. On the other hand, a study conducted by
Wang (2013) revealed that students of different ethnicities varied in their intent to
pursue a STEM major. For example, ethnic minority students (i.e., African Americans,
Latino/as, Southeast Asians, and Native Americans) are negatively affected by school
and classroom climates including racial discrimination, which caused low self-
confidence in pursing STEM majors (Cabrera, Colbeck, & Terenzini, 2001).

Compared with other factors, gender has been found to exert an especially
pronounced effect in regard to the selection of a STEM major. In populations evenly
split in terms of gender, it has been shown that fewer women than men select STEM
majors (King, 2010; Prayor, Hurtado, DeAngelo, Blake, & Tran, 2009). As a result,
women are underrepresented in the STEM workforce (National Science Board, 2008).
Studies have indicated that formal and informal educational settings have led to a lack
of female STEM majoring students and professionals (Prayor et al., 2009; Shapiro
& Sax, 2011). The formal educational environment, encompassing curricula and
academic interactions with teachers and peers, influences women’s interests in STEM
and, in turn, whether they select a STEM major (Kinzie, 2007; Shapiro & Sax, 2011).
In addition, the informal educational setting increasing teachers and parents’ gender-
biased perception affected their female students/daughters in selecting their majors.

SES is also a critical factor influencing attitudes toward learning and academic
achievement among students. As one environmental factor that affects students,
economic status has often been measured based on the educational and income

532
Han / Korean Students’ Attitudes toward STEM Project-Based Learning and Major Selection

levels of students’ parents, students’ involvement in loan programs, and students’

eligibility for free or reduced-cost meals. The impact of students’ SES on their intent
to pursue a STEM major has varied among studies with different underlying contexts
(Maltese & Tai, 2011; Wang, 2013). According to the results from Malteses and Tai’s
(2011) study, SES measured through parental education level and participation in
loan programs and work-study was not statistically significant in predicting students’
college persistence. Moreover, Wang (2013) found that students’ college readiness in
mathematics and science was associated with their STEM major selection, and this
relationship was not significantly different across varied SES groups.

In contrast to other factors aforementioned, students’ demographic characteristics

cannot be altered through the adoption of an educational approach. However,
teachers, educators, and policymakers may be able to facilitate the development of
appropriate instructional strategies, enhance learning and teaching environments,
and promote students’ interest and motivation in and outcome expectations regarding
STEM majors. Among the diverse educational approaches that have been taken to
stimulate students’ interest, belief in the benefits of, and intentions regarding STEM
majors, the current study focused on PBL. In the following section, why the current
study illuminates PBL and how PBL components enhance students’ intent to pursue
a STEM major are elucidated.

Project-Based Learning for STEM

PBL is an instructional approach providing “the contextualized, authentic
experiences necessary for students to scaffold learning and build meaningfully
powerful science, technology, engineering, and mathematics concepts” (Capraro,
Capraro, & Morgan, 2013, p. 2). PBL has been discussed alongside and integrated
into STEM education because it can bridge secondary STEM courses with post-
secondary specializations and STEM professions (Capraro et al., 2013). In the real
world, STEM professionals are expected to solve diverse, ill-defined, and non-routine
problems that have multiple possible solutions. Similarly, PBL provides students
with ill-defined and authentic tasks and encourages them to experience problematic
situations in a school-based learning environment. Ultimately, PBL improves the
quality of education and enables students to become ideal STEM professionals with
21st century skills after graduating from secondary and post-secondary institutions
(Bell, 2010). For this reason, PBL has been introduced to teachers as an appropriate
instructional approach and has been implemented into STEM classrooms.

The impact of PBL implementation in STEM classrooms has been investigated. The
effects of PBL on students have been discussed within three domains: the affective,
behavioral, and cognitive domains. Regarding the affective domain, students’ interest,
self-confidence, and self-efficacy have been examined. STEM classrooms integrating

533
EDUCATIONAL SCIENCES: THEORY & PRACTICE

PBL were found to exert a positive influence on students’ interest in learning and their
conviction in the utility of STEM subjects (Baran & Maskan, 2010). Furthermore,
PBL activities increased student self-confidence and self-efficacy (Baran & Maskan,
2010). PBL is usually designed and implemented through collaborative group work;
hence, PBL has also been found to exert a positive impact on students’ communication
and collaboration skills (Domínguez & Jaime, 2010; Kaldi et al., 2011; van Rooij,
2009). Finally, the positive effects of PBL on students in the affective and behavioral
domains influenced their academic achievement (Han, Capraro, & Capraro, 2015).
As students’ interest in and positive attitudes toward learning STEM disciplines
increased, their test scores and overall academic achievement improved.

Studies (i.e., Capraro et al., 2013; Han & Carpenter, 2014) have investigated the
reasons behind the positive impact of PBL on students in the affective, behavioral, and
cognitive domains. PBL exerts a positive influence because of its unique constructs
and components of PBL. With Korean middle-school students as participants,
Han and Carpenter (2014) validated five constructs of STEM PBL: self-regulated
learning, collaborative learning environment, interdisciplinary learning environment,
technology-based learning, and hands-on activities. Capraro et al. (2013) theoretically
investigated components of PBL. It remains necessary to investigate how the
constructs and components of PBL account for its effect on students’ decisions
regarding STEM majors. In this sense, the current study analyzed data pertaining
to students’ attitudes toward the constructs of PBL and examined the relationships
among the constructs and the selection of a STEM major.

