Content uploaded by Andrew J Martin
Author content
All content in this area was uploaded by Andrew J Martin on Aug 17, 2021
Content may be subject to copyright.
Frontiers in Psychology | www.frontiersin.org 1 August 2021 | Volume 12 | Article 702163
ORIGINAL RESEARCH
published: 17 August 2021
doi: 10.3389/fpsyg.2021.702163
Edited by:
Sina Fackler,
Leibniz Institute for Educational
Trajectories (LG), Germany
Reviewed by:
Fotini Polychroni,
National and Kapodistrian University
of Athens, Greece
Hongbiao Yin,
The Chinese University of Hong Kong,
China
*Correspondence:
Andrew J. Martin
andrew.martin@unsw.edu.au
Specialty section:
This article was submitted to
Educational Psychology,
a section of the journal
Frontiers in Psychology
Received: 29 April 2021
Accepted: 05 July 2021
Published: 17 August 2021
Citation:
Martin AJ, Collie RJ and
Nagy RP (2021) Adaptability and
High School Students’ Online
Learning During COVID-19: A Job
Demands-Resources Perspective.
Front. Psychol. 12:702163.
doi: 10.3389/fpsyg.2021.702163
Adaptability and High School
Students’ Online Learning
During COVID-19: A Job
Demands-Resources Perspective
AndrewJ.Martin*, RebeccaJ.Collie and RobinP.Nagy
School of Education, University of New South Wales, Sydney, NSW, Australia
The present study investigated the role of adaptability in helping high school students
navigate their online learning during a period of COVID-19 that entailed fully or partially
remote online learning. Drawing on Job Demands-Resources theory and data from a
sample of 1,548 Australian high school students in nine schools, weexamined the role
of adaptability in predicting students’ online learning self-efcacy in mathematics and
their end of year mathematics achievement. It was found that beyond the effects of
online learning demands, online and parental learning support, and background
attributes, adaptability was signicantly associated with higher levels of online learning
self-efcacy and with gains in later achievement; online learning self-efcacy was also
significantly associated with gains in achievement—and significantly mediated
the relationship between adaptability and achievement. These ndings conrm the
role of adaptability as an important personal resource that can help students in
their online learning, including through periods of remote instruction, such as during
COVID-19.
Keywords: adaptability, job demands-resources, online learning, remote instruction, COVID-19, achievement
INTRODUCTION
e COVID-19 pandemic led to an unexpected and rapid shi to remote learning for
students around the world. In the space of a few weeks, the very nature of learning and
instruction was transformed (Australian Academy of Science, 2020). Learning and instruction
moved to remote online modes at speed and scale. e extent to which students have
successfully responded and adjusted to these disruptions has been key to how they have
coped academically (Australian Academy of Science, 2020). is being the case, adaptability
may be a personal attribute that is highly relevant through times of online remote
learning and instruction, such as during COVID-19 and any other future periods of
disrupted learning.
Adaptability is the capacity to regulate one’s behaviors, thoughts, and feelings in response
to novel, variable, uncertain, and unexpected situations and circumstances (Martin et al.,
2012, 2013). Adaptability has been identied as an important capacity for students’ academic
and personal development, including their motivation, engagement, achievement, and social-
emotional wellbeing (Martin etal., 2013; Holliman et al., 2018, 2019, 2021). Given adaptability
Martin et al. Adaptability, Online Learning, and COVID-19
Frontiers in Psychology | www.frontiersin.org 2 August 2021 | Volume 12 | Article 702163
is specically aimed at successfully navigating change,
uncertainty, and novelty, it is also likely a vital personal attribute
to support students during periods of novelty, variability, and
uncertainty, such as with COVID-19 restrictions and lockdowns,
including periods of online learning through these times. To
the extent that adaptability is associated with positive educational
processes and outcomes during online learning, it may be an
important area of focus for educational interventions.
e aim of this research was to expand current knowledge
of adaptability by focusing on its role in students’ academic
development and online learning during a period of COVID-19
that entailed fully or partially remote online learning. Drawing
on Job Demands-Resources theory (JD-R theory; Bakker and
Demerouti, 2017, 2018) and focusing on learning and instruction
in mathematics, we examined the role of adaptability in
predicting students’ online learning self-ecacy and their end
of year achievement. We were particularly interested in the
extent to which adaptability (a personal resource) played a
role in students’ online learning self-ecacy and achievement
beyond the eects of any online learning demands,
online and parental learning support, and background
attributes. Figure 1 demonstrates the hypothesized model
under examination.
THEORETICAL BACKGROUND AND
LITERATURE REVIEW
Adaptability
As described above, adaptability is the capacity to adjust
behaviors, thoughts, and feelings in response to novel, variable,
uncertain, and unexpected situations and circumstances (Martin
et al., 2012, 2013). It is thus a tripartite perspective composed
of behavioral, cognitive, and emotional dimensions (Martin
et al., 2012, 2013). Research among school students has
demonstrated links between adaptability and students’
engagement and achievement (Martin et al., 2012, 2013; Collie
etal., 2017), identied the role of adaptability in young people’s
responses to climate change (Liem and Martin, 2015),
demonstrated the role of adaptability in reducing students’
failure dynamics (Martin etal., 2015), shown links with university
students’ engagement and longer-term achievement (Holliman
etal., 2018), and validated adaptability across diverse international
contexts (Martin etal., 2017). ere is, then, a strong evidence
base for the role of adaptability in students’ positive development.
e present study is an opportunity to expand on this by
investigating the role of adaptability in assisting students’ online
learning experiences and outcomes during a period of substantial
JOB DEMANDS
-Online Learning Barriersin Math
BUFFERING EFFECT
- Adaptability x Online Barriers in
Math
ONLINE LEARNING
SELF-EFFICACY IN
MATH
PERFORMANCE
- End of Year Math Test
Achievement
JOB RESOURCES
-Online Learning Supportin Math
JOB RESOURCES
-Parent/Home Help in Math
PERSONAL RESOURCES
-Adaptabilityin Math
BACKGROUND
ATTRIBUTES
- Age
- Gender (M/FM)
- Parent Education
- Non-English Speak B’ground
- Math Self-efficacy
- Prior Test Achievement
FIGURE1 | Hypothesized JD-R process in online mathematics.
Martin et al. Adaptability, Online Learning, and COVID-19
Frontiers in Psychology | www.frontiersin.org 3 August 2021 | Volume 12 | Article 702163
novelty, variability, and uncertainty—specically, online learning
during the COVID-19 pandemic. Given there is likely to
be substantial novelty, variability, and uncertainty ahead due
to the evolving nature of the pandemic (Australian Academy
of Science, 2020), it is important to identify modiable psycho-
behavioral attributes that may assist students through this and
through future periods of disrupted learning. e present study
focuses on adaptability as one such attribute.
Online Learning and Instruction
Online learning encompasses the use of desktop computers,
laptops, tablets, virtual reality devices, mobile phones, personal
digital assistants, and more (Sung etal., 2017). Online learning
methods traverse staged programs of instruction, animation,
gaming, simulations, video instruction, collaborative documents,
chatrooms, etc. ere are also many content and learning
management systems (e.g., Canvas, Moodle, and Blackboard)
that facilitate online learning. Online learning activity
predominantly comprises synchronous instruction that is in
real-time (such as live video interaction) and asynchronous
instruction that may bepre-recorded or a standalone self-paced
online program (alheimer, 2017).
When appraising the eectiveness of online learning, there
is a mixed evidence base. On the positive side, there is meta-
analytic evidence demonstrating the eectiveness of various
online learning approaches, yielding generally small to moderate
eect sizes (Yuwono and Sujono, 2018). ere is also meta-
analytic evidence that mobile-computer-supported learning can
enhance collaborative learning (Sung et al., 2017). On the
negative side, there is research suggesting that online learning
approaches are not as eective as real-time in-class learning.
