Master’s Study Duration: The Eects of Active Learning
Based on the Belief-Action-Outcome Model*
Jon-Chao Hong Jian-Hong Ye
Department of Industrial Education and Institute for
Research Excellence in Learning Sciences,
National Taiwan Normal University
Faculty of Education,
Beijing Normal University
Yu-Feng Wu Zhen He
Oce of Physical Education,
Ming Chi University of Technology
Faculty of Education,
Beijing Normal University
Influenced by Confucianism, the social role is postponed from school to work in Taiwan, most young adults enter the job
market after completing a higher education degree. However, in recent years, delayed graduation by postgraduate students has
become a problem. To understand this phenomenon, this study recruited a mix of participants who had already graduated and
participants who were about to graduate (individuals who had completed their courses and thesis). The aim of the study was to
explore (1) how individuals’ academic self-ecacy aects their active learning strategies and academic self-condence and (2)
how this is reected in the duration of their studies. A total of 245 valid questionnaires were collected, comprising the responses
of 91 men and 154 women. Among the participants, 34.3% graduated on time, whereas 51% did not graduate on time because
of incomplete theses. A conrmatory factor analysis approach was adopted in this study. The results demonstrated that academic
self-ecacy was positively related to active learning strategies (higher-order, integrative, and reective strategies) and active
learning strategies were positively related to academic self-condence, whereas academic self-condence was negatively related
to an extended duration for completing a master’s degree.
Keywords: academic self-efficacy, academic self-confidence, active learning strategies, belief-
action-outcome model, study duration
＊ 1. Corresponding Author: Jian-Hong Ye, firstname.lastname@example.org
2. This work was financially supported by the “Institute for Research Excellence in Learning
Sciences” of National Taiwan Normal University (NTNU) from The Featured Areas Research
Center Program within the framework of the Higher Education Sprout Project by the Ministry of
Education (MOE) in Taiwan.
Bulletin of Educational Psychology, 2022, 53(4), 879–900
National Taiwan Normal University, Taipei, Taiwan, R. O. C.
The relatively low graduation rate of higher education in traditional practice has received widespread
attention (Roksa, 2010). For example, the average completion time for a master�s degree in Uganda is 3
years and 8 months, significantly higher than the required 2-year period (Wamala & Oonyu, 2012). In
addition, most Belgian research students receive a master’s degree after 5 years of study (Dupont et al.,
2013). Many students hesitate to enter the job market as early as their parents did (Hirschi, 2018). As a
result, universities around the world are working to shorten the study duration to prevent high dropout rates
(Geven et al., 2018). However, according to the from-school-to-work model, postponing graduation means
deferring taking up one’s social role (Danziger & Ratner, 2010). In Western societies, adulthood tends to be
dened by a social role that involves a balance among independence, interdependence, responsibility, and
productivity (Smith et al., 2016).
In contrast, under Confucianism, if people are educated, they will eventually have a better social role,
so continuing to study without stopping for any reason is embedded in most Taiwanese people’s thinking.
So as not to have to get a job, tertiary education has become very common in Taiwan. Many young people
try dierent ways to continue their studies and thus postpone taking up their social role of making money to
help support their families (Smith et al., 2016).
Although in recent years the status of graduation postponement of master’s study has not been
announced by the Statistical Department of Ministry of Education, Taiwan, The Chief Executive Oce of
the Executive Yuan in Taiwan pointed out that in 2011 there were 619 doctoral students and 4,482 master’s
students who had to extend their studies. In 2009, the numbers were 495 and 2,607 for doctoral and master’s
degree students, respectively. When comparing these two years, the delayed graduation rate increased by
25.1% for doctoral students and 71.9% for master’s students (Directorate-General of Budget, Accounting
and Statistics, Executive Yuan, 2012). However, what causes them to delay is not clear and needs to be
studied. According to the dominance of Confucianism, stopping their study to earn money to help support
their families is not the main reason why those Taiwanese students delay their graduation (Agu, 2014).
Some research has indicated that psychological characteristics are critical to academic learning and success
(Galla et al., 2014); thus, this study explored some relevant factors from the perspectives of educational
The concept of active learning was highlighted by Bonwell and Eison (1991) who dened it as learning
activities that engage students and encourage them to think deeply about what they are doing (Hyun et al.,
2017). Active learning is a key aspect of educational psychology, given that learners perform cognitive
processes when they face a learning task, and they also refer to motivational matters that force the student
to control their own learning process (Martin-Lobo et al., 2018). Active learning promotes students’ higher
order learning with critical thinking, and involves them in integrating knowledge; with these skills, they are
more willing to work through challenging material (Adkins, 2018). However, dierent learning strategies
tend to emerge in environments which change signicantly over the period of individual lifetimes (Bullinaria,
2018). In this sense, how active learning aects those individuals’ postponement of their master’s study is
the focus of this study.
This study draws on the belief-action-outcome (BAO) theory, which can eectively explain individuals’
behaviors and nal outcomes (Melville, 2010), and how people’s beliefs inuence their subsequent actions,
which will in turn have a signicant impact on their behavioral outcomes. For the purpose of this study,
BAO was adapted to explore the correlates between academic self-ecacy, active learning strategies, and
academic self-confidence to understand how those antecedents relate to the delay of graduation. It was
expected that the results of this study could help readers explore the role of academic self-efficacy and
understand the signicance of active learning strategies adapted by those students who defer completion of
their graduate studies.
Master’s Study Duration 881
In educational psychology, academic self-ecacy has been identied as an important factor inuencing
the academic performance of college students. Academic self-efficacy was defined by Odaci (2011) as
the belief students have in their ability to handle academic tasks. Without it, students may experience
procrastination or inability to learn (Jung et al., 2017). Related research shows that academic self-ecacy
has a predictive eect on successful learning in universities (Gore, 2006). Students with high self-ecacy
can enhance their achievements in the eld (Schöber et al., 2018). In addition, it is well known that low
levels of academic self-ecacy also prevent students from conducting research and learning with a strong
desire (Love et al., 2007). Based on this, academic self-ecacy is seen as an important regulator of student
learning behavior to complete their studies (Meluso et al., 2012). Thus, how academic self-ecacy aects
those post-graduate students in working on their master’s degree was explored in this study.
Active Learning Strategies
There are many factors influencing learning, one of which is active learning, the learning strategy
considered to explain one of the successful learning factors (Poondej & Lerdpornkulrat, 2016). These
learning factors can be a great help to graduate students since they are required to write a dissertation.
Dissertations are the most enduring, autonomous, and creative research task students must undertake, and
require the adoption of high-level and in-depth strategies (Dupont et al., 2013). The active learning strategy
is the approach whereby students make eorts to have some practical experience in their study (Warburton,
2003). Active learning can be taken to mean how students practice deep inquiry into the meaning of
learning, focusing on integration and reection methods (Vos et al., 2011). Previous research has pointed
out that the completion of a master’s degree and active learning approaches are highly related (Drennan,
2010). Thus, this study explored the learners’ strategical actions for active learning when they conducted
academic study. Moreover, how active learning strategies (i.e., critical analysis, reection, and integration)
can aect master’s students’ learning in relation to how long it takes them to obtain a degree was the topic
of interest of this study.