Case of Korea
The factors that influence Korean students selecting a post-secondary major have
been investigated (Kim & Oh, 2013; Lee, 2014; Yeo & Lee, 2013). Gender was found
to be a significant factor predicting students’ collegiate major selection, especially
among students selecting STEM fields (Kim & Oh, 2013). Men were more likely
to select engineering fields than were women. Outside of STEM fields, gender was
shown not to have a critical effect in determining college students’ major selection
(Yeo & Lee, 2013). Parental income and education level were shown to exert an
impact on Korean students’ selection of a major (Kim & Oh, 2013). Korean students
whose parents were of higher educational backgrounds tended toward the humanities
and social sciences rather than engineering. Finally, among Korean students, academic
achievement in high school was a decisive predictor of college major selection.
Korean students with higher levels of academic achievement in mathematics and
science were more likely to select STEM majors, whereas those who were good at
Korean and English were more inclined to choose majors in the humanities and social
sciences (Kim & Oh, 2013; Song, 2014). Among Korean students, interest in STEM

534
Han / Korean Students’ Attitudes toward STEM Project-Based Learning and Major Selection

courses has not been identified as a factor in predicting college major selection.
However, based on existing literature (i.e., Song, 2014), it was surmised that Korean
students’ interest might be a factor in their decision to pursue a STEM major.

The current study focused specifically on PBL and its impact on students’ intent
to pursue a STEM major because PBL has been around for many years in Korea and
has been examined as an effective instructional approach for the STEM disciplines.
Classes integrating PBL in Korea have exerted a positive impact on students’ self-
directed learning skills, interest in subjects, and attitudes toward learning itself (Cho
& Kim, 2010; Lee, 2014). Therefore, the present study sought to link the effect of
PBL on students’ interests to their intent to pursue a STEM major.

Theoretical Framework and Research Questions

The present study is based on Bandura’s (1986) general social cognitive theory, the
SCCT (Lent et al., 1944, 2000; Lent et al., 2010), and previous studies on students’
selection of a STEM major (Byars-Winston et al., 2010; Lent, Lopez, Lopez, & Sheu,
2008; Maltese & Tai, 2011; Sahin, 2013; Shapiro & Sax, 2011; Wang, 2013; Zhe,
Doverspike, Zhao, Lam, & Menzemer, 2010). The proposed model (see Figure 1)
focused on the relationship between students’ intent to pursue a STEM major and
PBL as an instructional model that can influence students’ interest in and outcome
expectations regarding the selection of a STEM major (i.e., belief in the benefits of
STEM fields). According to previous studies regarding the causal effects of PBL in
STEM classes on students’ interest and outcome expectations (i.e., Baran & Maskan,
2010; Domínguez & Jaime, 2010; Kaldi et al., 2011; van Rooij, 2009), students’
attitudes toward instructional approaches for STEM were found to be linked with their
belief in the benefits of pursuing STEM. The construct “belief in STEM benefits” was
postulated as a mediator based on the SCCT literature and, in the model, was linked
to students’ intention to pursue a STEM major.

  Belief in STEM
benefits (students’
outcome
expectations
regarding the
selection of a STEM
Learning
major)
environment, Students’ intent to
instructional model pursue a STEM
(PBL) major
   

Figure 1. Hypothesized model.

CO   BE  
535
EDUCATIONAL SCIENCES: THEORY & PRACTICE

Guided by the aforementioned studies and theoretical framework, the current study
examined the direct and indirect impacts of students’ attitudes toward STEM PBL on
their interest in majoring in STEM disciplines. Specifically, this study addressed the
following research questions:
Research Question 1: What are the relationships among the constructs concerning students’
attitudes toward STEM PBL and their intent to pursue STEM majors in post-secondary
education?
Research Question 2: Is there a mediator on the construct established in Question 1?
Research Question 3: How do the relationships described in Question 1 vary by gender?

Methods

Participants
The study participants were students enrolled at six secondary schools in Korea.
The schools and students were selected by the availability sampling approach.
Among the schools and teachers available to the author, six teachers at six schools
consented to administer the survey in their classrooms; surveys were administered by
the teachers rather than by the author. The schools were scattered throughout a large
city in Korea and the areas in which the schools were located were, on average, in the
middle for the city in terms of SES. A total of 840 students were surveyed; 24 students’
responses were excluded because they did not answer the items or intentionally gave
the same answers to all the items. Among the remaining participants (n = 816), 168
(20.6%) and 648 (79.4%) were middle school and high school students, respectively.
Female and male students accounted for 60.7% (n = 495) and 39.3% (n = 321) of the
sample, respectively.

