British Journal of Educational Technology Vol 0 No 0 2019 1–18

doi:10.1111/bjet.12747

Exploring autonomous learning capacity from a self-regulated

learning perspective using learning analytics

Zacharoula Papamitsiou and Anastasios A. Economides

Zacharoula Papamitsiou is a Computer Science and Informatics teacher in secondary education, and a junior
researcher at SMILE (Smart and Mobile Interactive Learning Environments) Lab. Currently she is a Ph.D. candidate
in Information Systems at the University of Macedonia. Her domain of expertise is on adapting and personalizing the
learning services and on supporting learners’ decision making using Learning Analytics. Anastasios A. Economides
is Full Professor on Computer Networks & Telematics Applications and Director of the SMILE (Smart & Mobile
Interactive Learning Environments) laboratory at the University of Macedonia, Thessaloniki, Greece. His research
interests include personalized and collaborative learning & assessment, user experience and acceptance of smart
systems and services and networking techno-economics. Address for correspondence: Zacharoula Papamitsiou,
University of Macedonia, 156 Egnatia street, Thessaloniki 54636, Greece. Emails: papamits@uom.edu.gr,
papamitsiou.xara@gmail.com

Abstract
Practising self-regulated learning (SRL) has been proposed to develop learning
autonomy. However, there is lack of empirical evidence on how SRL strategies affect
autonomous learning capacity. This study attempts to bridge that gap by utilizing the
learners’ trace data for measuring the learners’ autonomous interactions, and
investigates the effects of four SRL strategies on learners’ autonomous choices. The goal
is to explain how the employed SRL strategies impact autonomous control (in terms of
frequencies of self-enforced decisions, as well as time-spent on decision making). The
results from an exploratory study with undergraduate learners (N = 113) shown that
goal-setting and time-management have strong positive effects on autonomous control,
effort-regulation moderately positively affects learners’ autonomy, while help-seeking
has a strong negative effect. These findings provide empirical evidence and contribute
to clarifying the role of each one of the SRL strategies in the development of autonomous
learning capacity, from a learning analytics perspective. Limitations and potential
implications for research and practice are also discussed.

Introduction
Mariana is an electrical engineer professional. Since her early high-school years, she was inter-
ested in astrophysics and the universe science. She was encouraged by her teachers to determine
her learning goals in line with her values and interests, and to control and regulate her behav-
ior and actions accordingly, for attaining those goals. As such, she took some relevant courses
during her university studies and she reads books and articles online. Recently, she enrolled
herself in a Massive Open Online Course (MOOC) related to cosmology, motivated by her own in-
terest on the topic. Rather than merely “reacting” to situations that provide her with the oppor-
tunity to learn (eg, responding to teacher-controlled instruction), Mariana “proactively” seeks
out knowledge: she sets her learning goals, regulates her efforts, allocates study-time, monitors
her progress, seeks help when needed, critically reflects on her learning, and gradually, becomes

© 2019 British Educational Research Association

2  
British Journal of Educational Technology  Vol 0 No 0 2019

Practitioner Notes
What is already known about this topic
• Autonomous learning is more possible to be efficient because it encompasses the
learner’s freedom of choice and control over learning, according to what is important
to the learner herself.
• Self-regulated learning is a process comprising of strategies that can be trained, while
autonomy is a psychological need for experiencing volition and is difficult to be
trained.
• Empirical research on how learning strategies affect the development of capacity for
autonomous learning is based solely on learners’ self-reported perceptions and not on
measurements of autonomous interactions.
What this paper adds
• The current study takes us a step ahead by considering the learners’ autonomous
choices and provides an example of modeling autonomy using learning analytics.
• The moderate effect of self-regulated learning strategies on autonomous choices con-
tributes to understanding that exercising self-regulation is not enough for becoming
autonomous learner.
• This study opens the discussion on how to measure autonomous interactions, how
autonomous learning capacity can be developed, and how autonomous learning ca-
pacity development can be assessed.
Implications for practice and/or policy
• Practitioners shall be able to design and integrate specific features into online learn-
ing environments, that allow for autonomous choices, in a controlled manner, to
train and guide learners towards effectively using the self-regulated learning
strategies.
• Learners shall have opportunities to exercise control over their learning towards
aligning their own self-set learning goals with their achievements.
• The role of other traits (eg, personality) and strategies (eg, autonomy-supportive
teaching) should be explored in the scope of understanding learning autonomy.

aware of her learning needs. In the open online learning environment of the MOOC, Mariana
has the overall control of her learning trajectory: from the selection of the topic, to the selection
of the course, and to the degree of completion of the course, according to the goals she has set for
herself, Mariana systematically specifies and tracks her path.
In this example, the learner is an autonomous and self-regulated learner. When the learner is
autonomous, the learning is possible to be efficient because it is primarily important to herself.
Central to Self-Determination Theory (SDT; Deci & Ryan, 2002) is the notion that autonomy
relates to greater learning benefits. The role of autonomy on motivating learner to acquire self-reg-
ulation skills has been extensively explored in literature (eg, Lüftenegger et al., 2012; Pintrich,
2000; Reeve, 2006; Wolters, 2003). However, how learners develop their autonomous learning
capacity is still a “black box.” The adoption of autonomy-supportive environments that provide
learners the opportunity to exercise self-regulation strategies might contribute to increasing the
efficiency of autonomous choices (eg, Andrade, 2014; Benson, 2011; Jossberger, Brand-Gruwel,
Boshuizen, & van de Wiel, 2010; Loyens, Magda, & Rikers, 2008). This study empirically explores
© 2019 British Educational Research Association
 Effects of self-regulation on autonomous learning    3

how specific self-regulated learning strategies (ie, goal-setting, effort regulation, help-seeking and
time-management) are externalized as autonomous interactions within an online learning envi-
ronment that allows freedom of choice (selection of tasks), using learning analytics.