A psychological factor related to learning eectiveness is self-condence, and a major feature of self-
confidence is clear personal beliefs (White, 2009). Self-confidence is a subjective post-experience that
stems from the individual’s deterministic judgment of self-expression and which prompts the individual
to immediately respond to current performance perceptions (Jiang & Kleitman, 2015). For example, self-
condence is considered to predict the future performance of individuals (Hong et al., 2017). Academic
self-confidence is defined as an individual’s self-confidence perception of his/her own academic ability
(Laird, 2005) (e.g., condence in solving academic writing or research questions). In the eld of education,
academic self-condence is used to predict academic success or failure (Nokelainen et al., 2007). Because
academic self-condence is an important facet in the context of learner-centered learning (Maclellan, 2014),
this may imply the eect of academic self-condence on the duration of graduate study, and thus needs
to be explored. Therefore, this study explored learners' perceptions of self-confidence in their academic
abilities when conducting academic research.
The study duration is the time it takes for learners to obtain their degree. Most countries in the world
have a time limit for degree studies (Almgren, 2014). In the context of the growing number of graduate
students, universities around the world are working to shorten the duration of their studies at institutes
(Geven et al., 2018). Ho et al. (2020) considered study duration as a form of academic performance, and
study duration can be considered as the overdue performance of the participants in the study. Looking at
the example of Finland, the average time to complete a master's degree is between 6 and 6.5 years, instead
of the expected 5 years (Haarala-Muhonen et al., 2017), whereas most students in Belgium need 5 years to
earn a master's degree (Dupont et al., 2013). Given this situation, understanding students’ attitudes toward
postponing their master's degree graduation is an important indicator of whether a student will experience
a delay in graduation. In addition, the duration of obtaining a master’s degree refers to the total time
within which students are required to complete their master’s study, and is the topic explored in this paper.
Therefore, this study examined study duration, meaning the time that students spend to complete their
The control-value theory predicts that students will invest more eort if they perceive activities to be
valuable and if they anticipate success (Cooper et al., 2017). Academic self-ecacy is a kind of expectation
belief that is closely related to the construction of successful expectations (Bergey et al., 2018). At the same
time, research has shown that self-ecacy has a major impact on the use of in-depth processing strategies
(Berger & Karabenick, 2011). In the eld of educational psychology, there is a strong relationship between
self-condence and academic achievement (Stankov et al., 2012). However, given the dissimilarities in the
bases of self-condence and self-ecacy, it is essential to explore both constructs to fully understand the
relationship between these structures (Munroe-Chandler et al., 2008). In line with this, the term “condence”
is used, and the dierence between the two constructs lies in the point in time, with self-ecacy being the
construct before (hypothetical) cognitive behavior, and self-condence after cognitive behavior (Morony
et al., 2013). Therefore, since self-condence and self-ecacy are dierent concepts and are measured in
dierent ways, including both is an important component of delivering a more comprehensive examination
of dierent levels of self-condence-related constructs (Munroe-Chandler et al., 2008) and can provide a
more complete understanding of the impact of academic self-condence on completion of graduate studies.
This study included both as important components to better understanding the impact of academic self-
condence on completion of graduate school.
However, the interaction between those factors has not been studied in relation to the postponement of
the graduation of master’s students; thus, the purpose of this study was to answer the following research
questions: (1) Is academic self-ecacy related to the three active learning strategies? (2) Are those three
active learning strategies related to academic self-condence? and (3) Is academic self-condence related to
the length of delay of graduation?
Research Model and Hypotheses
Melville�s (2010) belief-action-outcome (BAO) theory can effectively explain individuals� behaviors
and nal outcomes. In other words, the BAO framework explains how beliefs facilitate the execution of
actions and how actions further influence outcomes (Molla et al., 2014). The study by Ho et al. (2020)
Master’s Study Duration 883
found that the model constructed through BAO is useful for understanding learning performance issues in
higher education. Thus, BAO can help explain how people�s cognitive beliefs inuence their actual actions
(behaviors) and, ultimately, their outcomes or performance in educational settings. In this study, academic
self-ecacy was considered as a belief, active learning strategies as an action, and academic self-condence
and study duration as outcomes. Accordingly, this study proposes a theoretical model for linking academic
self-ecacy with three types of active learning (higher order learning, integrative learning, and reective
learning), academic self-condence, and study duration. Thus, seven research hypotheses were proposed to
construct the research model found in Figure 1 below.
Academic Self-ecacy Related to Three Types of Active Learning Strategies
Relevant research has indicated that active learning methods can improve students’ participation in
study by improving their analytical and conceptual thinking skills (Hall et al., 2004). Active learning has
a positive impact on performance because of higher education. To focus on knowledge synthesis, active
learning is required (Chotitham et al., 2014), and there is a correlation between self-ecacy and learning
strategies (Wang et al., 2008). Self-ecacy and productivity also have a positive relationship (Phillips &
Russell, 1994). Having a strong sense of self-ecacy allows learners to adhere to specic strategies (Anam
& Stracke, 2016). Research suggests that online self-ecacy has a predictive eect on active learning, and
that there is a signicant positive correlation between them (Zhang, 2015). In addition, studies have shown
a signicant positive correlation between self-ecacy and active learning (Liem et al., 2008). Therefore,
based on belief-action in BAO, this study proposed the hypotheses for exploring how academic self-ecacy
relates to active learning strategies:
H1: Academic self-ecacy is positively related to higher order learning.
H2: Academic self-ecacy is positively related to integrative learning.
H3: Academic self-ecacy is positively related to reective learning.
Three Types of Active Learning Strategies Related to Academic Self-condence
Self-condence is a judgment of one’s ability and quality. Condent people rely on their ability to face
any event they encounter (Heinström, 2010). A change in self-condence aects motivation and predicts the
quality of performance (Georion et al., 2013). A specic form of condence is academic self-condence,
which refers to a person’s condence in their academic performance (Laird, 2005). Strong academic self-
confidence comes from effective learning strategies (Valiee et al., 2016) as using more post-cognitive
strategies can improve learners’ self-confidence (Kisac & Budak, 2014). Therefore, based on action-
outcome in BAO, how active learning strategies relate to academic self-confidence is hypothesized as
H4: Higher order learning is positively related to academic self-condence.
H5: Integrative learning is positively related to academic self-condence.
H6: Reective learning is positively related to academic self-condence.