Measures
The measurement instrument utilized was adopted from a previous study by Han
and Carpenter (2014) and revised for the current study. The survey instrument was
developed to investigate Korean students’ attitudes toward STEM PBL and was
verified in terms of reliability and validity. For reliability, Cronbach’s Alpha was
calculated. The validity of the instrument was established using exploratory factor
analysis (EFA) and confirmatory factor analysis (CFA). The survey inquiring into
students’ perspectives on STEM PBL as a teaching-learning method initially consisted
of 51 Likert-formatted items. Thirty-eight items were adopted from previous studies
regarding self-regulated learning (Kim, 2011), collaborative learning (Lim, Cha, &
Noh, 2001), beliefs in STEM benefits (Little & Hoel, 2011; Najafi, Ebrahimitabass,
Dehghani, & Rezaei, 2012), technology-based learning (Najafi et al., 2012), and
hands-on activities (Holstermann, Grube, & Bogeholz, 2010; Miller, 1990); 13 items

536
Han / Korean Students’ Attitudes toward STEM Project-Based Learning and Major Selection

were created by the author. A five-point scale ranging from “strongly agree,” “agree,”
“either agree or disagree,” “disagree,” and “strongly disagree” was used. Based on the
EFA and CFA analysis results, 26 items and 5 constructs (collaborative learning (CO),
technology-based learning (TE), self-regulated learning (SR), hands-on activities
(HO) and belief in STEM benefits (BE)) were extracted and confirmed. There were
five, five, six, five, and five items for CO, TE, SR, HO, and BE, respectively. These
five factors pertaining to students’ attitudes toward the STEM PBL teaching-learning
method were assumed to be potential predictors of students’ major selection. The
values of reliability of each construct were acceptable for five constructs, having
Cronbach’s Alphas of .838 (CO), .846 (TE), .796 (SR), .876 (HO), and .887 (BE).

Two further questions were posed to students. One inquired as to their gender and
the other asked whether they intended to pursue a STEM major in post-secondary
school. The responses to these two items were coded dichotomously (female = 1,
male = 0, and yes = 1, no = 0, respectively).

Data Analysis
Descriptive analysis. In the current study, composite variables were used in the
structural equation modeling analyses. Based on the CFA results, composite variables
were computed as mean values of the items involved in each construct (i.e., CO,
TE, SR, HO, and BE). For the composite variables, descriptive statistics including
means, standard deviations, skewness, and kurtosis were computed (Table 1).
Means, standard deviations, and correlation coefficients were useful in constructing
a proposed structure representing the relationships among the factors of the STEM
PBL teaching-learning method and the outcome variable of major selection. For the
dependent variable (i.e., STEM major selection), frequencies were reported because
the variable was binary-coded.
Table 1
Descriptive Statistics
Mean SD Skewness Kurtosis
Collaborative Learning (CO) 3.676 0.733 -0.622 1.217
Technology-Based Learning (TE) 3.579 0.681 -0.305 0.442
Self-Regulated Learning (SR) 3.404 0.647 -0.330 0.759
Hands-On Activities (HO) 3.554 0.714 -0.218 0.127
Belief in STEM Benefits (BE) 3.785 0.655 -0.480 1.353
Frequency STEM Major Non-STEM Major
STEM Major Selection 273 543

Structural equation modeling. Based on the theoretical framework, constructs

composing the measurement instrument were assumed and a CFA was performed for
these constructs. Subsequently, SEM analyses were conducted to test the proposed
conceptual model. The analyses were implemented using Mplus 7.0 (Muthén

537
EDUCATIONAL SCIENCES: THEORY & PRACTICE

& Muthén, 2010). For probit regression models, the weighted least square with
adjustment in mean and variance estimator was used because the outcome variable
was binary (Muthén & Muthén, 2010). Moreover, three fit indices (the chi-square
and the degree of freedom [χ2/df], root mean square error of approximate [RMSEA],
and comparative fit index [CFI]) were reported to analyze the overall model fit. The
cutoff values for RMSEA and CFI for good fit were .07 and .95, respectively (Byrne,
1998; Hu & Bentler, 1999; Kline, 2005). The chi-square test was expected to be
rejected in the event that the fit of the model was good.

Indirect effects. Among the five constructs of STEM PBL teaching-learning

method, “hands-on activities” was hypothesized to serve as a mediator, converting
the effects of students’ attitudes toward STEM PBL into the intent to major in STEM
disciplines. Therefore, direct and indirect paths from four constructs of the STEM
PBL teaching-learning method to the outcome variable were estimated and tested for
statistical significance.

Multiple-group analysis: Differences by gender. To answer the research question

“How do the relationships described in Question 1 vary by gender?” multiple-group
path analysis was employed (Kline, 2011). Three multiple-group path analyses were
conducted. The first multiple-group path analysis was utilized to test whether there
was an overall difference between male and female students. Next, a series of eight
separate multiple-group path analyses were conducted to identify those paths in which
male and female students differed significantly from one another. Finally, a third
multiple-group path analysis, which incorporated the results of these eight separate
analyses, was conducted. The first analysis only indicated the extent to which the
assumed baseline model was aligned to each male and female student group, but the
third analysis confirmed that there were statistically significant differences between
male and female students on some specified paths.