For example, Clinton (2019); see also Delgado et al. (2018)
found that students reading material in paper-based form
showed greater comprehension than students reading the same
material in digital form. Findings from PISA 2012 (Peña-López,
2015) found that students who used computers very frequently
at school performed more poorly than students with other
levels of computer use. Moreover, it seems that many teachers
are not highly trained in harnessing technology to help students
learn (Peña-López, 2015). ere is also a line of research
demonstrating generally null or minimal eects when comparing
online and in-class modes. In an online coaching program
for teachers, there were no signicant eects for student
achievement (Kra and Hill, 2020). In a study of online distance
education, Cavanaugh et al. (2004) found comparable student
achievement across online and in-class instructional modes.
One reason why there are such mixed ndings is because
there are many factors that are implicated in the success of
online modes. Factors related to technology access, technology
skills, instructional and resource quality, parent/home support,
ethnicity, socioeconomic status, and learning support needs
have all been identied as inuencing the extent to which
online learning is eective or not (AITSL, 2020; Australian
Academy of Science, 2020). Importantly, however, given the
substantial novelty, variability, and uncertainty associated with
online learning during COVID-19, it is also likely that various
personal psychological attributes have potential to assist
students’ learning during this time and in future periods of
disrupted learning. Adaptability is hypothesized as one such
factor and is the focus of our investigation into online learning
experiences during a period of COVID-19 in Australia when
students were variously engaged in fully or partially remote
online learning.
Job Demands-Resources Theory
We draw on JD-R as a means to explore and understand the
role of adaptability in students’ online learning experiences
during COVID-19. Before doing so, we summarize JD-R as
traditionally formulated in workplace research. en,
we extrapolate from this to explore its relevance to students’
online learning and to frame the present study.
Job Demands-Resources Theory in the
Workplace
Job Demands-Resources theory holds that there are specific
contextual factors in jobs and work roles that help or hinder
employees’ outcomes (Schaufeli and Bakker, 2004). Job
demands are aspects of work that require psychological or
physical exertion (e.g., performing under a heavy workload
and addressing mounting deadlines) and that are linked
with psychological or physical costs (e.g., poor mental and
physical health aspects of burnout; Bakker and Demerouti,
2017; Collie etal., 2020a). Job resources are aspects of work
that help employees attain desired work-related goals and
growth (e.g., peer support; Demerouti etal., 2001) and are
linked with positive outcomes (e.g., motivation and health;
Skaalvik and Skaalvik, 2018).
In recent years, JD-R theory has recognized that there are
also personal resources that determine employees’ work-related
functioning (Xanthopoulou et al., 2007; Collie et al., 2020a).
Personal resources are modiable, personal capacities that reect
an individual’s potential to inuence their working environment;
similar to job resources, personal resources are linked with
positive outcomes (Schaufeli and Taris, 2014). Collie et al.
(2020a); see also Granziera etal. (2021) proposed that adaptability
can be considered a personal resource, as it is a modiable
capacity that can help an individual navigate change in the
workplace and eect positive outcomes.
In addition to these “main eects” of demands and resources,
there is also a buering possibility suggested by JD-R theory
(Bakker and Demerouti, 2017)—and adaptability may be an
important part of this. For example, Granziera et al. (2021)
proposed that adaptability may buer the negative eects of
job demands such that employees high in adaptability are less
likely to experience the negative eects of job demands. Granziera
et al. (2021) demonstrated support for this by showing that
adaptability oset the negative eect of role conict on emotional
exhaustion in teachers (see also Dicke et al., 2018).
Alongside the need to consider potential buering eects,
we also draw attention to more recent renements of JD-R
theory that speak to how demands and resources may
be perceived dierently by individuals: A given job demand
or job resource may beperceived in dierent ways by dierent
Martin et al. Adaptability, Online Learning, and COVID-19
Frontiers in Psychology | www.frontiersin.org 4 August 2021 | Volume 12 | Article 702163
people—not all individuals perceive a demand as a hindrance
and not all individuals perceive a resource as a help
(Bakker and Demerouti, 2017; Yin et al., 2018). is may
bethe case for numerous reasons, such as the level of control
individuals have in their role, the prestige of their role, the
extent to which the demand benets them, etc. (Bakker and
Demerouti, 2017). is being the case, we remain open to
the possibility that demands and/or resources may have
apparently counter-intuitive eects.
JD-R and Learning and Instruction
Although JD-R is centered on workplace processes, it is evident
the same factors and processes implicated in workplace
functioning are implicated in students’ learning. ere are
specic contextual factors in academic learning that help or
hinder students’ educational outcomes (Martin and Marsh,
2009). is being the case, job demands in the educational
setting refer to aspects of learning that require psychological
or physical exertion (e.g., performing under a heavy study
load and meeting multiple due dates) and are linked with
psychological or physical costs (e.g., stress, dropout, and
underachievement). Correspondingly, job resources in the
educational setting are aspects of learning that help students
attain desired academic goals and growth (e.g., teacher/
instructional support) and are linked with positive outcomes
(e.g., engagement and achievement). In relation to personal
resources, in line with Collie et al. (2020a), adaptability can
be considered a modiable capacity that can help students
navigate change and eect positive learning outcomes. Indeed,
there may also be a buering role for adaptability in the
learning context such that adaptable students may beless likely
to experience the negative eects of job demands.
us, although JD-R theory is a well-established approach
for understanding employees’ workplace functioning (Bakker and
Demerouti, 2017), wepropose it can also be applied to learning
and instruction. Moreover, although there is substantial research
harnessing JD-R to investigate teachers’ workplace experiences
(e.g., Collie etal., 2020a; Granziera etal., 2021), there is signicant
scope for investigating the same dynamics among school students.
Demands and Resources in the Present Study
In addition to our focus on adaptability (as a personal resource),
our study comprised one job demand and two job resources.
e job demand, online learning barriers, refers to the challenges
that students experience when learning online at home. It is
well documented that factors, such as unreliable internet,
diculties accessing appropriate computing and technology,
and distracting home environments, present barriers to students’
online learning (Peña-López, 2015; Australian Academy of
Science, 2020). In relation to job resources, online learning
support refers to the quality of the online learning resources
and learning opportunities made available to students by their
schools (Yukselturk and Bulut, 2007; Means etal., 2009; Escueta
et al., 2017; Gregori et al., 2018; AITSL, 2020). e other job
resource is parent/home help, which refers to the extent to
which parents provide help with schoolwork and the necessary
routines and resources are available at home to assist learning
(Galpin and Taylor, 2018).
Although wehypothesize that online learning barriers (job
demand) will yield negative eects and that online learning
support and parent/home help (job resources) will yield
positive eects, we are open to the possibility that this may
not be so—in keeping with recent developments in JD-R
theory stating that there is variability between individuals in
how they perceive demands and resources (Bakker and
Demerouti, 2017; Yin etal., 2018; Han etal., 2020). Indeed,
recent research by Martin et al. (2021) showed that students
in high school science perceive and experience a dicult
task in dierent ways, some seeing it as a challenge and
some seeing it as a threat. In the case of the present study
we might ask, at what point does parent/home help move
from being supportive (yielding a positive motivational eect)
to being controlling (yielding a negative motivational eect;
Neubauer et al., 2020)?
In terms of JD-R’s contended buffering eect, we can model
the interaction between adaptability and online learning barriers
to ascertain the extent to which adaptability may moderate
the negative eects of job demands (Collie et al., 2020a;
Granziera et al., 2021). ese factors are all demonstrated in
Figure 1 as key predictors of student outcomes that take the
forms of online learning self-ecacy and end of year test
achievement—links now discussed.
Linking the Resources and Demands With Online
Learning Self-Efcacy
Collie et al. (2020a) argued that the nature of individuals’
demands and resources impacts their domain-specic ecacy,
which in turn impacts important outcomes, such as
performance. Online learning self-efficacy refers to students’
perceived and experienced competence in online learning. A
large body of research has demonstrated the importance of
perceived ecacy for a range of outcomes, including
performance (e.g., Bandura, 1997; Martin, 2007, 2009; Klassen
and Chiu, 2010; Marsh and Martin, 2011). In JD-R models,
the positioning of ecacy can dier, with some models placing
it as a personal resource alongside job demands and resources
(e.g., Xanthopoulou etal., 2007), while others having ecacy
predicted by demands and resources—but notably still referring
to it as a personal resource (Collie et al., 2020a). We adopt
the latter position because (in line with Collie et al., 2020a)
we wanted to focus on what demands and resources lay a
foundation for online learning self-ecacy given it is a desirable
outcome in itself (as well as being a means to desirable ends,
such as achievement; Collie et al., 2020a). Indeed, other
researchers have also identied perceived ecacy as an outcome
of job demands, job resources, and other personal resources
(e.g., Chang, 2013).