Academic Self-condence is Positively Related to Study Duration
Condence usually increases the motivation of a person’s abilities (Bénabou & Tirole, 2002), and it
is believed that increasing or decreasing self-confidence will improve or reduce performance (Hanton
& Connaughton, 2002). One study suggested that psychological characteristics are critical to academic
learning and success (Galla et al., 2014). Bearden et al. (2001) argued that self-confidence is related to
decisions and behavior. Self-condent people will assign a higher value to success and a lower value to
failure (Chaouali et al., 2017). For example, academic self-confidence can be an important predictor of
problems with study orientation and academic performance (van der Aar et al., 2019). In addition, excellent
performance in competitive sports is also associated with good self-confidence (Comeig et al., 2016).
High self-condence, known specically as athletic condence, is considered to be the key psychological
characteristic required by elite athletes in order to promote optimal performance (McGinn et al., 2018),
and self-condence is also considered to be the best predictor of academic subjects such as mathematics
and English (Stankov et al., 2012). Therefore, based on the condence-behavior model (Keller, 2009), how
academic self-condence and study duration are related is hypothesized as follows:
H7: Academic self-condence is positively related to study duration.
The questionnaires were targeted at full-time postgraduate students studying in Taiwan. An online
questionnaire (Google Forms) was developed and posted on some graduate student social networks, and
Facebook was used to inform the participants in those social networks. Adopting a snowball sampling
approach, we asked the questionnaire receivers to share the link of the questionnaire website with other
graduate students they were acquainted with. Questionnaires were collected from June 16 to September 30,
2018, until there were 350 returns.
Master’s Study Duration 885
The participants in this study were all master’s degree students enrolled in graduate study in that
semester. A total of 245 participants’ useful questionnaires were returned, of which 91 (37.1%) were from
males and 154 (62.9%) from females. Regarding the studying institutes, 194 (79.2%) participants were
from public universities, and 51 (20.8%) were from private universities. As for the types of university,
153 students were from general universities (62.4%) and 92 were from technological universities (37.6%).
Regarding their majors, 173 (70.3%) were majoring in social science and management, while 72 (29.7%)
were from natural science, engineering, and related majors. Among the participants, 84 (34.3%) graduated
on time; for those who did not, the reasons were: incompletion of thesis for 125 (51%), incompletion of
credits for 11 (4.5%), taking a teacher’s education program for nine (3.7%), and personal factors for 12
The questionnaire items were adapted from previous studies, and were translated into Chinese. Three
domain experts were invited to check the accuracy, and then the questionnaire was sent to 20 graduate
students to identify any unsuitable items. In line with this forward and backward process, statements were
corrected until face validity was ensured. The questionnaire adopted a 5-point Likert scale (1[strongly
disagree] to 5 [strongly agree]) for participants to self-rate themselves so as to obtain the reliability and
validity of the questionnaire items and constructs.
Academic self-efficacy measurement: Self-efficacy is an individual’s belief in their own ability to
eectively accomplish a task (Bandura, 1977), so according to this denition and reference, Hong et al.’s
(2015) questionnaire for measuring students’ self-ecacy in learning was adapted for use in this study. The
term academic self-ecacy is used in this study to refer to the postgraduate students’ belief in their ability
to complete their research (e.g., “In the course of my research, I can nd out the details of the problem
and have a solution to it” and “When faced with problems in the study process, I will develop multiple
Active learning strategies measurement: Active learning is a key strategy for students to extract meaning
and understanding from course materials and experience (Warburton, 2003). This study was based on the
Deep Strategy Learning Scale from Entwistle and McCune (2004). The active learning referred to in this
study refers to the use of strategies for applying active learning. Exemplary items include: First, “In the
process of solving a learning problem, I will classify and compare the problems before I take action” and “I
always look for dierent ways before making assumptions when solving problems.” Second, “I nd myself
thinking about the commonalities of different course content” and “I try to integrate ideas, information
or experiences into new and more complex explanations and relationships.” Finally, “I can reflect on
something new from mistakes and change the way I understand problems or concepts” and “I will reect on
my own views on a topic or issue, and nd out the advantages and disadvantages.”
Academic self-confidence measurement: Self-confidence is a judgment of one�s ability, quality, and
self-confidence that affects motivation and predicts the quality of performance (Geoffrion et al., 2013).
In this study, self-confidence enhancements are based on the above definitions and revisions, where the
term academic self-confidence as used in this study refers to the students� self-confidence perception of
conducting academic research. For example, “I am very self-learning oriented and have the condence to
do research” and “I am good at planning research design, so I am condent in my career development in
Items analysis was analyzed with rst-order conrmatory factor analysis (CFA) (Hong et al., 2020),
where items with a factor loading (FL) not larger than the value of .50 were deleted; the higher residual
values in each construct were also canceled, until they met the threshold suggestions of Hair et al. (2019) (see
Table 1). Then, the number of items was reduced from eight to six for academic self-ecacy, from seven to
four for higher order learning, from seven to ve for integrative learning, from eight to four for reective
learning, and from seven to ve for academic self-condence, as shown in Table 1.
Index Threshold Academic
χ2--- 28.1 2 31.1 .9 3
df --- 9 2 9 2 5
χ2/df < 5 3.12 1 3.46 .45 .60
RMSEA < .10 .09 .02 .09 .01 .01
GFI > .80 .96 .99 .96 .99 .99
AGFI > .80 .914 .98 .90 .99 .90
Reliability and Validity Analysis
In this study, the value of Cronbach's α ranged from .84 to .88 and the value of CR ranged from .83 to
.88, indicating that the internal consistency and composite reliability (CR), and average variance extracted
(AVE) ranged from .50 to .57 and the Factor Loadings (FL) ranged from .71 to .76, indicating that the
convergence validity of this construct was acceptable, as shown in Table 2.
Reliability and Validity Analysis
Constructs M SD αCR AV E FL
Threshold -- -- > .70 > .70 > .50 > .50
Academic self-ecacy 3.61 .57 .88 .88 .51 .74
High level learning 3.73 .58 .84 .84 .57 .75
Integrative learning 3.72 .55 .86 .86 .54 .74
Reective learning 3.85 .55 .84 .84 .57 .76
Academic self-condence 3.68 .52 .86 .83 .50 .71
Note. M = mean; SD = standard deviation; FL = factor loading; CR = composite reliability; AVE = average variance extracted.
Study Duration Analysis
The rst paragraph of Article 26 of Taiwan’s University Act (2015) stipulates that the term of study for
a master’s degree is 1 to 4 years. However, at present, most schools in Taiwan have dispersed the master's
degree courses over four semester-long classes and the analysis in this study showed that the average
Master’s Study Duration 887
duration of the course is 5.22 semesters, which means that the master’s students only obtained their degree
on average in the second semester of their 3rd year, as shown in Table 3. The number of participants who
studied for 4 semesters or less was 84 (34.3%), 5 semesters was 55 (22.4%), 6 semesters was 79 (32.2%), 7
semesters was 22 (9%), and 8 semesters or more was 5 (2%), as shown in Table 4.
Study Duration Analysis (semester)
Construct Min Max M SD Med.