Results

CFA Analysis Results

CFA was conducted in regard to the employed measurement instrument; the CFA
results indicated that the measurement model fit the data well (RMSEA = 0.041, CFI
= 0.981, χ2/df = 189.243/80 [rejected]). The chi-squared test was rejected because
it is extremely sensitive to sample size and the sample size of the current study was
comparatively large (Brown, 2006; Marsh, Hau, Artelt, Baumert, & Peschar, 2006).
All of the standardized factor loadings were above 0.680 and statistically significant
(p < .05). The model fit indices and factor loadings indicated that the measurement
model had good convergent validity (Kline, 2011). Based on the CFA results, the
hypothesized measurement model, which was based on the correlation coefficients

538
Han / Korean Students’ Attitudes toward STEM Project-Based Learning and Major Selection

of the variables and theoretical framework, was tested using SEM in order to answer
the three research questions.

SEM Analysis Results

To test the hypothesized structural model, SEM analyses (e.g., path analyses)
were conducted. A recursive path analysis model was developed in light of previous
research results that disclosed the relationships among factors influencing students in
their choice of a major. A composite variables model was used in the current study
because using manifest variables was more likely to reflect the characteristics of the
sample rather than of the population. Table 2 presents both covariance and correlation
matrices that were used to determine whether the model fit the data.

The correlation coefficients of the relationships among the variables CO, TE,
SR, HO, and BE were statistically significant. The relationships between STEM
Major Selection and others were statistically significant with the exception of the
relationships between STEM Major Selection and CO and between STEM Major
Selection and SR. The hypothesized model was proven to fit the data (see Figure
2). All of the path coefficients were statistically significant with the exception of the
path from SR to STEM major selection. Moreover, the model fit indices showed that
the model had a good fit with the data (χ2/df = 3.420/2 [not rejected], CFI = 0.994).
Given that to be acceptable, CFI value must exceed 0.90, the current model was
found to fit the data well (Byrne, 1998; Hu & Bentler, 1999; Kline, 2005).
Table 2
Correlation Matrices among Composite Variables
1 2 3 4 5
1. CO —
2. TE 0.199** —
3. SR 0.360** 0.158** —
4. HO 0.337** 0.277** 0.322** —
5. BE 0.323 **
0.265** 0.453** 0.379** —
6. STEM Major Selection -0.044 0.141** 0.007 0.194** 0.138**

Note. CO=collaborative learning; TE=technology-based learning; SR=self-regulated learning;

HO=hands-on activities; BE=belief in STEM benefits; **p < .01; *p < .05.

539
   instructional model   pursue a STEM
(PBL) major
   
EDUCATIONAL SCIENCES: THEORY & PRACTICE

CO   BE  
CO   BE  
TE  

TE  
 
SR   STEM  Major  
 
Selection  
SR   STEM  Major  
Selection  
HO  

HO  

a. Initial model

CO   0.108**  
BE  
CO   0.108**  
BE   0.154**  
0.140 **  

TE   0.154**  
-­‐0.139**  
0.140**  
TE   0.327**  
0.106**  
-­‐0.139**    
0.106**   STEM  Major  
SR   0.327 **  

 
Selection  
0.196**   0.245**   STEM  Major  
SR  
Selection  
0.196**   0.245**  
HO  

HO   b. Final model
Figure 2. Initial and final models. CO=collaborative learning, TE=technology-based learning, SR=self-
regulated learning, HO=hands-on activities, BE=belief in STEM benefits.

The correlation matrix suggested that the constructs of CO and SR were unrelated
with students’ intent to select a STEM major. In an attempt at model trimming,
structural models excluding either CO or SR were tested. Compared with the initial
model, the model without SR had better model fit indices (χ2/df = 3.778/3 [not
rejected], RMSEA = 0.018, CFI = 0.997). Because the latter model was nested within
the initial model, the chi-squared value was utilized as an indicator of the goodness
of the model fit and for evaluating whether it was a better model (Brown, 2006;
Marsh et al., 2006). The latter model was consequently selected as the final model
because the chi-square test was not rejected (p = .550). In sum, the results from the
two SEM structural models supported the hypothesis that students’ attitudes toward
PBL components (i.e., CO, TE, SR, and HO) were related to their intent to pursue a
STEM major indirectly through the belief in STEM benefits.

540
Han / Korean Students’ Attitudes toward STEM Project-Based Learning and Major Selection

Mediation effects. In the models, the construct “belief in STEM benefits” played
a role as a mediator between the four other constructs and the outcome variable (i.e.,
intent to pursue a STEM major). Mediation was tested using the delta method in Mplus
7.0 (MacKinnon, 2008; Muthén & Muthén, 2010; see Table 3). First, the direct effects
(paths from the constructs, of collaborative learning, technology-based learning, and
hands-on activities to the outcome variable) were statistically significant, with path
loadings of -0.139, 0.106, and 0.245, respectively. The indirect effects of students’
attitudes toward collaborative learning, technology-based learning, and hands-on
activities on the intent to pursue a STEM major were also statistically significant,
though the estimated values were much smaller than those of direct effects. In
addition, the indirect effect from self-regulated learning to the selection of a STEM
major was statistically significant with an estimated value of 0.050.
Table 3
Direct and Indirect Effects
Effects Estimates
Effects from collaborative learning to the selection of a STEM major
Indirect: STEM ← Belief in STEM benefits ← Collaborative learning 0.017**
Direct: STEM ← Collaborative -0.139**
Effects from technology-based learning to the selection of a STEM major
Indirect: STEM ← Belief in STEM benefits ← Technology-based learning 0.022**
Direct: STEM ← Technology-based learning 0.106**
Effects from self-regulated learning to the selection of a STEM major
Indirect: STEM ← Belief in STEM benefits ← Self-regulated learning 0.050**
Effects from hands-on activities to the selection of a STEM major
Indirect: STEM ← Belief in STEM benefits ← Hands-on activities 0.030**
Direct: STEM ← Hands-on activities 0.245**
Note. **p < .05.