Of particular interest in our research is the role of adaptability
in predicting online learning self-ecacy. According to Collie
et al. (2020a); see also Collie and Martin (2016), adaptability
fosters mastery and ecacy experiences—and their research
among teachers demonstrated precisely this. Accordingly,
we hypothesize that adaptability during times of such
Martin et al. Adaptability, Online Learning, and COVID-19
Frontiers in Psychology | www.frontiersin.org 5 August 2021 | Volume 12 | Article 702163
uncertainty, variability, and novelty (viz. online learning during
COVID-19) will be associated with higher levels of online
learning self-ecacy. In addition to this, we suggest that the
presence of online learning barriers (job demands) will lead
to lower online learning self-ecacy, whereas job resources
in the forms of online learning support and parent/home
help will be associated with higher online learning
self-ecacy.
Achievement as an Outcome of Online Learning
Self-Efcacy
In most JD-R models, workplace outcomes reflected in
diverse forms of performance (e.g., retention and achievement)
are the final part of the process (though, the process is
cyclical over time; Collie et al., 2020a). Extrapolating to
learning and instruction processes under a JD-R framework,
academic achievement is contended as an analogous
performance outcome (see Figure 1). Thus, the final part
of the process examined in our hypothesized model considers
the association between online learning self-efficacy and
subsequent achievement. This component is also supported
by conceptualizing from social cognitive theory (Bandura,
1997) and supported by a long line of empirical research
in education (Martin, 2007, 2009; Lee et al., 2014; Schunk
and DiBenedetto, 2014). Wetherefore hypothesize a positive
link between online learning self-efficacy and achievement.
Moreover, given our focus on adaptability as a predictor
of online learning self-efficacy, wealso explore the indirect
association between adaptability and achievement via online
learning self-efficacy.
Mathematics: The Subject Area for This
Investigation
For several reasons, mathematics was our focus for this
investigation. ere is evidence of declining achievement
and participation in high school mathematics in Australia
(e.g., omson et al., 2016; OECD, 2018). ere are also
concerns that rst-year university STEM students are not
suciently prepared for the level of mathematics skill required
at the tertiary level (Nicholas et al., 2015). It is also the
case that students can struggle with online formats in
mathematics. For example, when assessing online and paper-
based tests, Backes and Cowan (2019) found paper-based
tests yielded higher mathematics results than online tests.
Hassler Hallstedt et al. (2018) found that engaging with a
mathematics program on a tablet yielded a small positive
eect size for basic arithmetic, but not for arithmetic transfer
and problem solving; they also found the positive eects
faded over the course of 6 to 12 months. Notwithstanding
this, other research has found more positive evidence for
online mathematics learning (e.g., Sung et al., 2017). Taken
together, mathematics is an area of national priority and
one for which there is mixed evidence for eective instruction
in online modes. It is, thus, a potentially illuminating focus
for research investigating factors that may assist students’
online learning experiences.
The Role of Salient Background Attributes
In assessing the unique eects of demands and resources, it
is important to account for the following background attributes
(covariates) that are known to be associated with one or more
of this study’s substantive variables: age, gender, language
background, parent education, mathematics self-ecacy, and
prior mathematics achievement. Older students seem to achieve
more highly in technology-assisted learning (Escueta et al.,
2017; Sung et al., 2017). Girls tend to score higher in the
self-regulatory attributes (Martin, 2007) important for self-
directed/autonomous remote online learning (Kirschner and
De Bruyckere, 2017). Ethnicity has been found to moderate
the eects of online learning on achievement (Nguyen, 2015).
In periods of remote learning during COVID-19, parents have
struggled with the motivational and learning demands placed
on them (Garbe et al., 2020) and unfamiliarity with these
processes may be greater for parents with fewer years of
education themselves. Online learning self-ecacy and
achievement in mathematics are likely to be associated with
self-ecacy in mathematics more generally (not just in its
online aspects) and also with prior mathematics achievement
(e.g., Hattie, 2009).
AIMS OF THE PRESENT STUDY
Drawing on JD-R theory and set during a period of COVID-19
entailing fully or partially remote online learning, this research
investigated the role of adaptability in high school students’
online learning self-ecacy in mathematics and their end of
year mathematics achievement. Following our review of theory
and prior research, we pose numerous hypotheses and a
research question. Hypothesis 1: beyond the eects of online
learning demands, online and parental learning support, and
background attributes, adaptability will bepositively associated
with students’ online learning self-ecacy and gains in end
of year achievement. Hypothesis 2: beyond the eects of
adaptability, online learning demands, online and parental
learning support, and background attributes, online learning
self-ecacy will be positively associated with gains in end
of year achievement. Hypothesis 3: online learning self-ecacy
will signicantly mediate the relationship between adaptability
and gains in end of year achievement. Research Question 1:
what is the role of adaptability in buering the potentially
negative eects of online learning barriers.
MATERIALS AND METHODS
Participants
The sample comprised 1,548 Australian high school students
from nine schools. All schools were in the independent
school sector and located in or around major urban areas
of the state of New South Wales (NSW) on the east coast
of Australia. Of the nine schools, four were co-educational,
two were single-sex boys’ schools, and three were single-sex
girls’ schools. Just over half (53%) of students were boys.
Martin et al. Adaptability, Online Learning, and COVID-19
Frontiers in Psychology | www.frontiersin.org 6 August 2021 | Volume 12 | Article 702163
Students were in Year 7 (21%), Year 8 (34%), Year 9 (17%),
and Year 10 (28%)—the first 4 years of high school in
Australia. The average age was 13.77years (SD=1.16years).
Fourteen percent of students spoke a language other than
English at home. Students tended to be from educated
backgrounds, with parents/carers scoring 5.19 (SD = 1.77)
on a scale of 1 (no formal education) to 6 (university
education).
Procedure
e lead researcher’s university provided human ethics approval.
School principals then provided approval for their school’s
participation. Subsequently, parents/carers and participating
students provided consent. An online survey and mathematics
test were administered during school hours in the second term
(of four school terms) of 2020. As described in the introduction,
this was during a period of COVID-19 that entailed fully or
partially remote online learning. e end of year online
mathematics test was administered in the nal term of 2020
when all students had returned to school for in-class lessons.
Students were asked to complete the survey and tests on
their own.
Materials
Our substantive factors included job demands, job resources,
personal resources, online learning self-ecacy, and performance.
Descriptive, reliability, and factor analytic statistics are presented
in Tabl e 1. Wealso assessed background attributes as covariates,
comprising age, gender, parent education, and
language background.
Job Demands, Resources, and Outcomes
JD-R factors comprised job demands (online learning barriers),
job resources (online learning support, parent/home help),
personal resources (adaptability), a buering factor (online
demands x adaptability), ecacy (online learning self-ecacy),
and performance (end of year achievement test)—all in relation
to mathematics. Descriptive and measurement statistics are
shown in Tab l e 1. Online learning barriers were a formative
sum (from 0 to 3) representing the accumulation of barriers
to students’ online learning at home, including unreliable
Internet, inadequate computing/technology, and little/no access
to a quality area for concentration. Online learning support
comprised ve items asking students about the quality of
support/resourcing for their online learning (e.g., “How satised
are youwith your online learning platform for mathematics?”),
rated on a scale from 1 (very dissatised) to 5 (very satised).