Bringing self-regulated learning and autonomous learning on the same page:

similarities, differences and where they intersect
Mariana is a self-regulated learner. Self-regulated learning (SRL) is conceptualized as an “active,
constructive process whereby learners set goals for their learning and then attempt to monitor,
regulate, and control their cognition, motivation, and behavior, guided and constrained by their
goals and the contextual features in the environment” (Pintrich, 2000, p. 435). Self-regulated
learners are aware of their learning processes and adjust their behavior (self-corrections) to keep
themselves on track towards their desired outcomes (Carver & Scheier, 2011; Pintrich, 2003;
Reeve, Ryan, Deci, & Jang, 2008). SRL is guided by motivation, metacognition, strategic action
(planning, monitoring and evaluation), and the specifications of the learning environments
(McCardle & Hadwin, 2015; Pintrich, 2000; Zimmerman, 2002).
Mariana is also an autonomous learner. For Holec (1981, p. 3), autonomous learning is the
“capacity to take charge of one’s own learning.” This definition assumes accepting responsibility
over all spectrum of the learning process, regardless of the learning context or the specifications
of the learning environment. Recently, Huang and Benson (2013) argued that capacity to con-
trol learning comprises ability (knowledge and skills to plan, monitor, evaluate learning), desire
(motivation, volition, willingness) and freedom (permission to control).
Based on the generic definitions of autonomous learning and SRL per se, one gets the impression
that both terms point to a significant notional overlap: both concepts emphasize learners’ active
engagement, goal-directed behavior, control, metacognition and responsibility. Intrinsic motiva-
tion is also prominent in both (Deci & Ryan, 2002). Although these two terms somehow conflate,
they should not be confused with each other. Few studies attempted to bring these concepts on the
same page and focused on identifying their similarities, differences and where they intersect (eg,
Andrade & Bunker, 2009; Cosnefroy & Carré, 2014; Lewis & Vialleton, 2011; Loyens et al., 2008).
Autonomy is a psychological need tied to the learners’ sense of volition and control. Core to the
idea of autonomy is freedom of choice (Andrade & Bunker, 2009; Loyens et al., 2008; Pintrich
& Schunk, 1996), placing the learner at the outset of the learning task. On the contrary, SRL
emerged from educational psychology research informed by social cognitive theory and is a pro-
cess comprising a set of strategies and placing emphasis on how learner can be effective without
reliance on teacher structure/control (Reeve, 2009; Stefanou et al., 2013; Vansteenkiste, et al.,
2012). Furthermore, autonomy targets at fostering learners’ responsible self-initiative and allows
them to determine the selection of what shall be learned, as well as the critical evaluation (reflec-
tion) of the learning tasks that were selected (Candy, 1991): autonomous learner is able to define
what needs to be learned. On the other hand, SRL seems more concerned with the subsequent steps
in the learning process, such as setting goals, monitoring their progress, reflecting on the steps
that were taken and changing their plans accordingly. Self-regulated learner plans the learning
activity and enacts tactics and strategies before initiating the learning task. SRL is described as the
processes that the learner substantiates (Zimmerman, 2000).

The need to examine autonomous learning from a self-regulated learning

perspective
When the learner is self-regulated, the learning is very likely to be efficient: self-regulation
has been associated with deeper and more permanent learning (eg, Broadbent & Poon, 2015;
© 2019 British Educational Research Association
4  
British Journal of Educational Technology  Vol 0 No 0 2019

Pintrich, 2003; Tsai, Shen, & Tsai, 2011; Wang, Shannon, & Ross, 2013). For one to become effi-
cient self-regulated learner, the role of autonomous motivation has been extensively explored in
literature and is beyond question (eg, Lüftenegger et al., 2012; Maldonado-Mahauad et al., 2018;
Pintrich, 2000; Reeve et al., 2008; Wolters, 2003). Learners can acquire and improve self-regu-
lation skills, through guidance and practise (Dignath, Buettner, & Langfeldt, 2008) in learning
environments that promote self-initiative and support autonomous motivation (Pintrich, 2000).
But first, learners should develop their autonomous capacity.
However, learners do not intuitively know how to build their autonomous learning capacity and
make efficient learning choices; autonomy is not a skill, and therefore, it is difficult to be “trained.”
Within SDT, autonomy is perceived as the inner endorsement of one’s behavior, ie, a psychological
need for experiencing volition (Deci & Ryan, 2002). All learners are potentially autonomous, but
(a) the degree of autonomy each learner achieves depends on the characteristics of the learner
(eg, personality; Olesen, Thomsen, Schnieber, & Tønnesvang, 2010), and (b) the learner who
lacks autonomy can developing it under appropriate conditions (Benson, 2011).
Mariana was not always an autonomous learner; she was encouraged by her teachers to take
responsibility and control of the learning choices according to her values. For Little (1995),
autonomy does not imply learning in isolation and complete lack of support, but rather interde-
pendence among teachers and learners. Autonomy-supportive teaching and autonomy-support-
ive design of learning environments have been proposed to increase autonomous motivation,
engagement and self-regulation (eg, Andrade, 2014; Jossberger et al., 2010; Reeve, 2006; Sun
& Rueda, 2012; Vansteenkiste, Simons, Lens, Sheldon, & Deci, 2004). In these conditions, learn-
ers have more freedom to pursue their goals and undertake critical evaluation of the materials
they select. It has also been argued that one needs to exercise self-regulation within appropriately
configured environments to become a capable autonomous learner (eg, Andrade, 2014; Benson,
2011; Jossberger et al., 2010; Loyens et al., 2008). In other words, mastering the skills to self-reg-
ulate activities and tasks can be the first step to learning how to become autonomous.
Apparently, it is very likely that self-regulation might affect the self-enforced choices in terms of
interactions measured with learning analytics parameters: the question is how. How SRL strate-
gies contribute to developing learning autonomy is still a “black box.” It is critical to examine how
SRL strategies are externalized as learners’ autonomous choices, in learning environments that
allow learners to be consciously involved in their own learning by supporting freedom of choices,
and to investigate autonomous interactions from the SRL perspective.

Autonomous learning and SRL strategies in online learning environments

The more the learning turns online, the higher is the need for learners to develop and sustain
their autonomous learning trait. The reason behind this claim is that contemporary online learn-
ing environments provide learners more opportunities to freely choose what, where and how to
learn, eliminating the restriction of place, time and physical materials and, up-to a degree, giv-
ing the learners substantial control of their learning (compared to traditional classrooms or to
blended learning environments) (Broadbent & Poon, 2015; Xu & Jaggars, 2014). These environ-
ments inherently promote learners’ active engagement in the learning process and at the same
time require learners’ abilities to manage their own learning processes (Sun & Rueda, 2012; Tsai
et al., 2011) by providing less support and guidance on how to efficiently and deeply learn (Wang
et al., 2013). In order learners to be(come) capable to exploit the opportunities of online learning
environments to efficiently learn, they should be trained and develop their autonomy.
Kormos and Csizér (2014) examined the mediation effect of self-regulation on autonomy in
computer-assisted learning conditions and found that time-management predicted learners’
© 2019 British Educational Research Association
 Effects of self-regulation on autonomous learning    5