Multiple-Group Analysis: Differences by Gender

Based on the final SEM model, in which no path was constrained to be equal
across genders, the relationships among students’ attitudes toward STEM PBL
and major selection were investigated. The model fit indices for this first multiple-
group path analysis indicated a good model fit (χ2 = 0.527, df = 2; RMSEA < 0.001;
CFI = 1.000). To examine the paths on which male and female students showed
significant differences in terms of factor loadings, eight constrained path analyses
were employed. In each path analysis, one path was constrained. The result of each
constrained model was compared to that with all paths freed. The results of eight chi-
square tests indicated that male and female students differed significantly from one
another on the path leading from collaborative learning to belief in STEM benefits (p
< .05) and the path from hands-on activities to belief in STEM benefits (p < .05) (see
Table 4). The final multiple-group path model specifying these constrained paths was
compared to the model with all six paths constrained. On the final multiple group path

541
EDUCATIONAL SCIENCES: THEORY & PRACTICE

model, the path from hands-on activities to belief in STEM benefits was statistically
significant for both male (β = .308) and female (β = .141) students. In the case of the
path from collaborative learning to belief in STEM benefits, the path coefficient for
the male group was statistically significant (β = .224) whereas that for the female
group was not.
Table 4
Chi-Square Difference Test Results for Multi-Group Path Analysis
p-value
Constrained Path Chi-Square Value/df
(Chi-Square Test for Constrained Path)
CO→BE 12.994/3 p < .001**
TE→BE 0.693/3 p = .684
SR→BE 0.626/3 p = .753
HO→BE 14.885/3 p < .001**
CO→STEM major selection 1.787/3 p = .262
TE→STEM major selection 2.394/3 p = .172
SR→STEM major selection 1.319/3 p = .373
HO→STEM major selection 1.071/3 p = .461
Note. CO=collaborative learning; TE=technology-based learning; SR=self-regulated learning;
HO=hands-on activities; BE=belief in STEM benefits; **p < .01; *p < .05.

Discussion
The current study examined structural models to elicit the relationships among
students’ attitudes toward constructs of PBL and their intent to pursue a STEM
major. According to the findings of the study, the components of PBL have direct and
indirect effects on students’ intent to choose a STEM major. In the sense that PBL is
an appropriate and common pedagogical approach for STEM education in classrooms
from kindergarten through the 12th grade, the present study has educational implications
for research and practice regarding students’ selection of a STEM major.

The results of this study imply that it is possible that specific instructional
approaches influence students’ intentions regarding major selection. In the study,
students’ attitudes toward PBL were linked with their intent to select a STEM
major. The path coefficients from the PBL constructs (i.e., collaborative learning,
technology-based learning, and hands-on activities) to the intent to pursue a STEM
major indicated that students who felt positive about participating in PBL activities
tended to belief more in STEM benefits and tended to select STEM majors at the
college level. In other words, having classes integrated with PBL on offer might
encourage more students to choose a STEM major.

Regarding the factors that influence students’ intent to select or actual selection
of a STEM major, researchers have investigated and focused on students’ individual
demographic characteristics, such as gender, ethnicity, and SES, as well as

542
Han / Korean Students’ Attitudes toward STEM Project-Based Learning and Major Selection

environmental factors such as exposure to mathematics and science courses (King,

2010; Lent et al., 2010; Maltese & Tai, 2011; Prayor et al., 2009; Sahin, 2013;
Shapiro & Sax, 2011; Wang, 2013). However, no studies have examined the impacts
of instructional approaches on students’ intent to pursue a STEM major. Specifically,
it was necessary to scrutinize how PBL affects students’ intent to select a STEM
major because PBL has been illuminated as an effective instructional approach
to maximize students’ interest and academic achievement in STEM disciplines
(Capraro et al., 2013). For this reason, the findings of the current study are significant
in understanding the relationship between the instructional strategy implemented in
STEM courses and students’ intent to select a STEM major.