Because the nature of online learning elements (e.g., online
learning platforms, such as learning management systems) can
be quite variable (Tinmaz and Lee, 2020)—e.g., qualitative
responses in the present study revealed more than 20 online
learning platforms were used—a given online learning element
may not necessarily bea resource per se. us, to better ensure
wewere assessing it as a resource, weasked students to appraise
the resource via ratings of satisfaction. While weacknowledge
resources under JD-R are oen assessed in terms of the
characteristics or attributes of the resource, we adapted this
to assess it in a more nuanced and targeted fashion to establish
it more clearly as a resource. In fact, the idea to tap into
appraisals of job demands and resources is now being recognized,
with researchers suggesting it is only then that the help or
hindrance dimension of a job resource/demand can beassessed
(Liu and Li, 2018; Ma etal., 2021). Parent/home help comprised
ve items asking about the help they received at home for
their learning (e.g., “How oen do your parents or someone
else in your home help youwith your mathematics homework?”),
rated on a scale of 1 (never/hardly ever) to 5 (every day/
almost every day). Adaptability comprised three items (using
the Adaptability Scale—Short; Martin etal., 2016) asking students
about the extent to which they could adjust their behavior,
thinking, and emotion to eectively navigate novelty, variability,
and uncertainty (e.g., “In mathematics, to assist me in a new
situation, I am able to change the way I do things”), rated
on a scale of 1 (strongly disagree) to 7 (strongly agree). Buffering
was assessed via the interaction of online learning demands
and adaptability (an interaction term generated through the
cross-product of the two zero-centered main eects;
Aiken et al., 1991).
Online learning self-efficacy was a single item asking students
about their perceived competence in online learning (“Overall,
how condent are you as an online learner in mathematics?”),
rated on a 1 (not condent) to 4 (very condent) scale. Given
this was a single-item factor, we sought to account for
measurement error by creating an error-adjusted score using
the following equation: s
h
2
× (1 −ωh), where s
h
2
is the
variance of our online learning self-ecacy variable (0.827)
and ωh was the reliability of the same variable (Cole and
Preacher, 2014; Kline, 2016), which weconservatively estimated
TABLE1 | Descriptive and measurement statistics.
Possible
range M SD Reliability
(omega)
CFA
loading M
Online
learning
barriers (job
demands)
0–3 0.217 0.476 – –
Online
learning
support (job
resources)
1–5 3.711 0.708 0.795 0.659
Parent/
home help
(job
resources)
1–5 2.678 0.856 0.751 0.612
Adaptability
(personal
resources)
1–7 5.471 1.054 0.800 0.749
Online
learning
self-efcacy
1–4 2.888 0.910 0.700†0.837
End of year
test
achievement
0–10 5.745 1.948 – –
All measures are in relation to mathematics; M, mean; SD, standard deviation; CFA,
conrmatory factor analysis; dash, formative score/single-item indicator.†reliability
estimated for this single item indicator and used to generate error-adjusted score.
Martin et al. Adaptability, Online Learning, and COVID-19
Frontiers in Psychology | www.frontiersin.org 7 August 2021 | Volume 12 | Article 702163
at 0.70 in this study. In so doing, unreliability was accounted
for in this factor, as would be the case if we had multiple
items and estimated a latent factor. is error-adjusted score
was used in the conrmatory factor analysis (CFA) and structural
equation modeling (SEM; described below). End of year
achievement was assessed via a 10-item mathematics test and
operationalized as a formative summed score. Achievement
scores were standardized by year level (M = 0; SD = 1).
Questions were structured in 4-answer multiple-choice format,
graduated in diculty and designed to assess underlying
mathematical competencies (as opposed to knowledge recall)
from the Australian National Curriculum (Kindergarten-10),
and associated state syllabus outcomes (e.g., addition, subtraction,
patterns, algebra, time, fractions, decimals, percentages, ratio,
probability, and area). An example question was “Which of
the following is correct? (A: 0.0409 > 0.041, B: 0.21 > 0.200,
C: 0.00004 > 0.0003, and D: 0.123 > 0.124),” to assess a part
of the syllabus material covering decimals, fractions,
and percentages.
Background Attributes
In assessing the unique eects of demands and resources, it
is important to account for numerous background attributes
in modeling. For these background attributes, participants
reported age (a continuous measure), gender (0 = male and
1 = female), language background (0= English speaking and
1 = non-English speaking), and parent education (scale from
1=no formal education to 6=university education). Descriptive
statistics for these are presented in Participants section, above.
We also assessed mathematics self-ecacy (single item from
the domain-specic version of the Motivation and Engagement
Scale High School Short, Martin, 2020; validated by Martin
et al., 2020): “I believe I can do well in mathematics” rated
(1 = strongly disagree to 7 = strongly agree; M = 5.40,
SD = 1.94) and prior achievement (10-item mathematics test
parallel to the end of year test described above; M = 5.52,
SD = 1.86).
Data Analysis
Conrmatory factor analysis and SEM were the central analyses,
conducted with Mplus version 8 (Muthén and Muthén, 2017).
We used the MLR (maximum likelihood robust to
non-normality) estimator that provides parameter estimates
with standard errors and a chi-square test statistic that are
robust to non-normality (Muthén and Muthén, 2017). To
assess model t, a Comparative Fit Index (CFI) and Tucker
Lewis Index (TLI) greater than 0.90, a Root Mean Square
Error of Approximation (RMSEA) and Standardized Root
Mean Square Residual (SRMR) less than 0.08 indicated
acceptable t (Hu and Bentler, 1999; Kline, 2016). Missing
data were dealt with using the Mplus default, Full Information
Maximum Likelihood (FIML; Arbuckle, 1996).
For the CFA, the following factors were included:
online learning barriers (formative score), online learning
support (latent factor), parent/home help (latent factor),
adaptability (latent factor), online learning self-efficacy
(error-adjusted score), end of year achievement (formative
summed score), and background attributes (each a single
indicator, with loading set at 1.00 and residual at 0)—thus,
a 12-factor CFA.
The hypothesized structural model (Figure 1) was tested
using SEM. In this model, (a) online learning demands,
online learning support, parent/home help, adaptability, the
interaction of online demands and adaptability (buffering
effect), and all background attributes predicted online learning
self-efficacy and in turn, (b) these factors—including online
learning self-efficacy—predicted end of year achievement
(thus, a “fully-forward” model). Because we included prior
achievement as a predictor in the model, wecould interpret
paths to end of year achievement in terms of gains (or
declines). Our data also enabled tests of indirect (mediation)
effects which were conducted in subsidiary analyses. A
parametric bootstrapping approach was used to test mediation.
Here, weexplored the extent to which online learning self-
efficacy mediated the relationship between the various
demands and resources and students’ end of year achievement.
Analyses were based on bootstrapped standard errors
with 1,000 draws (MacKinnon et al., 2002; Shrout and
Bolger, 2002).
RESULTS
Conrmatory Factor Analysis and
Correlations
e 12-factor CFA tested the dimensionality and measurement
properties underlying the hypothesized model and also generated
bivariate correlations that were the rst insight into the
relationships of interest in Figure 1. is CFA yielded an
acceptable t to the data, χ2 (152) = 453.25, p < 0.001,
CFI= 0.956, TLI=0.933, RMSEA=0.036, and SRMR =0.033.
Factor loading means are shown in Tabl e 1 and correlations
are presented in Table 2 . Here, wesummarize only signicant
correlations among substantive factors that are key to the
hypothesized model (all other signicant and non-signicant
correlations are in Table 2 ). e following were signicantly
correlated with online learning self-ecacy: online learning
barriers (r = −0.247, p < 0.001), online learning support
(r=0.689, p<0.001), parent/home help (r= 0.153, p<0.001),
and adaptability (r = 0.529, p < 0.001). us, online learning
barriers were associated with lower online learning self-ecacy,
whereas online learning support, parent/home help, and
adaptability were associated with higher online learning self-
ecacy. e following were signicantly correlated with end
of year achievement: online learning self-ecacy (r = 0.256,
p < 0.001), online learning barriers (r = −0.097, p < 0.001),
online learning support (r = 0.140, p < 0.001), parent/home
help (r = −0.090, p < 0.01), and adaptability (r = 0.272,
p < 0.001). us, online learning barriers and parent/home
help were associated with lower end of year achievement,
whereas online learning self-ecacy, online learning support,
and adaptability were associated with higher end of
year achievement.