perception of autonomous use of learning resources, but a strong motivation was a prerequi-
site for the adoption of the strategy. Τhe metacognitive regulation of cognition was also pos-
itively related to students’ perceptions of autonomy support (Schuitema, Peetsma, & van der
Veen, 2016; Sierens, Vansteenkiste, Goossens, Soenens, & Dochy, 2009), and learners’ deep-
level learning strategies and effort-regulation during the learning process were also positively
related with perceived autonomy support (Vansteenkiste et al., 2012). The results of another
study shown that effort-regulation was the most important SRL strategy recommended by suc-
cessful MOOC learners (Kizilcec, Pérez-Sanagustín, & Maldonado, 2017). Furthermore, effec-
tive help-seeking has been viewed as an important strategy that enables students to maintain
engagement and leads to long-term mastery and autonomous learning (Puustinen, 1998;
Ryan, Hicks, & Midgley, 1997). However, it was found that students in online learning con-
ditions perceived desire for autonomy as one of the main reasons for avoiding seeking help
(Kizilcec et al., 2017).
There is a lack of empirical knowledge on how SRL strategies are related to learners’ exercise of
autonomous control. The present study aims to fill this gap by explaining the variance in auton-
omous interactions based on SRL strategies, using learning analytics. This knowledge is expected
to improve our understanding of the self-regulation mechanisms that can build up autonomy,
as well as to allow us to adjust the design of the learning environments. Therefore, the research
question (RQ) this study aims to address is defined as follows:
RQ: What is the effect of self-regulated learning strategies on learners’ autonomous interactions
(measured as utilized learning analytics)?

The research model and hypotheses

This study is contextualized in online self-assessment conditions. Self-assessment promotes the
development of learners’ capacity to self-regulate their behavior and to preserve their autonomy
(McMillan & Hearn, 2008). A recent contextualization of SRL that mostly apply in online learn-
ing environments includes six prevalent strategies, ie, goal-setting, time-management/study-en-
vironment, help-seeking, task-strategies (eg, effort regulation and rehearsal), peer-learning and
self-evaluation (Barnard, Lan, To, Paton, & Laic, 2009). In this study, we formulate hypotheses
about the four (out of the six) self-regulation strategies that have been associate with autono-
mous learning in literature. In the present study, peer-learning and self-evaluation do not apply
because the online learning environment is self-assessment oriented (peer-learning does not
hold, and self-evaluation is inherent by definition).
As such, in order to address the research question, the research hypotheses on the relationships
between the considered factors are outlined as follows.
Based on the results from Vansteenkiste et al. (2012) and Kizilcec et al. (2017) we believe that the
sense of acting freely and making independent choices can be seen as an opportunity for students
who properly regulate their efforts according to the requirements of the tasks to exhibit higher
autonomous control, as well. Thus:
H1:  Effort-regulation will have a positive effect on Autonomous Control.
Goal-setting is a skill associated with the learners’ self-set learning expectations, whereas self-set
goals produce higher goal commitment (Zimmerman, 2000). We believe that the learners who
are aware of their goals and have high learning and achievement expectations, will take advan-
tage of autonomous control and will select tasks that will facilitate their goals. Therefore:
H2:  Goal-setting will have a positive effect on Autonomous Control.

© 2019 British Educational Research Association

6  
British Journal of Educational Technology  Vol 0 No 0 2019

Based on previous arguments (Nelson-Le Gall, 1985; Newman, 2002; Puustinen, 1998) that
seeking understanding-aided help can lead to greater autonomy, we believe that learners who
have strong help-seeking skills will feel free to ask for multiple levels of hints during dealing with
learning tasks. Therefore:
H3:  Help-seeking will have a positive effect on Autonomous Control.

Time-management is a meta-skill aiming at efficiently allocating time on each of the tasks, within
a limited amount of time (Michinov, Brunot, Bohec, Juhel, & Delaval, 2011). In line with Kormos
and Csizér (2014), students who believe that have good time-management skills are expected to
allocate time on tasks they have chosen and fine-tune their time-management practices. Thus:
H4:  Time-management will have a positive effect on Autonomous Control.
Figure 1 illustrates the causal relationships among the considered factors.

Methods
Participants and system
Overall, one hundred and thirteen (113) undergraduate students (51 females [45.1%] and 62
males [54.9%], aged 19–26 years-old [M = 20.74, SD = 1.755, N = 113]) at a European University
were enrolled in an online self-assessment procedure for the Management Information Systems
I course (related to Information Systems, databases and Business Intelligence), at the University
lab, for 60 minutes. The participation to the procedure was optional. The self-assessment tests
were offered to facilitate the students’ self-preparation before the final exams, to help them track
their progress and align with their learning goals and the scores on these tests had no participa-
tion to the final grade (ie, no rewards as external motivation).

Procedure
Calibration of the tasks
For the needs of the self-assessment, 150 multiple-choice questions (tasks) in total were cali-
brated before being available to the students. The purpose of the calibration was to allocate each

Figure 1:  Overall research model and factor relationships with hypotheses
© 2019 British Educational Research Association
 Effects of self-regulation on autonomous learning    7

task to one category of difficulty and discrimination ability. The discrimination ability of a task
corresponds to the probability of students in a given mastery class responding correctly to each
task. For the tasks’ discrimination ability configuration, three mastery classes of students were
used (ie, Class A: final grade ≥ 7, Class B: final grade ≥ 4 and Class C: final grade < 4) and the
respective probabilities were computed, using the results from prior self-assessment tests (involv-
ing students who have already been classified).
Furthermore, two experts agreed on the tasks’ difficulty (easy, medium and hard).

The self-assessment procedure

For the self-assessment, the students had to complete up to 12 tasks within the 60 minutes.
The rationale behind this restriction is that we wanted to keep the students in line with the four
self-regulation strategies explored in the study: to freely select tasks according to their goal-set-
ting, to regulate their effort according to the difficulty of the tasks and the limited available time,
to manage their time-allocation per task within limited time-settings, and to question help-seek-
ing by downgrading task difficulty (in a “stepping-back” manner, lowering performance aspira-
tion [Karabenick & Knapp, 1991]). Each task had four possible answers, but only one was correct.
The first task was randomly assigned to the students. The students had full-autonomy to select
the next self-assessment task according to the desired level of difficulty of that task. As shown in
Figure 2, the students could ask either for a task of the same difficulty with the current one, or for
an easier or harder task, or for a random task (delivered to them according to the discrimination
ability of the task and the students’ currently diagnosed mastery level—the estimation of the stu-
dents’ mastery class is updated every time they submit an answer). The students could skip a task,
and ask for a new one, but they could not revise previous (already answered or skipped) tasks. The
self-assessment was finalized either when the 12 tasks were completed (80.1% of the participants),
or when a maximum possible score was achieved (ie, 10) (11.6% of the participants), or when the
available time of 60 minutes was exceeded (8.3% of the participants). Figure 2 synopsizes this pro-
cedure.For the score computation, only the correct answers were considered up-to-that moment,
without penalizing the incorrect answers. Each task’s participation on the score was according to
its difficulty, varying from 0.5 points (easy) to 1 point (medium) to 1.5 points (hard). The students
knew that they could not achieve a maximum score (ie, 10) by selecting only easy tasks, and that if
they achieved the maximum score, the quiz would finalize.
The online self-assessment system employed in the study is illustrated in Figure 3 (Papamitsiou &
Economides, 2013).