Study duration 4 8 5.22 1.08 5
Descriptive Analysis of Study Duration
Study duration Frequency (percentage)
4 semesters (or less)
8 semesters (or more)
Model Goodness of Fit Test
A good model adaptation represents the validity of the model construction. Relevant scholars have
recommended a value of χ2/df to be less than 5 (Hair et al., 2010), RMSEA should be less than .1, GFI,
AGFI, NFI, NNFI, IFI, and RFI should be greater than .80, while RMR should be greater than 0.05 (Abedi
et al., 2015). PNFI and PGFI should be greater than .50 (Abedi et al., 2015). In this Model goodness of
t test, χ2 = 520.25, df = 269, χ2/df = 1.93, RMSEA = .06, RMR = .06, GFI = .86, AGFI = .83, NFI = .85,
NNFI = .91, IFI = .92, RFI = .83, PNFI = .76, and PGFI = .71. The values of the tting indicators in this
study are in line with scholars' recommendations.
In structural equation modeling (SEM), path analysis is an extension of multivariate regression because
it includes numerous multivariate regression models or calculations that are assessed at the same time.
This delivers a more eective and direct way to model mediations, indirect eects, and other multifarious
relations between variables (Lei & Wu, 2007). Although multi-group validation factor analysis can help
assess and validate measurement invariance to ensure that the instrument is used fairly and is not biased
towards certain groups of key questions (Ullman & Bentler, 2012), this study was based on verifying the
eects of the BAO framework on graduate study duration, so no comparison between dierent models was
The SEM verification results showed that academic self-efficacy has a significant impact on higher
order learning (β = .45***, t = 5.75); academic self-ecacy has a signicant impact on integrative learning
(β = .41***, t = 5.34); academic self-ecacy has a signicant impact on reective learning (β = .534***,
t = 6.74); higher order learning has a signicant impact on academic self-condence (β = .18 **, t = 2.68);
integrative learning has a signicant impact on academic self-condence (β = .44 ***, t = 6.02); reective
learning has a signicant impact on academic self-condence (β = .55***, t = 6.95); and nally, academic
self-condence has a signicant impact on study duration (β = -.63***, t = -10.48). Academic self-ecacy
has an explanatory power of 20.2% for higher order learning, 17% for integrative learning, and 29% for
reective learning. Academic self-condence has an explanatory power of 70%, with an explanatory power
of 40% for study duration, as shown in Figure 2.
Research Model Verication
(t = 5.75)
(t = 5.33)
(t = 6.74)
(t = -10.48)
(t = 2.68)
(t = 6.02)
(t = 6.95)
R2 = .20
R2 = .17
R2 = .29
R2 = .70 R2 = .40
** p < .01. *** p < .001.
Indirect Eects Analysis
The results of this study veried that the 95% condence interval for neither the direct nor the indirect
effects crossed zero (Nakagawa & Cuthill, 2007). In terms of the indirect effect, academic self-efficacy
is indirectly positively related to academic self-condence (β = .55**), and academic self-ecacy has an
indirect negative relationship to study duration (β = -.35**); higher order learning and study duration have
an indirect negative relationship (β = -.11*); and integration learning has an indirect negative relationship to
study duration (β = -.27**). Reection learning has an indirect negative relationship to study duration (β =
-.34**), as shown in Table 5.
Indirect Eect Analysis
Constructs Academic self-ecacy Higher order learning Integrative learning Reective learning
β 95% CI β 95% CI β 95% CI β 95% CI
Self-condence .55** [.43, .65]
Study duration -.35** [-.43, -.26] -.11* [-.23, -.03] -.27** [-.37, -.19] -.344** [-.44, -.24]
* p < .05. ** p < .01. *** p < .001.
Master’s Study Duration 889
The results of the t test revealed signicant dierences in the constructs of reective learning, academic
self-condence, and study duration between the participants of the two subject areas. Among them, students
majoring in natural science and engineering elds showed a greater reective learning and academic self-
condence when compared to students majoring in social science and management. In addition, students
majoring in natural science and engineering showed a lower study duration when compared to social
science and management students, as shown in Table 6.
Constructs Group N M t Compare d
Academic self-ecacy 1173 3.57 -1.83 .25
2 72 3.71
Higher order learning 1173 3.69 -1.56 .21
2 72 3.82
Integrative learning 1173 3.68 -1.67 .24
2 72 3.81
Reective learning 1173 3.79 -2.39* 2 > 1 .35
2 72 3.98
Academic self-condence 1173 3.62 -2.68** 2 > 1 .37
2 72 3.81
Study duration 1173 5.35 2.89** 1 > 2 .40
2 72 4.92
Note. 1. Social science and management, 2. Natural science and engineering.
Honicke and Broadbent (2016) pointed out that academic self-efficacy is a self-recognition or an
individual’s belief in his/her ability to perform at a specied level. Even in the face of academic challenges,
it is not easy to compromise, and the participants in this study have good academic self-efficacy (M =
3.72, SD = 4.24). Scouller (1998) indicated that students are likely to adopt active learning methods when
preparing term papers. In this study, participants’ perceptions of using active learning strategies were higher
order learning (M = 3.61, SD = .57), integrative learning (M = 3.27, SD = .58), and reective learning (M
= 3.85, SD = .55). Galla et al. (2014) showed that psychological characteristics are critical to academic
learning and success. The participants in this study were found to have academic self-condence (M = 3.68,
SD = .52).
Academic Self-ecacy was Positively Related to Active Learning
Anam and Stracke (2016) stated that having a strong sense of self-ecacy allows learners to adhere
to specic strategies, while Wang et al. (2008) showed that there is a correlation between self-ecacy and
learning strategies in distance education-related research. In addition, Anam and Stracke stated that students
with higher English efficacy and self-regulating learning effectiveness reported learning strategies more
frequently than students without higher significance. In an educational study, self-efficacy is a positive
predictor of the relationship between active learning methods (Liem et al., 2008). At the same time, Zhang
(2015) found that online self-ecacy has a predictive eect on active learning, and also found a signicant
positive correlation between the two. The results of this study show that academic self-ecacy does have a
positive correlation with active learning, echoing the literature above.
Active Learning Strategies were Positively related to Academic Self-conﬁdence
Valiee et al. (2016) found that good self-condence comes from eective learning strategies. Smith et
al. (2018) stated that the specic impact of active learning techniques on condence and test scores will
certainly provide more insight, but due to institutional constraints, the approach is complicated. Kisac
and Budak (2014) used more post-cognitive strategies to be more successful in improving learners’ self-
condence. In addition, Gordon and Debus (2002) stated that the promotion of the use of in-depth methods
has brought varying degrees of success. The results of this study show that the strategy of using active
learning is positively related to academic self-condence.