Martin et al. Adaptability, Online Learning, and COVID-19
Frontiers in Psychology | www.frontiersin.org 8 August 2021 | Volume 12 | Article 702163
TABLE2 | Correlations from CFA.
1 2 3 4 5 6 7 8 9 10 11 12
JD-R factors
1. Online learning
barriers
–−0.239*** −0.058 −0.152*** −0.247*** −0.097*** 0.038 0.008 −0.044 0.024 −0.149*** −0.114***
2. Online learning
support
– 0.080*0.408*** 0.689*** 0.140*** 0.050 −0.054 0.031 0.018 0.258*** 0.167***
3. Parent/home
help
– 0.230*** 0.153*** −0.090** −0.177*** −0.077** 0.082** 0.040 0.166*** −0.096**
4. Adaptability – 0.529*** 0.272*** −0.111*** −0.200*** 0.063*0.047 0.556*** 0.263***
5. Online learning
self-efcacy
– 0.256*** −0.107** −0.098** 0.108** 0.047 0.406*** 0.235***
6. End of year
achievement
–−0.029 −0.097*** 0.174*** 0.144*** 0.308*** 0.561***
Background attributes
7. Age – 0.074** −0.019 0.039 −0.066** −0.012
8. Gender (M/FM) – −0.021 −0.060** −0.193*** −0.147***
9. Parent
education
– 0.030 0.116*** 0.161***
10. NESB – 0.061** 0.153***
11. Math self-
efcacy
– 0.309***
12. Prior
achievement
–
All JD-R factors are in relation to mathematics; NESB, non-English speaking background; M, male; and FM, female. *p<0.05, **p<0.01, and ***p<0.001.
Martin et al. Adaptability, Online Learning, and COVID-19
Frontiers in Psychology | www.frontiersin.org 9 August 2021 | Volume 12 | Article 702163
Structural Equation Modeling
We then tested the model in Figure1 using SEM. is yielded
an acceptable t to the data, χ2 (163) = 509.76, p < 0.001,
CFI=0.949, TLI=0.921, RMSEA=0.037, and SRMR=0.036.1
Tabl e 3 and Figure 2 show results. Here, we summarize only
signicant paths among substantive factors. All other signicant
and non-signicant paths are in Tabl e 3 . Signicant predictors
of online learning self-ecacy (beyond the eects of all
background attributes) were as follows: online learning demands
(β = −0.062, p < 0.05), online learning support (β = 0.562,
p < 0.001), and adaptability (β = 0.202, p < 0.001). us,
online learning demands were predictive of lower online learning
self-ecacy, whereas online learning support and adaptability
were predictive of higher online learning self-ecacy. In turn,
beyond the eects of background attributes, signicant predictors
of end of year achievement gains were as follows: online learning
self-ecacy (β=0.118, p<0.05), parent/home help (β=−0.103,
p<0.001), and adaptability (β = 0.079, p<0.05). us, online
learning self-ecacy and adaptability were predictive of gains
in end of year achievement, whereas parent/home help was
predictive of declines in end of year achievement (discussed
in further detail below).
Finally, we examined the indirect paths from demands and
resources to end of year achievement gains via online learning
self-ecacy. ere were two signicant indirect paths: online
learning support → online learning self-ecacy → end of year
achievement, β=0.066, p<0.05; adaptability → online learning
self-ecacy → end of year achievement, β= 0.024, p < 0.05.
us, online learning self-ecacy mediated the relationship
between online learning support and end of year achievement
gains; it also mediated the relationship between adaptability
and end of year achievement gains. Tabl e 3 also presents total
eects, showing that adaptability has the largest net positive
eect on achievement gains of all predictors (β = 0.103,
p<0.001), while parent/home help has the largest net negative
eect, being signicantly associated with achievement declines
(β = −0.100, p < 0.001).
DISCUSSION
Adaptability is a personal resource that has potential to assist
students through times of novelty, variability, and uncertainty—
such as what they have experienced during COVID-19. Drawing
on JD-R theory and a large sample of Australian high school
1
Because there were diverse modes of online mathematics instruction during
this period of COVID-19 (e.g., dierent combinations of class-based/online
learning, small group learning, and solo learning on any given school day),
students were not always nested in their mathematics classrooms. us, wedid
not statistically account for clustering/nesting in our analyses. For completeness,
however, when wetested the hypothesized model (Figure 1) using the Mplus
Type = Complex command (adjusting standard errors for the nesting of students
within classrooms), wederived good t (χ2 [163] = 493.38, p<0.001, CFI=0.948,
TLI = 0.919, RMSEA = 0.036, and SRMR = 0.036) and the same substantive
eects as our unadjusted model, with the minor exception that the online
learning barriers → online learning self-ecacy path was signicant at p< 0.10
(β = −0.062, p = 0.067), not p < 0.05.
students, we examined the role of adaptability (a personal
resource) in predicting students’ online learning self-ecacy
and the role of their online learning self-ecacy in predicting
their end of year achievement during a period of COVID-19
that entailed fully or partially remote online learning. Wefound
that adaptability was signicantly associated with greater online
learning self-ecacy and with gains in achievement (supporting
Hypothesis 1); online learning self-ecacy was also signicantly
associated with gains in achievement (supporting Hypothesis
2)—and signicantly mediated the relationship between
adaptability and achievement (supporting Hypothesis 3).
TABLE3 | Standardized direct and indirect effects for JD-R process in online
mathematics.
Online learning
self-efcacy
End of year test
achievement
β β
JD-R factors
Online learning barriers
(job demands)
−0.062* −0.012
Online learning support
(job resources)
0.562*** −0.072
Home/parent help (job
resources)
0.022 −0.103***
Adaptability (personal
resources)
0.202*** 0.079*
Adaptability × Barriers
(buffering)
−0.008 −0.001
Online learning self-
efcacy
– 0.118*
Background attributes
Age −0.101 −0.011
Gender (M/FM) 0.011 0.011
Parent education 0.053 0.079***
Non-English speaking
background
0.018 0.063**
Math self-efcacy 0.117** 0.095**
Prior achievement 0.039 0.463***
Indirect effects
Online learning barriers → Online learning self-
efcacy → End of year test achievement
−0.007
Online learning support → Online learning self-
efcacy → End of year test achievement
0.066*
Home/parent support → Online learning self-efcacy
→ End of year test achievement
0.003
Adaptability → Online learning self-efcacy → End of
year test achievement
0.024*
Adaptability × Barriers → Online learning self-efcacy
→ End of year test achievement
−0.001
Total effects
Online learning barriers → Online learning self-
efcacy → End of year test achievement
−0.020
Online learning support → Online learning self-
efcacy → End of year test achievement
−0.006
Home/parent support → Online learning self-efcacy
→ End of year test achievement
−0.100***
Adaptability → Online learning self-efcacy → End of
year test achievement
0.103***
Adaptability × Barriers → Online learning self-efcacy
→ End of year test achievement
−0.002
All JD-R factors are in relation to mathematics. *p<0.05, **p<0.01, and
***p<0.001.
Martin et al. Adaptability, Online Learning, and COVID-19
Frontiers in Psychology | www.frontiersin.org 10 August 2021 | Volume 12 | Article 702163
ese eects were signicant beyond any variance attributable
to online learning demands, online learning support, parent/
home help, and background attributes. Our ndings therefore
conrm the hypothesized role of adaptability as an important
personal resource and have practical implications for better
supporting students in their online learning, including through
periods of remote online instruction, such as during COVID-19.
Findings of Particular Note
In line with hypotheses, ndings showed that adaptability (a
personal resource) was signicantly associated with greater
online learning self-ecacy—beyond the eects of online learning
barriers (job demands), online learning support and parent/
home help (job resources), and background attributes. In fact,
adaptability not only predicted online learning self-ecacy as
hypothesized, but also directly predicted gains in end of year
test achievement—and signicantly indirectly predicted end of
year achievement via the mediating role of online learning
self-ecacy. Adaptability thus presents as an important factor
in how students navigate their online learning during periods
of signicant novelty, variability, and uncertainty (in this case,
during a period of COVID-19 that entailed fully or partially
remote online learning). We can infer that the adjustments
required by students to navigate these uncertain circumstances
were well met by the psychological attribute of adaptability.
is expands on the pre-COVID-19 evidence base for the
positive eects of adaptability on students’ educational outcomes
(Martin et al., 2013; Holliman et al., 2018, 2019, 2021). us,
in line with Collie et al. (2020a); see also Collie and Martin,
2016), it seems that adaptability fosters mastery and ecacy
experiences—manifested in our research by online learning
self-ecacy.