Figure 2:  Synopsis of the self-assessment procedure

© 2019 British Educational Research Association
8  
British Journal of Educational Technology  Vol 0 No 0 2019

Figure 3:  The online self-assessment environment

Before taking the self-assessment, each participant had to answer to a pretest questionnaire
that measures the SRL strategies explored in this study, ie, their perceptions of effort-regulation,
goal-setting, help-seeking and time-management. All participants signed an informed consent
form prior to their participation. The informed consent explained to them the procedure and
was giving the right to researchers to use the data collected for research purposes. Students were
aware that their interactions had been tracked and anonymized prior to being analyzed, and that
the collected data would be stored for 3 years.

Data collection
Measures
Data were collected with the online self-assessment environment illustrated in Figure 3, accord-
ing to the process described in previous section. Measures commonly used in the field of learning
analytics (eg, response-times and frequencies) (Maldonado-Mahauad et al., 2018; Papamitsiou
& Economides, 2019; Papamitsiou, Economides, Pappas, & Giannakos, 2018), indicative of the
learners’ autonomous control (AC) interactions, were computed from the logged clickstreams.
Table 1 illustrates the measures captured and coded, and their descriptive statistics.
The AC variable was modeled using Partial Least Squares (PLS) for dimensionality reduction. In
the initial model, for AC the factor loadings for FESC, FTEH, FTHE and SST were lower than 0.7

© 2019 British Educational Research Association

 Effects of self-regulation on autonomous learning    9

Table 1:  Measurements used in the study

Variable Mean (Std. Dev.) Name Description

FEAS 4.887 (1.549) Frequency of How many times the student asks for easier
choosing easier questions (compared to the current one)
FHAR 3.494 (1.631) Frequency of How many times the student asks for harder
choosing harder questions (compared to the current one)
FSAM 3.173 (1.281) Frequency of How many times the student asks for question
choosing same of the same difficulty with the current one
FRAN 1.708 (1.005) Frequency of How many times the student asks for random
choosing random questions (automatically assigned)
FESC 1.223 (1.012) Frequency of How many times the student ignores a question
skipping
FTEH 2.133 (1.127) Frequency of How many times the student asks for a harder
transition from question after an easier one (ie, from easy to
easier to harder medium or hard, and from medium to hard)
FTHE 2.843 (1.224) Frequency of How many times the student asks for an easier
transition from question after a harder one (ie, from hard to
harder to easier medium or easy and from medium to easy)
TTDM 62.236 (32.072) Time-spent on The time interval between answering a
decision making question and choosing the next one
SST 1386.04 (437.694) Study session time The total study session time for each student

and as such they were removed from the model. The consistency reliability of the factor that mea-
sures autonomy was confirmed (a = 0.791).

Instruments
In order to develop the instrument for measuring the four SRL strategies, we adapted some items
of the constructs from previously validated instruments. For students’ effort-regulation (ER) we
adopted three items from the Motivated Strategies for Learning Questionnaire (MSLQ; Pintrich,
Smith, Garcia, & Mckeachie, 1993). For goal-setting, we used the goal-expectancy construct, and
configured three items from Terzis and Economides (2011). Goal-expectancy (GE) is a measure
of goal-setting particularized on assessment procedures. For help-seeking (HS), three items were
adopted from the Online Self-Regulated Learning Questionnaire (OSLQ; Barnard et al., 2009).
Finally, for time-management (TM) we configured three items from the OSLQ, as well. The items
from each instrument were selected and modified for the context of the self-assessment (for all
constructs, Cronbach’s was above 0.7, ensuring internal consistency). The questionnaire was
first developed in English and then translated into the students’ native language, by certified
translators to ensure linguistic equivalence. All items were measured in a 7-point Likert-like
scale (1 = strongly disagree to 7 = strongly agree, Appendix Table A1).

Data analysis
Structural and measurement model
For addressing the research question, the construction of a path diagram that contains the struc-
tural and measurement model was conducted with the Partial least-squares Structural Equation
Modeling (PLS-SEM) technique (Chin, 1998; Tenenhaus, Vinzi, Chatelin, & Lauro, 2005). PLS-
SEM is an exploratory technique that allows for estimating and directly testing theoretically
proposed chains of cause and effect between the latent variables (Sarstedt, Ringle, & Hair, 2017).
© 2019 British Educational Research Association
10  British Journal of Educational Technology  Vol 0 No 0 2019

PLS is a non-parametric, distribution-free method and was selected to reduce the predictors to
a smaller set of uncorrelated components and perform least-squares regression on them. Our
sample of 113 participants exceeds the recommended value of 50, ie, (a) 10 times larger than the
number of items for the most complex construct (AC with five items), and (b) 10 times the largest
number of independent variables impact a dependent variable (ER, GE, HS, TM to AC).

Measures and evaluation criteria

The structural model evaluates the relationship between exogenous and endogenous latent variables
by examining the variance measured (R2). R2 values of 0.67, 0.33 and 0.19 are substantial, moderate
and weak respectively (Chin, 1998). The quality of path model can be evaluated by the Stone-Geisser’s
Q2 value (Geisser, 1974; Stone, 1974), an evaluation criterion for the cross-validated predictive rele-
vance of the PLS path model. The Q2 statistic measures the predictive relevance of the model by repro-
ducing the observed values by the model itself. A Q2 greater than 0 means the model has predictive
relevance; Q2 statistic less than 0 mean that the model lacks predictive relevance. Finally, a bootstrap
procedure evaluates the significance of the path coefficients (β value) and total effects, by calculating
t-values. For the measurement and the structural model we used SmartPLS 3.2.

Results
Convergent validity—discriminant validity
The results support the measurement model. All criteria for convergent validity are met: the items’
factor loadings on the corresponded constructs are higher than 0.7 (Chin, 1998), Cronbach’s
a and composite reliability of all constructs are higher than 0.7 and confirm reliability of the
measurement model, and Average Variance Extracted (AVE) is higher than 0.5, exceeding the
variance due to measurement error for that construct. Table 2 displays these results.
Discriminant validity is also confirmed since the AVE of each construct is higher than the construct’s
highest squared correlation with any other construct (Fornell & Larcker, 1981). Table 3 presents the
variables’ correlation matrix; the diagonal elements are the square root of the AVE of a construct.

Testing hypotheses
A bootstrap procedure with 3000 resamples was used to test the statistical significance (t-value) of
the path coefficients (β value) in the model. Table 4 summarizes the results for the hypotheses testing.
As seen from this table, two of the initial hypotheses are supported, one is not supported and for
one hypothesis, its negation is supported. These results are discussed in next section.