Academic Self-conﬁdence was Negatively Related to Study Duration
There is ample evidence that self-condence aects performance, while positive self-condence can
improve performance (Compte & Postlewaite, 2004). Other studies have pointed out that if students lack
self-condence, their performance will certainly be hindered (Saini, 2016). Many studies have conrmed
that self-condence strongly inuences learning and performance achievements in various elds. In order to
achieve excellent athletic performance, for example, it is usually thought that athletes are required to have a
great amount of self-condence (Vealey & Chase, 2008). As another example, Ganley and Lubienski (2016)
conrmed the potential reciprocal relationship between mathematics achievement and self-condence. Ku
et al. (2014) indicated that many students lack condence in learning mathematics, which may lead them to
abandon the pursuit of more mathematics knowledge, which will directly and indirectly aect their learning
performance and achievements. Students’ self-condence will also have a solid eect on their high school
achievements (Tavani & Losh, 2003). The study duration was also considered as the overdue performance
of the participants in accordance with the denition proposed by Ho et al. (2020). Based on the condence-
behavior model (Keller, 2009), the results of this study show that academic self-confidence was indeed
negatively related to study duration.
The Dierence in Academic Field
Human’s life learning is a difficult trip over observation and involvement of the world. However,
the procedure of learning in engineering and natural sciences is a route which needs much observational
involvement and physical experiments (Montáns et al., 2019). Natural science and engineering jobs involve
much interaction with data and things (Ballesteros et al., 2021). On the other hand, social science and
humanities jobs involve much interaction with data and people (Shu et al., 1996). Based on the dierence in
interacting with people, data, and things, there are dierences in relation to students’ studying eld (Smeby,
Social inuence is dened as how a person in a social network is aected by the behavior of others
to adapt to group behavior patterns (Venkatesh & Brown, 2001). Social inuence lies in making explicit
the interactivism of the action-based conception and proposing an account of the fundamental aspects of
social identity (Mirski & Bickhard, 2021). A “we perspective” is essential both for the daily coordination
of human actions and in assisting to constitute people’s collective social identity and actions (Goddard &
Wierzbicka, 2021). The frame of coordinating social interaction in the community encultures individuals to
their mental-state attribution to the culture of that community (Mirski & Bickhard, 2021). In line with this,
Master’s Study Duration 891
individuals in dierent frames of communities may have dierent “we perspectives” on graduation.
The present study also validates the arguments of the above study. The present study compared the
duration of graduate study between subject areas, and found that students enrolled in graduate school in
the management and social sciences took longer to complete their degrees than those in natural science and
engineering. The reason may be due to the fact that students majoring in management and social science
bridge and bond human or culture-related learning materials, so they claim that they have little time to
learn so much knowledge, which makes them have less confidence in their ability to meet occupation
requirements (Yang, 2018). Moreover, the results of this study revealed that the constructs of reflective
learning strategies and academic self-condence were signicantly higher for students in natural science
and engineering than for those in social science and management; this result is supported by a study of
Horton et al. (2021) which indicates that the incorporation of reective journaling for students can be seen
as a mechanism for promoting self-condence to overcome challenges.
Current research on higher education tends to focus on performance in particular subject areas. Based on
this, the purpose of this study was to explore the relevant factors aecting the duration of master’s courses,
and to use the BAO theory to verify a theoretical model through structural equation modeling. The results
showed that academic self-ecacy and active learning (high level, integration, and reection) are positively
related; active learning (high level, integration, and reection) was also positively related with academic
self-condence, while academic self-condence was negatively related with study duration. Active learning
was also indirectly negatively related with study duration. In addition, although the three types of active
learning were most strongly related with academic self-condence and study duration, reective learning
was most strongly related with these two constructs, integrative learning was second, and higher order
learning was last.
Although in this study SEM helped us to understand the factors contributing to the study delay in study
duration, this analysis is a confirmatory rather than exploratory statistical technique and also has some
interpretative limitations. Specifically, the validation results of this study are suitable for explaining the
effect of active learning strategies on delayed graduation. Other intrinsic or extrinsic factors that affect
graduate students’ study duration were not combined in the model of this study to jointly explain the eects
of delayed years of study.
In addition, this study found that the average duration of a master’s degree program in Taiwan was more
than 5.22 semesters, and on average, students did not receive their master’s degree until their third year,
indicating that many master’s students tried to postpone assumption of their social roles. This phenomenon
is a problem that is worthy of attention, and action should be taken to ameliorate the situation.
Each discipline has its unique cultural characteristics, but in the research and policy formulation of
higher education, the assessment of departmental dierences is often neglected (Becher, 1994). There is
a special phenomenon in Taiwan’s higher education system. Under the same rules of practice, graduate
students in the humanities and social sciences generally spend more semesters than graduates in the natural
sciences. In addition, because of the differences that exist in different disciplines, this also means that
dierent disciplines have dierent approaches or emphases in the mentoring of professors. In addition to
the level of competence of the active learning strategies of the graduate students, the mentoring process
should be taken into account. At the same time, it is also necessary to design a more appropriate supervision
mechanism by focusing on the soft dimension degree thesis to be able to explain and understand their
contributions and the hard dimension degree thesis to be able to prove the experimental results, so that the
students can reach the requirements of the graduate school more quickly and shorten the study duration.
For the multiple levels of mediation analysis, the two requirements were tested according to Baron
and Kenny (1986): (a) there should be a predictor as the level of independent variable to aect the level
of mediator; and (b) after entering the mediator, the relationship between the level of predictor and the
dependent variable will come to an end. The results of this study revealed that three types of active learning
approaches play the role of mediator between academic self-ecacy and the study duration, indicating that
those approaches should be considered timely by those students who cannot complete their master’s degree
in the expected number of semesters.
The result indicated that reective learning was highest and was negatively related to postponement of
schooling duration; accordingly, the school administration system and faculties may guide those students
with less ability to reect on their learning, or provide them with counseling by academic advisors to help
them develop better approaches.
Additionally, the results of the study confirm that academic self-efficacy positively affects active
learning strategies, so when instructors identify students in the low self-ecacy group, they should rst
help them develop a good sense of self-ecacy to help them cope with academic activities.
In addition, in the past, the BAO model was generally used to explain people�s use of information
technology, and is a theoretical model that effectively explains phenomena from the macro level to the
micro level. BAO has been less applied in the eld of education in the past. However, in this study, it was
reconrmed that BAO can be used to explain the relationship between people’s beliefs about education-
related behaviors and subsequent outcomes in specific educational contexts, as well as to expand the
understanding of learning or educational behaviors by developing cognitive-behavior-based research
models for educational scholars.
Limitations and Future Study
Supervising is the mainstay of improving the learning environment for students throughout higher
education. A number of studies have shown that supervision is one of the most important factors in
promoting graduate students’ successful academic and postgraduate academic experience (Beaudin et al.,
2016). In a follow-up study, if the number of weekly supervision discussions or supervision styles can be
studied and analyzed, then more factors aecting the duration of taking courses at the postgraduate level
can be analyzed.
In social psychology, it is increasingly recognized that social influences play an important role in
students' academic performance (Fan, 2011), such as group culture among students, which may also be a
factor in delaying completion of studies. However, this component was not explored in this study. Therefore,
in future research, the inuence of peer group culture and learning atmosphere on study completion can be
Abedi, G., Rostami, F., & Nadi, A. (2015). Analyzing the dimensions of the quality of life in hepatitis B
patientsusing conrmatory factor analysis. Global Journal of Health Science, 7(7), 22–31. https://
Adkins, J. K. (2018). Active learning and formative assessment in a user-centered design course.