We can also now add to what we know about factors that
may enhance the eectiveness of online learning. As described
earlier, there is a mixed evidence base for the eectiveness of
online learning modes, representing a diversity of positive eects
(Yuwono and Sujono, 2018), negative eects (Peña-López, 2015;
Delgado et al., 2018; Clinton, 2019), and null eects
(Cavanaugh et al., 2004; Kra and Hill, 2020). It has been
suggested that part of this diversity is due to the variety of
factors that inuence online learning eectiveness. Research has
previously identied factors, such as technology access, technology
skills, instructional and resource quality, parent/home support,
ethnicity, socioeconomic status, and learning support needs
(AITSL, 2020; Australian Academy of Science, 2020).
To this, we can now add adaptability which predicted online
JOB DEMANDS
-Online Learning Barriers
BUFFERING EFFECT
-Adaptability x Online Barriers
ONLINE LEARNING
SELF-EFFICACY
PERFORMANCE
-End of Year Test Achievement
JOB RESOURCES
-Online Learning Support
JOB RESOURCES
-Parent/Home Help
PERSONAL RESOURCES
-Adaptability
BACKGROUND
ATTRIBUTES
- Age
- Gender (M/FM)
- Parent Education
- Non-English Speak B’ground
- Math Self-efficacy
-Prior Achievement
.56***
-.06*
.20***
.12*
.08**
-.10***
See Table 3 for
significant covariate
effects
See Table 3 for
significant covariate
effects
FIGURE2 | Standardized beta coefcients for JD-R process in online mathematics. All JD-R factors are in relation to mathematics; *p<0.05, **p<0.01, and
***p<0.001. See Table3 for indirect and covariate effects.
Martin et al. Adaptability, Online Learning, and COVID-19
Frontiers in Psychology | www.frontiersin.org 11 August 2021 | Volume 12 | Article 702163
learning self-ecacy and also achievement via online learning
self-ecacy. Indeed, because adaptability is a modiable
psychological attribute (Martin et al., 2013; Granziera et al.,
2021), it represents a viable direction for assisting students’
online learning experience.
In addition to the positive role of adaptability, online learning
self-ecacy was associated with gains in end of year test
achievement. us, the extent to which students perceived and
experienced competence in online learning was important for
their subsequent academic performance (beyond prior academic
performance). is is consistent with contentions under classic
conceptualizing (e.g., social cognitive theory; Bandura, 1997)
and research (e.g., Martin, 2007, 2009; Lee et al., 2014; Schunk
and DiBenedetto, 2014). Particularly noteworthy is the fact
that online learning self-ecacy predicted gains in achievement
beyond the eects of general mathematics self-ecacy on
achievement—thus, students’ ecacy in online mathematics
learning itself (net general mathematics self-ecacy) was linked
to their later mathematics achievement. ese ndings
demonstrate that achievement is not only a function of subject-
specic mathematics self-ecacy (consistent with prior research;
Green et al., 2007) but also a function of domain-specic
ecacy within the subject: in this case, online learning
self-ecacy.
Unexpected Findings of Note
Following prior research among teachers, we modeled the
interaction between personal resources (adaptability) and job
demands (online learning barriers) to ascertain the extent to
which adaptability may buer the negative eects of job demands
(Collie etal., 2020a; Granziera etal., 2021)—to address Research
Question 1. is interaction (buering) eect was not statistically
signicant; instead, it was the main eects of adaptability
(positive eect) and online learning barriers (negative eect)
that predicted online learning self-ecacy. is is nonetheless
important, as it shows that adaptability yields a positive eect
beyond the barriers that students experience in online learning.
us, adaptability surmounts the negative eects of online
learning barriers, even if it does not buer them.
It was also initially surprising to identify a negative path
between parent/home help and end of year test achievement—
higher levels of help from parents at home were associated
with lower end of year achievement. We suspect this may
beexplained by the reality that academically struggling students
are likely to require more help from their parents—thus, lower
achieving students reported higher levels of parent/home help.
But how do we reconcile this with other research showing
that low parental involvement is associated with lower
achievement (e.g., Lara and Saracostti, 2019)? Wecannot rule
out the possibility that the more intense parental involvement
with their adolescent child while at home during COVID-19
may have been perceived by the student as controlling and
giving rise to a reduction in autonomy-supportive parenting
practices (e.g., Neubauer et al., 2020)—leading to reduced
achievement. Further research is needed to understand this
better, but it does align with recent developments in JD-R
theory and research identifying variability between individuals
in how they perceive demands and resources (Bakker and
Demerouti, 2017; Yin et al., 2018; Han et al., 2020; Martin
et al., 2021), with some seeing resources as more a hindrance
than a help. In the case of our study, perhaps there was a
controlling role for parent/home help that was perceived as a
hindrance, and which evinced a negative eect for achievement.
Similar apparently counter-intuitive eects of parental
involvement and attitudes on students’ academic outcomes have
been found in other studies. Murayama et al. (2016) suggested
that overly positive parental judgments may bedisadvantageous
because they are associated with over involvement, controlling
behavior, and excessive pressure. Other studies explore parental
“intrusive support” of students. For instance, Gunderson et al.
(2012) explain how expectations of parents, based on their
own anxieties and stereotypical beliefs, can lead to lower
achievement, via intrusive support during homework.
Furthermore, we suggest it is important to better understand
the nature and impact of parental involvement as relevant to
the COVID-19 pandemic itself. For example, additional research
is needed to explore diverse dimensions of parental involvement
in their children’s schoolwork during the pandemic with particular
interest in the factors that determine whether this involvement
is perceived as a help or a hindrance.
ere were two non-signicant main eects also worth
noting (but they were not the substantive focus and we did
not formulate hypotheses for them): a non-signicant predictive
path between online learning barriers and achievement and a
non-signicant predictive path between online learning support
and achievement. Wesuggest this is noteworthy because these
two predictors were signicantly correlated with achievement
(see Tabl e 2), but aer including the signicant predictive
roles of adaptability and parent/home help on achievement,
online learning barriers and support explained no further
variance in achievement. Moreover, because adaptability yielded
a unique net positive eect on achievement relative to the net
negative eect of parent/home help (see total eects in Ta bl e 3 )
and because adaptability shared more variance with online
learning barriers and support than did parent/home help (see
Tabl e 2), we suggest it is the presence of adaptability that
played a major role in mitigating the predictive paths from
online learning barriers and support to achievement. e two
non-signicant paths also underscore an important mediating
role for self-ecacy, in similar vein to prior research nding
that teacher self-ecacy fully mediates the link between teachers’
adaptability and students’ outcomes (Collie etal., 2020a). ese
ndings, we suggest, further highlight the importance of
considering adaptability as a personal resource in JD-R models
generally (in line with emerging research: Collie etal., 2020a;
Granziera et al., 2021), and in models exploring disruptive
circumstances, such as COVID-19 more specically.
Implications for Theory and Practice
Based on the ndings, webelieve wehave successfully adapted
JD-R theory to the (online) learning and instruction setting
in high school mathematics. Weshowed that personal resources
by way of adaptability positively impacted students’ online
learning experiences and outcomes (consistent with research
Martin et al. Adaptability, Online Learning, and COVID-19
Frontiers in Psychology | www.frontiersin.org 12 August 2021 | Volume 12 | Article 702163
showing the positive impacts of adaptability among teachers;
Collie et al., 2020a; Granziera et al., 2021). We showed that
job demands by way of online learning barriers were associated
with lower online learning self-ecacy (consistent with research
showing such barriers impede online learning; e.g., Peña-López,
2015; Australian Academy of Science, 2020). We also showed
that job resources by way of online learning support and parent/
home help were associated with higher online learning self-
ecacy (consistent with prior research demonstrating a supportive
role for these factors; e.g., Yukselturk and Bulut, 2007; Means
et al., 2009; Escueta et al., 2017; Galpin and Taylor, 2018;
Gregori et al., 2018; AITSL, 2020).
e salient role of adaptability in this study also suggests
it as an important point for educational intervention. As
adaptability is an emerging area of research, suggested practice
directions have drawn on existing related frameworks, such
as the resilience research by Rutter (1987) and Morales (2000).