Overall model fit

According to these results, the suggested model explains almost the 33% of the variance in au-
tonomous control, which is statistically moderate. The cross-validated predictive relevance of
the model was also confirmed (Q2 = 0.261). Table 5 synopsizes the total effects of the selected
factors, as well as the variance (R2) and cross-validated predictive relevance (Q2) explained by
the proposed model.
The measurement results are summarized in Figure 4. This figure illustrates the path coefficients
for the initial hypotheses of the research model.

Discussion
Findings and interpretations
Contemporary online learning environments provide learners more opportunities to freely
choose what, where and how to learn, eliminating the restriction of place, time and physical

© 2019 British Educational Research Association

 Effects of self-regulation on autonomous learning    11

Table 2:  Results for the latent constructs of the measurement model

Factor loadings Cronbach’s a Composite reliability Average variance

Construct items (>0.7)a (>0.7)a (>0.7)a extracted (>0.5)a

ER 0.777 0.857 0.602

ER1 0.818
ER2 0.853
ER3 0.823
GE 0.871 0.920 0.794
GE1 0.885
GE2 0.893
GE3 0.895
HS 0.797 0.871 0.693
HS1 0.906
HS2 0.857
HS3 0.725
TM 0.853 0.911 0.773
TM1 0.908
TM2 0.851
TM3 0.877
AC 0.791 0.864 0.625
FEAS 0.803
FHAR 0.871
FSAM 0.702
FRAN 0.709
TTDM 0.716
ER: Effort-regulation, GE: Goal-expectancy, HS: Help-seeking, TM: Time-management, AC: Autonomous
Control, FEAS: Frequency of choosing easier, FHAR: Frequency of choosing harder, FSAM: Frequency of
choosing same, FRAN: Frequency of choosing random, TTDM: Time-spent on decision making.
a
Indicates an acceptable level of reliability and validity.

Table 3:  Measurement model (discriminant validity)

Mean (Std. Dev.) 1 2 3 4 5

1. Effort-regulation 4.553 (1.212) 0.831

2. Goal-expectancy 4.539 (1.342) 0.604 0.891
3. Help-seeking 5.454 (1.307) 0.334 0.595 0.833
4. Time-management 4.684 (1.296) 0.616 0.708 0.480 0.879
5. Autonomous control 2.637 (1.012) 0.422 0.501 0.187 0.488 0.776

materials (Broadbent & Poon, 2015; Xu & Jaggars, 2014). However, learners do not intuitively
know how to build their autonomous learning capacity and make efficient learning choices.
Autonomy-supportive teaching and autonomy-supportive design of learning environments
have been proposed to increase autonomous motivation, engagement and self-regulation (eg,
Andrade, 2014; Jossberger et al., 2010; Reeve, 2006; Sun & Rueda, 2012; Vansteenkiste et al.,
2004). The question is how SRL strategies influence the learners’ autonomous interactions,
within inherently autonomous online learning environments.
Previous studies relied on learners’ perceptions of autonomy. The present study targets at pro-
viding empirical evidence on how SLR strategies predict and justify the learners’ self-enforced
choices, using learning analytics.
© 2019 British Educational Research Association
12  British Journal of Educational Technology  Vol 0 No 0 2019

Table 4:  Hypothesis testing results

Hypothesis Path β t P Result

H1 Effort-regulation → Autonomous 0.105 1.076 0.283 Not support

control
H2 Goal-expectancy → Autonomous 0.374 3.388* 0.001 Support
control
H3 Help-seeking → Autonomous −0.190 2.154* 0.032 Support opposite
control
H4 Time-management → 0.250 2.244* 0.025 Support
Autonomous control
*
p < 0.05.

Table 5:  R2, Q2 and total effects

Endogenous R2 Q2 Exogenous Total effect t

Autonomous Control 0.332 0.261 Effort-regulation 0.105 1.076

Goal-expectancy 0.374 3.388*
Help-seeking −0.190 2.154*
Time-management 0.250 2.244*
*
p < 0.05.

Figure 4:  Path coefficients of the research model, overall variance explained (R2) for test score and cross-validated
predictive relevance (Q2)

In particular, in the demonstrated approach, autonomy was modeled in terms of frequencies

of interaction types and time-spent on decision making, both commonly used in the learning
analytics research practise (Maldonado-Mahauad et al., 2018; Papamitsiou & Economides,
2019; Papamitsiou et al., 2018). Next, a set of specific SRL strategies that have been identified in
© 2019 British Educational Research Association
 Effects of self-regulation on autonomous learning    13

previous research as critical in online learning contexts (Broadbent & Poon, 2015; Kizilcec et al.,
2017; Kormos & Csizér, 2014; Sierens et al., 2009), and that could be associated to or infer auton-
omous behavior, were explored through a structural and measurement model. These factors were
measured with configured versions of previously validated instruments, and their consistency
reliability was also confirmed (Table 2).
The overall prediction accuracy of the model proposed in this study was 33.2%, which is mod-
erate, and the cross-validated predictive relevance was confirmed (Q2 = 26.1%) (Table 5). This
finding contributes to understanding the diversity of the two concepts and provides empirical evi-
dence that exercising self-regulation strategies is not enough for experiencing autonomous control.
Apparently, additional factors should be considered as well. For example, the degree of autonomy
each learner achieves depends on the characteristics of the learner. Olesen et al. (2010) argued
that this relationship is catalyzed by personality traits.
Moreover, the analysis revealed a strong direct positive effect of goal-setting and time-manage-
ment on autonomous control (β = 0.374, t = 3.388, p = 0.001; β = 0.250, t = 2.244, p = 0.025
for GE and TM respectively), confirming hypotheses H2 and H4, in line with and further confirm-
ing previous results (Kormos & Csizér, 2014). Practically, it extends previous findings by adding
empirical evidence, and it implies two posits: (a) learners who are aware of their goals and have
high learning and achievement expectations, will try to select tasks that will facilitate their goals,
and (b) learners who believe that they have good time-management skills, will take the opportu-
nity for autonomous control in order to adjust and fine-tune these skills.
In addition, a moderate positive effect of effort-regulation on autonomy (β = 0.105, t = 1.076,
p = 0.283) was detected, neither supporting nor rejecting hypothesis H1. In previous research
(Vansteenkiste et al., 2012) effort-regulation was found to be a strong determinant of perceived
autonomy. The slight divergence of our finding might be due to how autonomy was measured, high-
lighting the difference between intentions and real behaviors (Gollwitzer et al., 2009). Based on our
results, the role of this factor is not clear: although it seems that learners’ perceptions of persistence
in their engagement with the learning tasks might influence their self-enforced choice of tasks, how-
ever, in practise, effort expenditure on the selected tasks would possibly clarify more this relationship.
The most intriguing finding of this study concerns the strong direct negative effect of the
help-seeking factor on autonomous interactions (β = −0.190, t = 2.154, p = 0.032), resulting
in supporting the opposite of hypothesis H3 (Table 4). Due to the highly self-initiating nature of
help-seeking behavior, and in line with Nelson-Le Gall’s (1985) claims that help-seeking can pro-
mote autonomy, we assumed that help-seeking would positively influence autonomous control.
Online learning environments provide increased opportunities for free help-seeking and they can
preserve learners’ anonymity. However, this hypothesis was not confirmed, yet, the opposite result
was discovered. This finding is in line with the findings in Kizilcec et al. (2017), that help-seeking
contradicts autonomous control. A possible explanation of this result is that learners might per-
ceive help-seeking as a threat to their autonomy (Huet, Escribe, Dupeyrat, & Sakdavong, 2011).
This finding requires attention and should be further explored from different perspectives, includ-
ing actual measurements of help-seeking behavior in environments that facilitate autonomous
learning conditions.