The present study provides additional evidence supporting why PBL needs to be
implemented in secondary level courses. Many studies (Baran & Maskan, 2010;
Dominguez & Jaime, 2010; Johnson et al., 1998; Kaldi et al., 2011; van Rooij, 2009;
Veenman et al., 2000) have illuminated the positive impacts of PBL on students’
attitudes and academic achievement and, accordingly, the necessity of implementing
PBL for K–12 students. PBL activities elicit more positive attitudes among students
regarding actual learning, improved communication and collaboration skills, and
higher levels of academic achievement in STEM disciplines. The current study
implicitly supported these findings of previous studies by connecting students’
attitudes toward PBL to their intent to pursue a STEM major. The link between PBL
and students’ intent to select a STEM major might be explained through general social
cognitive theory (Bandura, 1986) and the SCCT (Lent et al., 1994, 2000; Lent et al.,
2010): PBL encourages students’ interest in STEM majors, which in turn, stimulates
their intent to major in STEM fields. In addition, this study provides a reason as
to why exposure to mathematics and science courses increases the likelihood that
students will select a STEM major (Wang, 2013): in other words, the current study
explains how students should be exposed to mathematics and science courses. To
encourage more students to select a STEM major, it is necessary not only to increase
the duration students are exposed to mathematics and science but also to integrate
mathematics and science courses with an appropriate approach utilizing PBL.

Additionally, this study provides other educational implications that might be applied
to other instructional strategies including the components such as collaborative learning,
technology-based learning, and hands-on activities. In the study, five PBL constructs
were extracted and examined in terms of how they were related to the intent to pursue a
STEM major. Certain components of PBL—collaborative learning, technology-based
learning, and hands-on activities—had direct effects on students’ intention to pursue a
STEM major. In other words, any instructional approaches that include the components
of collaborative learning, technology-based learning, and hands-on activities might
exert similar impacts on students’ intent to select a STEM major.

543
EDUCATIONAL SCIENCES: THEORY & PRACTICE

The model proposed and validated in this study indicated that beliefs in STEM
benefits played a role as a mediator between other PBL components (i.e., collaborative
learning, technology-based learning, self-regulated learning, and hands-on activities)
and students’ intent to pursue a STEM major. This finding is in line with those of
Maltese and Tai (2011) and Byars-Winston et al. (2010) in the sense that students’
outcome expectations regarding STEM fields play a critical role in their choice and
pursuit of a STEM major. For the constructs of collaborative learning, technology-
based learning, and hands-on activities, there were direct and indirect effects and the
indirect effects partially explained the total effect. On the other hand, for the construct
of self-regulated learning, the direct effect on students’ intent to select a STEM major
was not statistically significant and there was only an indirect effect. That is, the
mediation effect of ‘belief in STEM benefits’ was complete between the variables,
SR and STEM major selection. For the other relationships between independent
variables except SR and STEM major selection, the mediation effects were partial.
This might be because the construct, SR was less likely to include STEM components
compared to the other constructs such as TE and HO.

The four indirect effects from each PBL construct to students’ intent to major in
a STEM field were all statistically significant. According to the previous studies,
the relationships between STEM PBL components and students’ beliefs in STEM
fields (Baran & Maskan, 2010) and between students’ beliefs in STEM fields and
career choice (SCCT; Lent et al., 1944, 2000; Lent et al., 2010) have been disclosed.
However, there were very limited studies linking these three constructs. For example,
a previous study by Wang (2013) found that students’ belief in the benefits of STEM
had a direct effect on STEM major selection and did not find that relationship was
mediated. The current study contributes to finding that the relationship between
students’ belief in the benefits of STEM and STEM major selection was statistically
significant. This implies that promoting students’ belief in the benefits of STEM fields
is critical to increase the synergistic effect of PBL activities on students’ positive
attitudes toward majoring in STEM fields. To have more students choose STEM
majors, it is important to provide not only the learning environment where students
are encouraged in STEM PBL activities, but also opportunities to increase students’
beliefs in STEM benefits in schools and mathematics classrooms.

Differences by gender were found in analyzing the model describing the

relationships between students’ attitudes toward PBL components and their intention
to pursue a STEM major. Differences between female and male students were
identified on the paths from collaborative learning to belief in STEM benefits and
from hands-on activities to belief in STEM benefits. For male students, only the
direct effects of these two PBL components were statistically significant, whereas for
female students, the indirect effects were significant. This might explain why male

544
Han / Korean Students’ Attitudes toward STEM Project-Based Learning and Major Selection

students are more likely to select a STEM major than are female students (King, 2010;
Prayor et al., 2009). For example, it was found that male students with a positive
attitude toward technology-based learning were more likely to select a STEM major,
whereas female students were influenced indirectly by technology mediated by the
belief in STEM benefits. In the case of the path from hands-on activities to belief in
STEM benefit, the estimated path coefficient for male students was larger than that
for female students. This might be explained by the fact that male students were more
likely to have positive attitudes toward engineering components, including diverse
hands-on activities (Felder, Felder, Mauney, Hamrin, & Dietz, 1995).

This study contributes to the understanding how PBL should be implemented in

mathematics and science classrooms to encourage students to pursue STEM majors
at the secondary level. The current study acknowledges that students’ major selection
might be affected by instructional approaches used in schools and not only by innate
individual factors such as ethnicity, gender, and parental income and education.
The use of instructional approaches encouraging students’ interest in and belief in
the benefits of majoring in STEM enhance the possibility that such students will
desire to select STEM majors in college, contributing to the achievement of equity in
educational opportunity.