For example, Martin et al. (2013); see also Burns and Martin
(2014) and Martin and Burns (2014) identied the following
steps to boost students’ adaptability: (1) teach students how
to recognize novelty, variability, and uncertainty, (2) explain
to students how they can adjust their behavior, thinking, and/
or emotion to navigate the novelty, variability, and uncertainty
(strategies are detailed below), (3) encourage students to recognize
the benets of these psycho-behavioral adjustments, and (4)
explain to students that continued behavioral, cognitive, and/
or emotional responses to novelty, variability, and uncertainty
represent the “adaptability cycle” and that this cycle leads to
enhanced ongoing positive outcomes in the face of change.
Burns and Martin (2014) and Martin and Burns (2014)
propose that the second step of this process (adjusting behavior,
cognition, and emotion) is the most critical part of the adaptability
cycle. According to Martin (2014); see also Burns and Martin
(2014) and Martin and Burns (2014) and extrapolating his
guidance to online learning, (a) students can adjust their
cognition by thinking about a new online task in a dierent
way (e.g., considering the opportunities the new online option
might oer); (b) students can adjust their behavior by seeking
out new or more online information and resources, or asking
for help (e.g., asking a teacher to help with a new online
learning management system); and (c) students can adjust their
emotions by minimizing negative feelings (e.g., frustration)
when they need to juggle in-class and online learning modes
(e.g., choosing not to focus on disappointment if the teacher
engages an online learning approach that is not to the
student’s preference).
Our ndings also showed that adaptability is not the only
practical implication to take from this study; it is also important
to remove barriers to students’ online learning and to enhance
their online learning resources. Attending to the online learning
barriers would entail addressing Internet and connection issues,
ensuring students have access to appropriate computing and
technology, and identifying places for them to engage with
online learning so they can concentrate (Australian Academy
of Science, 2020). Attending to online learning support would
involve ensuring high quality learning management systems,
providing ample opportunity to interact with and receive help
from the teacher online, and being provided with the opportunity
to engage with peers online but also to work independently
as appropriate.
Limitations and Future Directions
ere are some limitations in this study that are important
to take into account when interpreting the ndings and which
also have implications for future research into online learning.
First, our correlational research data cannot be interpreted as
supporting causal conclusions. Experimental work that
manipulates adaptability and explores for any subsequent shis
in online learning self-ecacy would better establish (or not)
the causal role of adaptability. Indeed, Galpin and Taylor (2018)
and others (e.g., Means etal., 2009; Patrick and Powell, 2009;
Quesada-Pallarès et al., 2019) recommend more studies that
can test causality (including experimental studies) and the
factors that may moderate whether online learning is benecial
or not. Second, although our achievement data were based
on a mathematics test tapping into diverse aspects of mathematics
syllabus, it will be important to expand the outcome measures
to assess other aspects of mathematics performance. ird,
there tends to be more research into online learning among
post-school students (e.g., university/college) and to some extent
among high school students (such as in our study); there is
a need for more research among elementary school students
(Means et al., 2009; Clinton, 2019). Fourth, this study relied
on student reports of online learning barriers and support.
Additional indicators, such as parent and teacher ratings, might
be used in future to triangulate ndings with students’ reports
of constructs in our study. Also on the matter of measurement,
we assessed online learning resources in terms of student
appraisals (via ratings of satisfaction) and not in terms of
characteristics of the resources themselves. Findings and
conclusions regarding job resources in our study must take
this into account. Fih, we suggest research that can identify
dierent combinations of demands and resources and their
relationships to online learning self-ecacy and academic
achievement. As a case in point, latent prole analysis may
identify distinct typologies of students who balance the diverse
online demands and resources in dierent ways. Prior JD-R
research has conducted latent prole analysis among teachers
(Collie et al., 2020b) and expanding this to students would
be illuminating.
Sixth, it will be helpful to understand adaptability and its
role in online learning in real-time. For example, research has
identied the in-situ dimensions of students’ learning and
engagement (Schneider etal., 2016; Martin et al., 2020); online
learning demands and resources are also likely to have salient
in-situ aspects. Seventh, due to constraints of time and to
accommodate the fact students were located in diverse
combinations of online and in-class learning modes, wewanted
to guard against asking extensive batteries of questions about
their online experience. us, single-item indicators were used
in some cases. Although there is research suggesting single-
item scales have merit in cases where long scales are not able
to be used (e.g., Gogol et al., 2014) and wemodeled an error-
adjusted score for our central online learning self-ecacy factor,
Martin et al. Adaptability, Online Learning, and COVID-19
Frontiers in Psychology | www.frontiersin.org 13 August 2021 | Volume 12 | Article 702163
future research might look to administering more extensive
item sets. Eighth, our research was set in mathematics which
is a challenging school subject and one in which students can
struggle (omson et al., 2016; OECD, 2018). To the extent
this is so, there may be disproportionate challenges in online
mathematics learning—or, it may emerge there are unique
opportunities aorded to mathematics when in online learning
modes. It is thus important to expand the present study to
other school subjects. Ninth, students in our sample were from
above average SES backgrounds. As such, these students likely
had fewer online learning barriers and more online learning
support than some other cohorts of students. Our ndings
may be just the tip of the iceberg in terms of the role of
these demands and resources. Finally, online learning platforms,
programs, and content tend to be developed and published
faster than research can assess their eectiveness (Escueta etal.,
2017)—signaling a need to conduct more rapid research in
order for researchers and research to stay abreast of the fast
pace of developments in online learning.
CONCLUSION
e COVID-19 pandemic necessitated a rapid shi to remote
learning for students around the world. During this time,
in-class learning and instruction moved to remote online modes
at speed and scale. Harnessing JD-R theory, the present study
identied adaptability as a personal resource that may support
students’ online learning experience and achievement during
such times. Findings demonstrated that adaptability does indeed
play a signicant role in this process, and thus may be an
important personal resource to foster in students’ online learning
during COVID-19—and beyond.
DATA AVAILABILITY STATEMENT
e datasets presented in this article are not readily available
because consent from participants to share dataset is not
available; summative data (e.g., correlation matrix with standard
deviations) are available to enable analyses. Requests to access
the datasets should be directed to Andrew Martin, andr ew.
martin@unsw.edu.au.
ETHICS STATEMENT
e studies involving human participants were reviewed and
approved by the UNSW Human Ethics Committee. Written
informed consent to participate in this study was provided by
the participants’ legal guardian/next of kin.
AUTHOR CONTRIBUTIONS
AM shared in the development of the research design and
led data analysis and report writing. RC and RN shared in
the development of the research design and assisted with data
analysis and report writing. All authors contributed to the
article and approved the submitted version.
REFERENCES
Aiken, L. S., West, S. G., and Reno, R. R. (1991). Multiple Regression: Testing
and Interpreting Interactions. London: Sage.
AITSL (2020). What Works in Online/Distance Teaching and Learning? Melbourne:
AITSL.
Arbuckle, J. L. (1996). “Full information estimation in the presence of incomplete
data,” in Advanced Structural Equation Modeling: Issues and Techniques. eds.
G. A. Marcoulides and R. E. Schumacker (Mahwah, NJ: Lawrence Erlbaum
Associates), 243–278.
Australian Academy of Science (2020). Learning outcomes for online versus
in-class education. Australian Academy of Science. Available at: https://www.
science.org.au/covid19/learning-outcomes-online-vs-inclass-education
(Accessed March 15, 2021).