Conclusions
In the introductory example, the learner is intrinsically directed to study and learn about a topic,
and she freely yet responsibly makes self-enforced decisions and choices for her learning, guided
by her own volition. At a macro-level, it is the learner she who determines why, what, when, how
and how much to learn, and has the overall control of the learning process, and it is also the
© 2019 British Educational Research Association
14  British Journal of Educational Technology  Vol 0 No 0 2019

learner she who initiates the learning cycle (autonomy). At a meso-level, the learner regulates
her behavior and actions and makes “contextualized” decisions: she still has the control and
responsibility for her choices, but the options derive from the context and not from the inner self. At
this level, the learner regulates her behavior and decides the degree up-to which she follows the
subsequent steps available to her according to the specific guidelines and/or requirements of the
learning context and the learning environment (self-regulation). The latest ones are defined at
a micro-level by the tutor/designer of the course/environment. In a sense, autonomy is broader
than self-regulation, and the results of the present study provide strong indications towards sup-
porting this claim, but further research is required on this direction.

Implications for research and practice

The findings offer implications for research and practice. Firstly, exploring additional factors (eg,
personality) as moderators of this relationship is required. Furthermore, understanding how
learners’ self-regulation contributes to and reflects their autonomy can provide insight on how
to plan the learners’ SRL support in online learning environments (research). As the next step,
practitioners shall be able to integrate specific features into online learning environment, in
order to train learners on effectively using the SRL strategies, accordingly (practice). For exam-
ple, supporting learners’ effort-regulation (eg, with learning analytics visualizations) might help
the learners to persist on their on-task engagement and guide them to choose tasks that better
correspond to their learning goals.
Moreover, attention is required regarding the impact of help-seeking on self-directed learning,
and the relationship between both self-initiated behaviors should be further explored (research).
For example, providing a help-seeking functionality to the learners, measuring their interactions
with this facility, and exploring the next autonomous choices of learning tasks, would provide
additional empirical evidence on the effect of help-seeking on autonomy. Clarifying this relation-
ship shall next open new directions towards making decisions on how to efficiently guide the
learners to feel free to seek help (practice).

Limitations and future work

Of course, there are limitations as well. First, the sample size (N = 113) of the present study is
relatively small, yet sufficient for the analysis methods employed. Larger samples should be ex-
plored to further validate the demonstrated findings. Second, two SRL strategies, ie, peer-learn-
ing and self-evaluation did not apply on the context of this study, and need to be considered for
analysis, as well. Last, the 60 minutes of the present study is a very limited time interval. The
duration of experimentation with respective procedures should increase. This is within our fu-
ture work plans.

Statements on open data, ethics and conflict of interest

The data can be obtained by request, by contacting the corresponding author.
Participation was voluntarily and all the data collected anonymously. Appropriate permissions
and ethical approval for the participation requested and approved.
There is no potential conflict of interest in this study.

Acknowledgements
This research did not receive any specific grant from funding agencies in the public, commercial
or not-for-profit sectors.
© 2019 British Educational Research Association
 Effects of self-regulation on autonomous learning    15

References
Andrade, M. S. (2014). Dialogue and structure: Enabling learner self-regulation in Technology-Enhanced
Learning environments. European Educational Research Journal, 13(5), 563–574. https://doi.org/10.2304/
eerj.2014.13.5.563
Andrade, M. S., & Bunker, E. L. (2009). A model for self-regulated distance language learning. Distance
Education, 30(1), 47–61. https://doi.org/10.1080/01587910902845956
Barnard, L., Lan, W. Y., To, Y. M., Paton, V. O., & Lai, S.-L. (2009). Measuring self-regulation in online and
blended learning environments. The Internet and Higher Education, 12(1), 1–6. https://doi.org/10.1016/j.
iheduc.2008.10.005
Benson, P. (2011). Teaching and researching autonomy (2nd ed.). London: Pearson Education Limited.
Broadbent, J., & Poon, W. L. (2015). Self-regulated learning strategies & academic achievement in online
higher education learning environments: A systematic review. The Internet and Higher Education, 27,
1–13. https://doi.org/10.1016/j.iheduc.2015.04.007
Candy, P. C. (1991). Self-direction for lifelong learning. San Francisco, CA: Jossey-Bass.
Carver, C. S., & Scheier, M. F. (2011). Self-regulation of action and affect. In R. F. Baumeister & K. D. Vohs
(Eds.), Handbook of self-regulation: Research, theory, and applications (2nd ed., pp. 3–21). New York, NY:
Guilford Press.
Chin, W. W. (1998). The partial least squares approach for structural equation modeling. In G. A.
Marcoulides (Ed.), Modern methods for business research (pp. 295–336). Mahwah, NJ: Lawrence Erlbaum
Associates Publishers.
Cosnefroy, L., & Carré, P. (2014). Self-regulated and self-directed learning: Why don’t some neighbors com-
municate? International Journal of Self-Directed Learning, 11(2), 1–11.
Deci, E. L., & Ryan, R. M. (2002). Handbook of self-determination research. Rochester, NY: University of
Rochester Press.
Dignath, C., Buettner, G., & Langfeldt, H.-P. (2008). How can primary school students learn self-regu-
lated learning strategies most effectively? A meta-analysis on self-regulation training programmes.
Educational Research Review, 3(2), 101–129. https://doi.org/10.1016/j.edurev.2008.02.003
Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and
measurement error. Journal of Marketing Research, 18(1), 39–50.
Geisser, S. (1974). A predictive approach to the random effect model. Biometrika, 61(1), 101–107.
Gollwitzer, P. M., Sheeran, P., Michalski, V., & Seifert, A. E. (2009). When intentions go public: Does so-
cial reality widen the intention-behavior gap? Psychological Science, 20(5), 612–618. https://doi.
org/10.1111/j.1467-9280.2009.02336.x
Holec, H. (1981). Autonomy in foreign language learning. Oxford: Pergamon.
Huang, J., & Benson, P. (2013). Autonomy, agency and identity in foreign and second language education.
Chinese Journal of Applied Linguistics, 36(1), 7–28.
Huet, N., Escribe, C., Dupeyrat, C., & Sakdavong, J.-C. (2011). The influence of achievement goals and per-
ceptions of online help on its actual use in an interactive learning environment. Computers in Human
Behavior, 27(1), 413–420. https://doi.org/10.1016/J.CHB.2010.09.003
Jossberger, H., Brand-Gruwel, S., Boshuizen, H., & van de Wiel, M. (2010). The challenge of self-directed
and self-regulated learning in vocational education: A theoretical analysis and synthesis of require-
ments. Journal of Vocational Education & Training, 62(4), 415–440. https://doi.org/10.1080/13636820.2
010.523479
Karabenick, S. A., & Knapp, J. R. (1991). Relationship of academic help seeking to the use of learning strate-
gies and other instrumental achievement behavior in college students. Journal of Educational Psychology,
83(2), 221–230. https://doi.org/10.1037/0022-0663.83.2.221
Kizilcec, R. F., Pérez-Sanagustín, M., & Maldonado, J. J. (2017). Self-regulated learning strategies predict
learner behavior and goal attainment in Massive Open Online Courses. Computers & Education, 104,
18–33. https://doi.org/10.1016/j.compedu.2016.10.001
Kormos, J., & Csizér, K. (2014). The interaction of motivation, self-regulatory strategies, and autonomous
learning behavior in different learner groups. TESOL Quarterly, 48(2), 275–299. https://doi.org/10.1002/
tesq.129