Information Systems Education Journal, 16(4), 34–40.
Master’s Study Duration 893
Agu, N. (2014). Variables attributed to delay in thesis completion by postgraduate students. Journal of
Emerging Trends in Educational Research and Policy Studies, 5, 435–443.
Almgren, K. (2014). Using business intelligence to mitigate graduation delay issues. International Journal
of Applied Science and Technology, 4(3), 94–100.
Anam, S., & Stracke, E. (2016). Language learning strategies of Indonesian primary school students: In
relation to self-ecacy beliefs. System, 60, 1–10. https://doi.org/10.1016/j.system.2016.05.001
Ballesteros, M. Á., Sánchez, J. S., Ratkovich, N., Cruz, J. C., & Reyes, L. H. (2021). Modernizing the
chemical engineering curriculum via a student-centered framework that promotes technical,
professional, and technology expertise skills: The case of unit operations. Education for Chemical
Engineers, 35, 8–21. https://doi.org/10.1016/j.ece.2020.12.004
Bandura, A. (1977). Self-ecacy: Toward a unifying theory of behavioral change. Psychological Review,
84, 191–215. https://doi.org/10.1037/0033-295X.84.2.191
Baron, R. B., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological
research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social
Psychology, 51, 1173–1182. https://doi.org/10.1037/0022-35188.8.131.523
Bearden, W. O., Hardesty, D. M., & Rose, R. L. (2001). Consumer self-condence: Renements in
conceptualization and measurement. Journal of Consumer Research, 28, 121–134. https://doi.
Beaudin, A., Emami, E., Palumbo, M., & Tran, S. D. (2016). Quality of supervision: postgraduate dental
research trainees’ perspectives. European Journal of Dental Education, 20, 32–38. https://doi.
Becher, T. (1994). The signicance of disciplinary dierences. Studies in Higher education, 19, 151–161.
Bénabou, R., & Tirole, J. (2002). Self-condence and personal motivation. The Quarterly Journal of
Economics, 117, 871–915. https://doi.org/10.1162/003355302760193913
Berger, J. L., & Karabenick, S. A. (2011). Motivation and students’ use of learning strategies: Evidence of
unidirectional eects in mathematics classrooms. Learning and Instruction, 21, 416–428. https://doi.
Bergey, B. W., Parrila, R. K., & Deacon, S. H. (2018). Understanding the academic motivations of students
with a history of reading diculty: An expectancy-value-cost approach. Learning and Individual
Dierences, 67, 41–52. https://doi.org/10.1016/j.lindif.2018.06.008
Bonwell, C., & Eison, J. (1991). Active learning: Creating excitement in the classroom. ASHE-ERIC higher
education report (ED336049). ERIC. https://eric.ed.gov/?id=ED336049
Bullinaria, J. A. (2018). Evolution of learning strategies in changing environments. Cognitive Systems
Research, 52, 429–449. https://doi.org/10.1016/j.cogsys.2018.07.024
Chaouali, W., Souiden, N., & Ladhari, R. (2017). Explaining adoption of mobile banking with the theory
of trying, general self-condence, and cynicism. Journal of Retailing and Consumer Services, 35,
Chotitham, S., Wongwanich, S., & Wiratchai, N. (2014). Deep learning and its eects on achievement.
Procedia-Social and Behavioral Sciences, 116, 3313–3316. https://doi.org/10.1016/
Comeig, I., Grau-Grau, A., Jaramillo-Gutiérrez, A., & Ramírez, F. (2016). Gender, self-condence,
sports, and preferences for competition. Journal of Business Research, 69, 1418–1422. https://doi.
Compte, O., & Postlewaite, A. (2004). Condence-enhanced performance. American Economic Review, 94,
Cooper, K. M., Ashley, M., & Brownell, S. E. (2017). Using expectancy value theory as a framework to
reduce student resistance to active learning: A proof of concept. Journal of Microbiology & Biology
Education, 18(2), 1–8. https://doi.org/10.1128/jmbe.v18i2.1289
Danziger, S., & Ratner, D. (2010). Labor market outcomes and the transition to adulthood. The Future of
Children, 20(1), 133–158.
Directorate-General of Budget, Accounting and Statistics, Executive Yuan. (2012). National status statistics
bulletin (No. 179). https://www.dgbas.gov.tw/public/Data/291716311471.pdf
Drennan, J. (2010). Critical thinking as an outcome of a master’s degree in nursing programme. Journal of
Advanced Nursing, 66, 422–431. https://doi.org/10.1111/j.1365-2648.2009.05170.x
Dupont, S., Meert, G., Galand, B., & Nils, F. (2013). Postponement in the completion of the nal
dissertation: An underexplored dimension of achievement in higher education. European Journal of
Psychology of Education, 28, 619–639. https://doi.org/10.1007/s10212-012-0132-7
Entwistle, N., & McCune, V. (2004). The conceptual bases of study strategy inventories. Educational
Psychology Review, 16, 325–345. https://doi.org/10.1007/s10648-004-0003-0
Fan, W. (2011). Social inuences, school motivation and gender dierences: An application of the
expectancy-value theory. Educational Psychology, 31, 157–175. https://doi.org/10.1080/01443410.2
Galla, B. M., Wood, J. J., Tsukayama, E., Har, K., Chiu, A. W., & Langer, D. A. (2014). A longitudinal
multilevel model analysis of the within-person and between-person eect of eortful engagement
and academic self-ecacy on academic performance. Journal of School Psychology, 52, 295–308.
Ganley, C. M., & Lubienski, S. T. (2016). Mathematics condence, interest, and performance: Examining
gender patterns and reciprocal relations. Learning and Individual Dierences, 47, 182–193. https://
Georion, R., Lee, T., & Singer, J. (2013). Validating a self-condence scale for surgical trainees. Journal of
Obstetrics and Gynaecology Canada, 35, 355–361. https://doi.org/10.1016/S1701-2163(15)30964-6
Geven, K., Skopek, J., & Triventi, M. (2018). How to increase PhD completion rates? An impact evaluation
of two reforms in a selective graduate school, 1976-2012. Research in Higher Education, 59, 529–
Master’s Study Duration 895
Goddard, C., & Wierzbicka, A. (2021). “We”: Conceptual semantics, linguistic typology and social
cognition. Language Sciences, 83, Article 101327. https://doi.org/10.1016/j.langsci.2020.101327
Gordon, C., & Debus, R. (2002). Developing deep learning approaches and personal teaching ecacy
within a preservice teacher education context. British Journal of Educational Psychology, 72,
Gore, P. A. (2006). Academic self-ecacy as a predictor of college outcomes: Two incremental validity
studies. Journal of Career Assessment, 14, 92–115. https://doi.org/10.1177/1069072705281367
Haarala-Muhonen, A., Ruohoniemi, M., Parpala, A., Komulainen, E., & Lindblom-Ylänne, S. (2017). How
do the dierent study proles of rst-year students predict their study success, study progress and
the completion of degrees? Higher Education, 74, 949–962. https://doi.org/10.1007/s10734-016-
Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2010). Multivariate data analysis (7th ed.).