Limitations of the Study

One limitation of the study is related to the use of students’ self-reported measures:
it could be contended that students who reported that they intended to pursue a
STEM major could in fact have chosen a different major in the future. However, as
Maltese and Tai (2011) stated, “the major students have in mind when learning high
school is a strong predictor of their eventual degree field” (p. 899), students’ intent
to pursue a STEM major at the secondary level is a strong and significant predictor
of their actual later major selection in STEM fields. In this sense, the reliability of
the findings, based as they were on self-reported measures, might be guaranteed in
the current research context. However, it is strongly suggested that future studies use
self-reported measures and observed measures regarding the selection of a STEM
to enable findings from both measures might be compared. In addition, the findings
from the current study might not be generalizable to students in other countries
because the participants in this study were all Korean students. Future studies could
include students of different nationalities and following different curricula in order to
determine how nationality may yield different results.

545
EDUCATIONAL SCIENCES: THEORY & PRACTICE

References
Akalin, S., & Sucuoglu, B. (2015). Effects of classroom management intervention based on
teacher training and performance feedback on outcomes of teacher-student dyads in inclusive
classrooms. Educational Science: Theory & Practice, 15, 739–758.
Bandura, A. (1986). Social foundations of thought and action: A social cognitive theory. Englewood
Cliffs, NJ: Prentice Hall.
Baran, M., & Maskan, A. (2010). The effect of project-based learning on pre-service physics
teachers electrostatic achievements. Cypriot Journal of Educational Sciences, 5(4), 243–257.
Bell, S. (2010). Project-based learning for the 21st century: Skills for the future. The Clearing
House, 83(2), 39–43.
Brown, T. A. (2006). Confirmatory factor analysis for applied research. New York, NY: Guilford Press.
Byars-Winston, A. M., Estrada, Y. E., Howard, C. H., Davis, D., & Zalapa, J. (2010). Influence of
social cognitive and ethnic variables on academic goals of underrepresented students in science
and engineering. Journal of Counseling Psychology, 57(2), 205–218.
Byrne, B. M. (1998). Structural equation modeling with LISREL, PRELIS and SIMPLIS: Basic
concepts, applications and programming. Mahwah, New Jersey: Lawrence Erlbaum Associates.
Cabrera, A., Colbeck, C., & Terenzini, P. (2001). Developing performance indicators for assessing
classroom teaching practices and student learning: The case of engineering. Research in Higher
Education, 42, 327–352.
Capraro, R. M., Capraro, M. M., & Morgan, J. (2013). STEM project –based learning: An integrated
science, technology, engineering, and mathematics (STEM) approach (2nd ed.). Rotterdam,
Netherlands: Sense.
Cho, J., & Kim, J. (2010). Collaborative project curriculum applying project-based learning.
Journal of the Korea Academia-Industrial Cooperation Society, 11(1), 214–219.
Cho, H., & Kim, D. (2013). The relationship between cooperation learning in mathematics and their
attitudes towards mathematics. Korean Comparative Education Society, 23(4), 131–154.
Choi, N., & Lee, J. (2008). The comparative analysis on the rate of return to post-secondary
education between engineering and medical school graduates. The Journal of Educational
Administration, 26(1), 369–389.
Domínguez, C., & Jaime, A. (2010). Database design learning: A project-based approach organized
through a course management system. Computers & Education, 55(3), 1312–1320.
Felder, R. M., Felder, G. N., Mauney, M., Hamrin, C. E., & Dietz, E. J. (1995). A longitudinal study
of engineering student performance and retention. III. Gender differences in student performance
and attitudes. Journal of Engineering Education-Washington, 84, 151–164.
Han, S., & Carpenter, D. (2014). Construct validation of student attitude toward science, technology,
engineering, and mathematics project based learning: The case of Korean middle grade students.
Middle Grade Research Journal, 9(3), 27–41.
Han, S., Capraro, R., & Capraro, M. (2015). How science, technology, engineering, and mathematics
(STEM) project-based learning (PBL) affects high, middle, and low achievers differently: The impact
of student factors on achievement. International Journal of Science and Mathematics Education,
13(5), 1089–1113.
Holstermann, N., Grube, D., & Bogeholz, S. (2010). Hands-on activities and their influence on
students’ interest. Research in Science Education, 24, 743–757.