© 2019 British Educational Research Association

16  British Journal of Educational Technology  Vol 0 No 0 2019

Lewis, T., & Vialleton, E. (2011). The notions of control and consciousness in learner autonomy and
self-regulated learning: A comparison and critique. Innovation in Language Learning and Teaching, 5(2),
205–219. https://doi.org/10.1080/17501229.2011.577535
Little, D. (1995). Learning as dialogue: The dependence of learner autonomy on teacher autonomy. System,
23(2), 175–181. https://doi.org/10.1016/0346-251X(95)00006-6
Loyens, S. M. M., Magda, J., & Rikers, R. M. J. P. (2008). Self-directed learning in problem-based learn-
ing and its relationships with self-regulated learning. Educational Psychology Review, 20(4), 411–427.
https://doi.org/10.1007/s10648-008-9082-7
Lüftenegger, M., Schober, B., van de Schoot, R., Wagner, P., Finsterwald, M., & Spiel, C. (2012). Lifelong
learning as a goal—Do autonomy and self-regulation in school result in well prepared pupils? Learning
and Instruction, 22(1), 27–36. https://doi.org/10.1016/j.learninstruc.2011.06.001
Maldonado-Mahauad, J., Pérez-Sanagustín, M., Kizilcec, R. F., Morales, N., & Munoz-Gama, J. (2018). Mining
theory-based patterns from Big data: Identifying self-regulated learning strategies in Massive Open
Online Courses. Computers in Human Behavior, 80, 179–196. https://doi.org/10.1016/j.chb.2017.11.011
McCardle, L., & Hadwin, A. F. (2015). Using multiple, contextualized data sources to measure learn-
ers’ perceptions of their self-regulated learning. Metacognition and Learning, 10(1), 43–75. https://doi.
org/10.1007/s11409-014-9132-0
McMillan, J. H., & Hearn, J. (2008). Student self-assessment: The key to stronger student motivation and
higher achievement. Educational Horizons, 87(1), 40–49.
Michinov, N., Brunot, S., Bohec, O. Le, Juhel, J., & Delaval, M. (2011). Procrastination, participation, and
performance in online learning environments. Computers & Education, 56(1), 243–252. https://doi.
org/10.1016/j.compedu.2010.07.025
Nelson-Le Gall, S. (1985). Help-seeking behavior in learning. Review of Research in Education, 12, 55–90.
Newman, R. S. (2002). How self-regulated learners cope with academic difficulty: The role of adaptive help
seeking. Theory Into Practice, 41(2), 132–138. https://doi.org/10.1207/s15430421tip4102_10
Olesen, M. H., Thomsen, D. K., Schnieber, A., & Tønnesvang, J. (2010). Distinguishing general causality
orientations from personality traits. Personality and Individual Differences, 48(5), 538–543. https://doi.
org/10.1016/j.paid.2009.11.032
Papamitsiou, Z., & Economides, A. (2013). Towards the alignment of computer-based assessment outcome
with learning goals: The LAERS architecture. In 2013 IEEE Conference on e-Learning, e-Management and
e-Services, Kucing, Malaysia, (pp. 13–17). doi: https://doi.org/10.1109/IC3e.2013.6735958
Papamitsiou, Z., & Economides, A. (2019). Taking control of the self-assessment: A learning analytics ap-
proach on the impact of learners’ autonomy on performance, response-time and effort. Computers &
Education.
Papamitsiou, Z., Economides, A. A., Pappas, I. O., & Giannakos, M. N. (2018). Explaining learning per-
formance using response-time, self-regulation and satisfaction from content: An fsQCA approach.
8th International Conference on Learning Analytics and Knowledge (LAK18), 181–190. https://doi.
org/10.1145/3170358.3170397
Pintrich, P. R. (2000). The role of goal orientation in self-regulated learning. In M. Boekaerts, P. R. Pintrich
& M. Zeidner (Eds.), Handbook of self-regulation (pp. 451–502). San Diego, CA: Academic Press. https://doi.
org/10.1016/B978-012109890-2/50043-3
Pintrich, P. R. (2003). A motivational science perspective on the role of student motivation in
learning and teaching contexts. Journal of Educational Psychology, 95(4), 667–686. https://doi.
org/10.1037/0022-0663.95.4.667
Pintrich, P. R., & Schunk, D. H. (1996). Motivation in education: Theory, research, and application. Englewood
Cliffs, NJ: Prentice-Hall Inc.
Pintrich, P. R., Smith, D. A. F., Garcia, T., & Mckeachie, W. J. (1993). Reliability and predictive validity of the
Motivated Mtrategies for Learning Questionnaire (MSLQ). Educational and Psychological Measurement,
53(3), 801–813.
Puustinen, M. (1998). Help-seeking behavior in a problem-solving situation: Development of self-regula-
tion. European Journal of Psychology of Education, 13(2), 271. https://doi.org/10.1007/BF03173093
Reeve, J. (2006). Teachers as facilitators: What autonomy-supportive teachers do and why their students
benefit. The Elementary School Journal, 106(3), 225–236. https://doi.org/10.1086/501484
© 2019 British Educational Research Association
 Effects of self-regulation on autonomous learning    17