Hair, J. F., Black, W. C., Babin, B. J., & Anderson, R. E. (2019). Multivariate data analysis (8th ed.).
Hall, M., Ramsay, A., & Raven, J. (2004). Changing the learning environment to promote deep learning
approaches in rst-year accounting students. Accounting Education, 13, 489–505. https://doi.org/10.
Hanton, S., & Connaughton, D. (2002). Perceived control of anxiety and its relationship to self-condence
and performance. Research Quarterly for Exercise and Sport, 73, 87–97. https://doi.org/10.1080/027
Heinström, J. (2010). From fear to ow: Personality and information interaction. Chandos Publishing.
Hirschi, A. (2018). The fourth industrial revolution: Issues and implications for career research and practice.
The Career Development Quarterly, 66, 192–204. https://doi.org/10.1002/cdq.12142
Ho, Y.-J., Ye, J.-H., Wang, C.-M., & Ye, J.-N. (2020). Personalities and overcondence predict
academic procrastination and frequency of seeking research advice that inuence graduate
study. Journal of Research in Education Sciences, 65(4), 139–170. https://doi.org/10.6209/
Hong, J.-C., Chen, M.-L., Ye, J.-H., & Chen, P.-H. (2020). The relationship between adolescents' creative
self-ecacy, competition engagements and the value of participating in competition-making
the IEYI international exhibition for young inventors as an example. Bulletin of Educational
Psychology, 51, 457–481. https://doi.org/10.6251/BEP.202003_51(3).0006
Hong, J.-C., Hwang, M.-Y., Tai, K.-H., & Lin, P.-C. (2015). Self-ecacy relevant to competitive anxiety
and gameplay interest in the one-on-one competition setting. Educational Technology Research and
Development, 63, 791–807. https://doi.org/10.1007/s11423-015-9389-2
Hong, J.-C., Hwang, M.-Y., Tai, K.-H., & Tsai, C.-R. (2017). An exploration of students’ science learning
interest related to their cognitive anxiety, cognitive load, self-condence and learning progress using
inquiry-based learning with an iPad. Research in Science Education, 47, 1193–1212. https://doi.
Honicke, T., & Broadbent, J. (2016). The inuence of academic self-ecacy on academic performance:
A systematic review. Educational Research Review, 17, 63–84. https://doi.org/10.1016/
Horton, A. G., Gibson, K. B., & Curington, A. M. (2021). Exploring reective journaling as a learning
tool: An interdisciplinary approach. Archives of Psychiatric Nursing, 35, 195–199. https://doi.
Hyun, J., Ediger, R., & Lee, D. (2017). Students' satisfaction on their learning process in active learning and
traditional classrooms. International Journal of Teaching and Learning in Higher Education, 29,
Jiang, Y., & Kleitman, S. (2015). Metacognition and motivation: Links between condence, self-protection
and self-enhancement. Learning and Individual Dierences, 37, 222–230. https://doi.org/10.1016/
Jung, K.-R., Zhou, A. Q., & Lee, R. M. (2017). Self-ecacy, self-discipline and academic performance:
Testing a context-specic mediation model. Learning and Individual Dierences, 60, 33–39. https://
Keller, J. M. (2009). Motivational design for learning and performance: The ARCS model approach.
Springer Science & Business Media. https://doi.org/10.1007/978-1-4419-1250-3
Kisac, I., & Budak, Y. (2014). Metacognitive strategies of the university students with respect to their
perceived self-condence levels about learning. Procedia-Social and Behavioral Sciences, 116,
Ku, O., Chen, S. Y., Wu, D. H., Lao, A. C. C., & Chan, T.-W. (2014). The eects of game-based learning on
mathematical condence and performance: High ability vs. low ability. Educational Technology &
Society, 17(3), 65–78.
Laird, T. F. N. (2005). College students’ experiences with diversity and their eects on academic self-
condence, social agency, and disposition toward critical thinking. Research in Higher Education,
46, 365–387. https://doi.org/10.1007/s11162-005-2966-1
Lei, P.-W., & Wu, Q. (2007). Introduction to structural equation modeling: Issues and practical
considerations. Educational Measurement: Issues and Practice, 26(3), 33–43. https://doi.
Liem, A. D., Lau, S., & Nie, Y. (2008). The role of self-ecacy, task value, and achievement goals
in predicting learning strategies, task disengagement, peer relationship, and achievement
outcome. Contemporary Educational Psychology, 33, 486–512. https://doi.org/10.1016/
Love, K. M., Bahner, A. D., Jones, L. N., & Nilsson, J. E. (2007). An investigation of early research
experience and research self-ecacy. Professional Psychology: Research and Practice, 38, 314–
Maclellan, E. (2014). How might teachers enable learner self-condence? A review study. Educational
Master’s Study Duration 897
Review, 66, 59–74. https://doi.org/10.1080/00131911.2013.768601
Martín-Lobo, P., Martínez-Álvarez, I., Muelas, Á., Pradas, S., Magreñán, A. (2018). A study of 16 years old
student learning strategies from a neuropsychological perspective: An intervention proposal. Trends
in Neuroscience and Education, 11, 1–8. https://doi.org/10.1016/j.tine.2018.03.001
McGinn, S., Alcock, D., & Cameron, L. J. (2018). Straight from the horse’s mouth: Understanding
professional event riders’ mental preparation for maximising self-condence prior to competition
using thematic analysis. Comparative Exercise Physiology, 14, 261–270. https://doi.org/10.3920/
Meluso, A., Zheng, M., Spires, H. A., & Lester, J. (2012). Enhancing 5th graders’ science content
knowledge and self-ecacy through game-based learning. Computers & Education, 59, 497–504.
Melville, N. P. (2010). Information systems innovation for environmental sustainability. MIS Quarterly, 34,
Mirski, R., & Bickhard, M. H. (2021). Conventional minds: An interactivist perspective on social cognition
and its enculturation. New Ideas in Psychology, 62, Article 100856. https://doi.org/10.1016/
Molla, A., Abareshi, A., & Cooper, V. (2014). Green IT beliefs and pro-environmental IT practices among
IT professionals. Information Technology & People, 27, 129–154. https://doi.org/10.1108/ITP-10-
Montáns, F. J., Chinesta, F., Gómez-Bombarelli, R., & Kutz, J. N. (2019). Data-driven modeling and
learning in science and engineering. Comptes Rendus Mécanique, 347, 845–855. https://doi.