546
Han / Korean Students’ Attitudes toward STEM Project-Based Learning and Major Selection

Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis:
conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1–55.
Johnson, D. W., Johnson, R. T., & Holubec, E. (1998). Cooperation in the classroom. Edina, MN:
Interaction Book Company.
Kaldi, S., Filippatou, D., & Govaris, C. (2011). Project-based learning in primary schools: Effects
on pupils’ learning and attitudes. Education 3–13, 39(1), 35–47.
Kim, B. (2012). Instrument development and analysis of secondary students’ mathematical beliefs.
Journal of Educational Research in Mathematics, 22(2), 229–259.
Kim, E. J. (2011). 문화예술교육 프로그램에서 음악 테크놀로지 기반 수업 [The educational effects
of music technology-based learning] (Published doctoral dissertation). Kung-Hee University,
Seoul.
Kim, K., & Oh, B. (2013). College major selection: Gender and socioeconomic discrimination
among Korean matriculates. Korean Journal of Sociology of Education, 23(4), 59–84.
King, J. E. (2010). Gender equity in higher education: 2010. Washington, DC: American Council
on Education.
Kinzie, J. (2007). Women’s paths in science: A critical feminist analysis. New Directions for
Institutional Research, 133, 81–93.
Kline, R. B. (2005). Principles and practice of structural equation modeling (2nd ed.). New York,
NY: Guilford Press.
Kline, R. B. (2011). Principles and practice of structural equation modeling (3rd ed.). New York,
NY: Guilford Press.
Lee, J-C. (2014). The effects of thesis-writing activity based project method on the ability of
girl’s middle school student’s self directed learning and learning attitude. Journal of the Korea
Academic-Industrial Cooperation Society, 15(3), 1458–1464.
Lent, R. W., Brown, S. D., & Hackett, G. (1994). Toward a unifying social cognitive theory of
career and academic interest, choice and performance. Journal of Vocational Behavior, 45(1),
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., 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(1), 52–62.
Lent, R. W., Sheu, H., Gloster, C. S., & Wilkings, G. (2010). Longitudinal test of the social cognitive
model of choice in engineering students at historically Black universities. Journal of Vocational
Behavior, 76(3), 387–394.
Lim, H., Cha, J., & Noh, T. (2001). The relationships of verbal behaviors with learning variables
in cooperative learning environments, and middle school students’ perceptions of cooperative
learning. Journal of the Korean Association for Research in Science Education, 21(3), 487–496.
Little, A., & Hoel, A. (2011). Interdisciplinary team teaching: An effective method to transform
student attitudes. The Journal of Effective Teaching, 11(1), 36–44.
MacKinnon, D. P. (2008). Mediation analysis. In R. L. Cautin, & S. O. Lilienfeld (Eds.), The
Encyclopedia of clinical psychology (pp. 1–9). New York, NY: Wiley.

547
EDUCATIONAL SCIENCES: THEORY & PRACTICE

Maltese, A. V., & Tai, R. H. (2011). Pipeline persistence: Examining the association of educational
experiences with earned degrees in STEM among U.S. students. Science Education, 95, 877–907.
Marsh, H. W., Hau, K., Artelt, C., Baumert, J., & Peschar, J. L. (2006). OECD’s brief self-
report measure of educational psychology’s most useful affective constructs: Cross-cultural,
psychometric comparisons across 25 countries. International Journal of Testing, 6(4), 311–360.
http://dx.doi.org/10.1207/s15327574ijt0604_1
Miller, A. S. (1990). At-risk sixth-grade students’ attitude toward science, achievement in science,
goal orientation, and cognitive engagement in science (Doctoral dissertation). The University
of North Carolina, Greensboro.
Muthén, L. K., & Muthén, B. O. (2010). MPLUS user’s guide (6th ed.). Los Angeles, CA: Muthén
& Muthén.
Najafi, M., Ebrahimitabass, E., Dehghani, A., & Rezaei, M. (2012). Students’ attitude towards
science and technology. Interdisciplinary Journal of Contemporary Research in Business, 3(10),
129–134.
National Science Board. (2008). Science and engineering indicators 2008. Arlington, VA: National
Science Foundation.
Pryor, J. H., Hurtado, S., DeAngelo, L., Blake, L. P., & Tran, S. (2009). The American freshman:
National norms fall 2009. Los Angeles, CA: Higher Education Research Institute, UCLA.
Sahin, A. (2013). STEM clubs and science fair competitions: Effects on post-secondary
matriculation. Journal of STEM Education, 14(1), 7–13.
Sengül, S., & Dereli, M. (2013). The Effect of Learning Integers Using Cartoons on 7th Grade
Students’ Attitude to Mathematics. Educational Sciences: Theory and Practice, 13, 2526–2534.
Shapiro, C. A., & Sax, L. J. (2011). Major selection and persistence for women in STEM. New
Directions for Institutional Research, 2011(152), 5–18.
Shim, J. (2012). The avoidance of science and engineering majors for the last 15 years. The
disaster emerges [이공계기피 15년, 재앙이 시작됐다]. Retrieved from http://blog.joins.com/
media/folderlistslide.asp?uid=sjw33&folder=2&list_id=12906020
Song, Y-J. (2014). The effect of major selection motivation, major satisfaction, and professor-
student interaction of university students on their school life adjustment (Unpublished doctoral
dissertation). Dong-A University, South Korea.
Van Rooij, S. W. (2009). Scaffolding project-based learning with the project management body of
knowledge (PMBOK®). Computers & Education, 52(1), 210–219.
Veenman, S., Kenter, B., & Post, K. (2000). Cooperative learning in Dutch primary classrooms.
Educational Studies, 26, 281–302
Wang, X. (2013). Why students choose STEM majors: Motivation, high school learning, and
postsecondary context of support. American Educational Research Journal, 50(5), 1081–1121.
Yeo, W-S., & Lee, Y-J. (2013). The effects of the major selection factors and satisfaction on
according to the characteristics of college. Korea Journal of Tourism and Hospitality Research,
27(6), 441–455.
Zhe, J., Doverspike, D., Zhao, J., Lam, P., & Menzemer, C. (2010). High school bridge program: A
multidisciplinary STEM research program. Journal of STEM Education, 11(1&2), 61–68.

548