Reeve, J. (2009). Why teachers adopt a controlling motivating style toward students and how they
can become more autonomy supportive. Educational Psychologist, 44(3), 159–175. https://doi.
org/10.1080/00461520903028990
Reeve, J., Ryan, R., Deci, E. L., & Jang, H. (2008). Understanding and promoting autonomous self-regula-
tion: A self-determination theory perspective. In D. H. Schunk & B. J. Zimmerman (Eds.), Motivation and
self-regulated learning: Theory, research, and applications (pp. 223–244). Mahwah, NJ: Lawrence Erlbaum
Associates Publishers.
Ryan, A. M., Hicks, L., & Midgley, C. (1997). Social goals, academic goals, and avoiding seeking help in the class-
room. The Journal of Early Adolescence, 17(2), 152–171. https://doi.org/10.1177/0272431697017002003
Sarstedt, M., Ringle, C. M., & Hair, J. F. (2017). Partial least squares structural equation modeling. In C.
Homburg, M. Klarmann, & A. Vomberg (Eds.), Handbook of market research (pp. 1–40). Heidelberg: Springer.
Schuitema, J., Peetsma, T., & van der Veen, I. (2016). Longitudinal relations between perceived autonomy
and social support from teachers and students’ self-regulated learning and achievement. Learning and
Individual Differences, 49, 32–45. https://doi.org/10.1016/j.lindif.2016.05.006
Sierens, E., Vansteenkiste, M., Goossens, L., Soenens, B., & Dochy, F. (2009). The synergistic relationship of
perceived autonomy support and structure in the prediction of self-regulated learning. British Journal of
Educational Psychology, 79(1), 57–68. https://doi.org/10.1348/000709908X304398
Stefanou, C., Stolk, J. D., Prince, M., Chen, J. C., & Lord, S. M. (2013). Self-regulation and autonomy in
problem- and project-based learning environments. Active Learning in Higher Education, 14(2), 109–122.
https://doi.org/10.1177/1469787413481132
Stone, M. (1974). Cross-validatory choice and assessment of statistical predictions. Journal of the Royal
Statistical Society. Series B (Methodological), 36(2), 111–147.
Sun, J. C.-Y., & Rueda, R. (2012). Situational interest, computer self-efficacy and self-regulation: Their
impact on student engagement in distance education. British Journal of Educational Technology, 43(2),
191–204. https://doi.org/10.1111/j.1467-8535.2010.01157.x
Tenenhaus, M., Vinzi, V. E., Chatelin, Y.-M., & Lauro, C. (2005). PLS path modeling. Computational Statistics
& Data Analysis, 48(1), 159–205.
Terzis, V., & Economides, A. A. (2011). The acceptance and use of computer based assessment. Computers &
Education, 56(4), 1032–1044. https://doi.org/10.1016/j.compedu.2010.11.017
Tsai, C.-W., Shen, P.-D., & Tsai, M.-C. (2011). Developing an appropriate design of blended learning
with web-enabled self-regulated learning to enhance students’ learning and thoughts regarding on-
line learning. Behaviour & Information Technology, 30(2), 261–271. https://doi.org/10.1080/01449
29X.2010.514359
Vansteenkiste, M., Simons, J., Lens, W., Sheldon, K. M., & Deci, E. L. (2004). Motivating learning, performance,
and persistence: The synergistic effects of intrinsic goal contents and autonomy-supportive contexts.
Journal of Personality and Social Psychology, 87(2), 246–260. https://doi.org/10.1037/0022-3514.87.2.246
Vansteenkiste, M., Sierens, E., Goossens, L., Soenens, B., Dochy, F., Mouratidis, A., … Beyers, W. (2012).
Identifying configurations of perceived teacher autonomy support and structure: Associations with
self-regulated learning, motivation and problem behavior. Learning and Instruction, 22(6), 431–439.
https://doi.org/10.1016/j.learninstruc.2012.04.002
Wang, C.-H., Shannon, D. M., & Ross, M. E. (2013). Students’ characteristics, self-regulated learning, tech-
nology self-efficacy, and course outcomes in online learning. Distance Education, 34(3), 302–323. https://
doi.org/10.1080/01587919.2013.835779
Wolters, C. A. (2003). Regulation of motivation: Evaluating an underemphasized aspect of self-regulated
learning. Educational Psychologist, 38(4), 189–205.
Xu, D., & Jaggars, S. S. (2014). Performance gaps between online and face-to-face courses: Differences
across types of students and academic subject areas. The Journal of Higher Education, 85(5), 633–659.
https://doi.org/10.1353/jhe.2014.0028
Zimmerman, B. J. (2000). Attaining self-regulation: A social cognitive perspective. In M. Boekaerts, P.
R. Pintrich & M. Zeidner (Eds.), Handbook of self-regulation (pp. 13–39). San Diego, CA: Academic Press.
https://doi.org/10.1016/B978-012109890-2/50031-7
Zimmerman, B. J. (2002). Becoming a self-regulated learner: An overview. Theory Into Practice, 41(2),
64–70.
© 2019 British Educational Research Association
18  British Journal of Educational Technology  Vol 0 No 0 2019

Appendix

Table A1:  Constructs and items from the questionnaires

Construct Items Description

Effort-Regulation (ER)— ER1 I work hard to do well in this class even if I don’t like what we
MSLQ; Pintrich et al. are doing
(1993) ER2 Even when course materials are dull and uninteresting, I
manage to keep working until I finish
ER3 I often feel so lazy or bored when I study for this class that I
quit before I finish what I planned to do
Goal Expectancy (GE) GE1 Courses’ preparation was sufficient for the test
– CBAAM; Terzis and GE2 My personal preparation for the test was sufficient
Economides (2011) GE3 My performance expectations for the test
Help-Seeking (HS)— HS1 I am persistent in getting help from the instructor through
OSLQ; Barnard et al. HS2 e-mail
(2009) I find someone who is knowledgeable in course content so
that I can consult with him or her when I need help
HS3 I share my problems with my classmates online so we know
what we are struggling with and how to solve our problems
Time-Management TM1 I allocate extra studying time for my online courses because I
(TM)—OSLQ; Barnard know it is time-demanding
et al. (2009) TM2 I try to schedule the same time every day or every week to
study for my online courses, and I observe the schedule
TM3 Although we don’t have to attend daily classes, I still try to
distribute my studying time evenly across days

© 2019 British Educational Research Association