Morony, S., Kleitman, S., Lee, Y. P., & Stankov, L. (2013). Predicting achievement: Condence vs self-
ecacy, anxiety, and self-concept in Confucian and European countries. International Journal of
Educational Research, 58, 79–96. https://doi.org/10.1016/j.ijer.2012.11.002
Munroe-Chandler, K., Hall, C., & Fishburne, G. (2008). Playing with condence: The relationship between
imagery use and self-condence and self-ecacy in youth soccer players. Journal of Sports
Sciences, 26, 1539–1546. https://doi.org/10.1080/02640410802315419
Nakagawa, S., & Cuthill, I. C. (2007). Eect size, condence interval and statistical signicance: A
practical guide for biologists. Biological Reviews, 82, 591–605. https://doi.org/10.1111/j.1469-
Nokelainen, P., Tirri, K., & Merenti-Välimäki, H. (2007). Investigating the inuence of attribution
styles on the development of mathematical talent. Gifted Child Quarterly, 51, 64–81. https://doi.
Odaci, H. (2011). Academic self-ecacy and academic procrastination as predictors of problematic
internet use in university students. Computers & Education, 57, 1109–1113. https://doi.org/10.1016/
Phillips, J. C., & Russell, R. K. (1994). Research self-ecacy, the research training environment,
and research productivity among graduate students in counseling psychology. The Counseling
Psychologist, 22, 628–641. https://doi.org/10.1177/0011000094224008
Poondej, C., & Lerdpornkulrat, T. (2016). Relationship between motivational goal orientations, perceptions
of general education classroom learning environment, and deep apperaches to learning. Kasetsart
Journal of Social Sciences, 37, 100–103. https://doi.org/10.1016/j.kjss.2015.01.001
Roksa, J. (2010). Bachelor’s degree completion across state contexts: Does the distribution of enrollments
make a dierence? Research in Higher Education, 51, Article 1. https://doi.org/10.1007/s11162-009-
Saini, M. (2016). A study of self-condence and academic achievement of high school students in relation
to certain socio-demographic variables. International Journal of Research in Economics and Social
Sciences, 6(6), 38–48.
Schöber, C., Schütte, K., Köller, O., McElvany, N., & Gebauer, M. M. (2018). Reciprocal eects between
self-ecacy and achievement in mathematics and reading. Learning and Individual Dierences, 63,
Scouller, K. (1998). The inuence of assessment method on students' learning approaches: Multiple
choice question examination versus assignment essay. Higher Education, 35, 453-472. https://doi.
Shu, X., Fan, P. L., Li, X., & Marini, M. M. (1996). Characterizing occupations with data from the
Dictionary of Occupational Titles. Social Science Research, 25, 149–173. https://doi.org/10.1006/
Smeby, J. (2000). Disciplinary dierences in Norwegian graduate education. Studies in Higher Education,
25, 53–67. https://doi.org/10.1080/030750700116019
Smith, C., Crosnoe, R., & Chao, S.-Y. (2016). Family background and contemporary changes in young
adults’ school-work transitions and family formation in the United States. Research in Social
Stratication and Mobility, 46, 3–10. https://doi.org/10.1016/j.rssm.2016.01.006
Smith, K. J., Grundmann, O., & Li, R. M. (2018). The development and impact of active learning strategies
on self-condence in a newly designed rst-year self-care pharmacy course-outcomes and
experiences. Currents in Pharmacy Teaching and Learning, 10, 499–504. https://doi.org/10.1016/
Stankov, L., Lee, J., Luo, W., & Hogan, D. J. (2012). Condence: A better predictor of academic
achievement than self-ecacy, self-concept and anxiety? Learning and Individual Dierences, 22,
Tavani, C. M., & Losh, S. C. (2003). Motivation, self-condence, and expectations as predictors of the
academic performances among our high school students. Child Study Journal, 33, 141–151.
Ullman, J. B., & Bentler, P. M. (2012). Structural equation modeling. In I. B. Weiner (Ed.), Handbook of
psychology (2nd ed., pp. 661–690). John Wiley & Sons. https://doi.org/10.1002/9781118133880.
University Act. (2015). Laws & Regulations Database of the Republic of China (Taiwan). https://law.moj.
Master’s Study Duration 899
Valiee, S., Moridi, G., Khaledi, S., & Garibi, F. (2016). Nursing students' perspectives on clinical instructors'
eective teaching strategies: A descriptive study. Nurse Education in Practice, 16, 258–262. https://
van der Aar, L. P. E., Peters, S., van der Cruijsen, R., & Crone, E. A. (2019). The neural correlates of
academic self-concept in adolescence and the relation to making future-oriented academic choices.
Trends in Neuroscience and Education, 15, 10–17. https://doi.org/10.1016/j.tine.2019.02.003
Vealey, R. S., & Chase, M. A. (2008). Self-condence in sport. In T. S. Horn (Ed.), Advances in sport
psychology (pp. 68–97, 430–435). Human Kinetics.
Venkatesh, V., & Brown, S. A. (2001). A longitudinal investigation of personal computers in homes:
Adoption determinants and emerging challenges. MIS Quarterly, 25, 71–102. https://doi.
Vos, N., van der Meijden, H., & Denessen, E. (2011). Eects of constructing versus playing an educational
game on student motivation and deep learning strategy use. Computers & Education, 56, 127–137.
Wamala, R., & Oonyu, J. C. (2012). Completion time dynamics for master's and doctoral studies at
Makerere University. Contemporary Issues in Education Research, 5, 131–138. https://doi.
Wang, Y., Peng, H., Huang, R., Hou, Y., & Wang, J. (2008). Characteristics of distance learners: Research
on relationships of learning motivation, learning strategy, self‐ecacy, attribution and learning
results. Open Learning: The Journal of Open, Distance and e-Learning, 23, 17–28. https://doi.
Warburton, K. (2003). Deep learning and education for sustainability. International Journal of Sustainability
in Higher Education, 4, 44–56. https://doi.org/10.1108/14676370310455332
White, K. A. (2009). Self-condence: A concept analysis. Nursing Forum, 44, 103–114. https://doi.
Yang, L. (2018). Higher education expansion and post-college unemployment: Understanding the roles of
elds of study in China. International Journal of Educational Development, 62, 62–74. https://doi.
Zhang, Q. (2015). Correlation between self-ecacy and deep learning of college students in the
environment of e-learning. e-Education Research, 36, 55–61. https://doi.org/10.13811/j.cnki.
收稿日期：2020 年 09 月 09 日
一稿修訂日期：2021 年 05 月 17 日
二稿修訂日期：2021 年 08 月 02 日
三稿修訂日期：2021 年 08 月 04 日
四稿修訂日期：2021 年 08 月 10 日
接受刊登日期：2021 年 08 月 11 日
略、學術自信心，以及如何影響修業時程。本研究共獲得 245 份有效問卷，包括 91 位男性填答
者及 154 位女性填答者。此外，研究參與者中有 34.3% 準時畢業，因為論文而未準時畢業者則占
教育心理學報，2022，53 卷，4期，879–900 頁