Backes, B., and Cowan, J. (2019). Is the pen mightier than the keyboard? e
eect of online testing on measured student achievement. Econ. Educ. Rev.
68, 89–103. doi: 10.1016/j.econedurev.2018.12.007
Bakker, A. B., and Demerouti, E. (2017). Job demands-resources theory: taking
stock and looking forward. J. Occup. Health Psychol. 22, 273–285. doi:
10.1037/ocp0000056
Bakker, A., and Demerouti, E. (2018). “Multiple levels in job demands-resources
theory: implications for employee well-being and performance,” in Handbook
of Well-Being. eds. E. Diener, S. Oishi and L. Tay (Utah: DEF Publishers),
1–13.
Bandura, A. (1997). Self-Efficacy: e Exercise of Control. New York, NY: Freeman.
Burns, E. C., and Martin, A. J. (2014). ADHD and adaptability: the roles of
cognitive, behavioural, and emotional regulation. Aust. J. Guid . Couns. 24,
227–242. doi: 10.1017/jgc.2014.17
Cavanaugh, C., Gillan, K. J., Kromrey, J., Hess, M., and Blomeyer, R. (2004).
e Effects of Distance Education on k-12 Student Outcomes: A Meta-Analysis.
Naperville, IL: Learning Point Associates/North Central Regional Educational
Laboratory (NCREL).
Chang, M.-L. (2013). Toward a theoretical model to understand teacher emotions
and teacher burnout in the context of student misbehavior: appraisal, regulation,
and coping. Motiv. Emot. 37, 799–817. doi: 10.1007/s11031-012-9335-0
Clinton, V. (2019). Reading from paper compared to screens: a systematic
review and meta-analysis. J. Res. R ead. 42, 288–325. doi:
10.1111/1467-9817.12269
Cole, D. A., and Preacher, K. J. (2014). Manifest variable path analysis: potentially
serious and misleading consequences due to uncorrected measurement error.
Psychol. Methods 19, 300–315. doi: 10.1037/a0033805
Collie, R. J., Granziera, H., Martin, A. J., Burns, E. C., and Holliman, A.
(2020a). Adaptability among science teachers in schools: a multi-nation
examination of its role in school outcomes. Teac h. Tea ch . E duc . 26, 350–364.
doi: 10.1080/13540602.2020.1832063
Collie, R. J., Holliman, A. J., and Martin, A. J. (2017). Adaptability, engagement,
and academic achievement at university. Educ . Psychol. 37, 632–647. doi:
10.1080/01443410.2016.1231296
Collie, R. J., Malmberg, L.-E., Martin, A. J., Sammons, P., and Morin, A. J. S.
(2020b). A multilevel person-centered examination of teachers’ workplace
demands and resources: links with work-related well-being. Front . Psych ol.
11:626. doi: 10.3389/fpsyg.2020.00626
Collie, R. J., and Martin, A. J. (2016). Adaptability: an important capacity for
eective teachers. Educ. Pract . eor y 38, 27–39. doi: 10.7459/ept/38.1.03
Delgado, P., Vargas, C., Ackerman, R., and Salmerón, L. (2018). Don’t throw
away your printed books: A meta-analysis on the eects of reading
media on comprehension. Educ. Res. Rev. 25, 23–38. doi: 10.1016/j.
edurev.2018.09.003
Martin et al. Adaptability, Online Learning, and COVID-19
Frontiers in Psychology | www.frontiersin.org 14 August 2021 | Volume 12 | Article 702163
Demerouti, E., Bakker, A. B., Nachreiner, F., and Schaufeli, W. B. (2001). e
job demands-resources model of burnout. J. Appl. Psy chol. 86, 499–512.
doi: 10.1037/0021-9010.86.3.499
Dicke, T., Stebner, F., Linninger, C., Kunter, M., and Leutner, D. (2018). A
longitudinal study of teachers’ occupational well-being: applying the job
demands-resources model. J. Occup. Health Psychol. 23, 262–277. doi: 10.1037/
ocp0000070
Escueta, M., Quan, V., Nickow, A., and Oreopoulos, P. (2017). Education
Technology: An Evidence-Based Review (No. w23744). Cambridge, MA: National
Bureau of Economic Research.
Galpin, A., and Taylor, G. (2018). Changing Behaviour: Children, Adolescents
and Screen Use. Leicester: e British Psychological Society.
Garbe, A., Ogurlu, U., Logan, N., and Cook, P. (2020). COVID-19 and remote
learning: experiences of parents with children during the pandemic. Am.
J. Qual. Res. 4, 45–65. doi: 10.29333/ajqr/8471
Gogol, K., Brunner, M., Goetz, T., Martin, R., Ugen, S., Keller, U., et al. (2014).
“My questionnaire is too long!” e assessments of motivational-aective
constructs with three-item and single-item measures. Contemp. E duc. Psychol.
39, 188–205. doi: 10.1016/j.cedpsych.2014.04.002
Granziera, H., Collie, R. J., and Martin, A. J. (2020). “Understanding teacher
wellbeing through Job Demands-Resources theory,” in Cultivating Teacher
Resilience. ed. C. F. Manseld (New York, NY: Springer), 229–244.
Granziera, H., Collie, R. J., and Martin, A. J. (2021). Teachers’ Job Demands
and Resources. Submitted for publication.
Green, J., Martin, A. J., and Marsh, H. W. (2007). Motivation and engagement
in English, Mathematics and science high school subjects: towards an
understanding of multidimensional domain specicity. Learn. Indiv id. Differ.
17, 269–279. doi: 10.1016/j.lindif.2006.12.003
Gregori, E. B., Zhang, J., Galván-Fernández, C., and de Asís Fernández-Navarro, F.
(2018). Learner support in MOOCs: identifying variables linked to completion.
Comput. Educ. 122, 153–168. doi: 10.1016/j.compedu.2018.03.014
Gunderson, E. A., Ramirez, G., Levine, S. C., and Beilock, S. L. (2012). e
role of parents and teachers in the development of gender-related math
attitudes. Sex Roles 66, 153–166. doi: 10.1007/s11199-011-9996-2
Han, J., Yin, H., Wang, J., and Bai, Y. (2020). Challenge job demands and job
resources to university teacher well-being: the mediation of teacher ecacy.
Stud . High. E duc. 45, 1771–1785. doi: 10.1080/03075079.2019.1594180
Hassler Hallstedt, M., Klingberg, T., and Ghaderi, A. (2018). Short and long-
term eects of a mathematics tablet intervention for low performing second
graders. J. Educ. Psychol. 110, 1127–1148. doi: 10.1037/edu0000264
Hattie, J. (2009). Visible Learning: A Synthesis of Over 800 Meta-Analyses Relating
to Achievement. New York, NY: Routledge.
Holliman, A. J., Martin, A. J., and Collie, R. J. (2018). Adaptability, engagement,
and degree completion: a longitudinal investigation of university students.
Educ. Psych ol. 38, 785–799. doi: 10.1080/01443410.2018.1426835
Holliman, A., Sheriston, L., Martin, A. J., Collie, R. J., and Sayer, D. (2019).
Adaptability: does students’ adjustment to university predict their mid-course
academic achievement and satisfaction? J. Furth . High. Educ. 43, 1444–1455.
doi: 10.1080/0309877X.2018.1491957
Holliman, A., Waldeck, D., Jay, B., Murphy, S., Atkinson, E., Collie, J., et al.
(2021). Adaptability and social support: examining links with psychological
wellbeing among UK students and non-students. Front. Psychol. 12:636520.
doi: 10.3389/fpsyg.2021.636520
Hu, L., and Bentler, P. M. (1999). Cuto criteria for t indexes in covariance
structure analysis: conventional criteria versus new alternatives. Struct. Equ.
Model. Multidiscip. J. 6, 1–55. doi: 10.1080/10705519909540118
Kirschner, P. A., and De Bruyckere, P. (2017). e myths of the digital native
and the multitasker. Teac h. Tea ch . E duc . 67, 135–142. doi: 10.1016/j.
tate.2017.06.001
Klassen, R. M., and Chiu, M. M. (2010). Eects on teachers’ self-ecacy and
job satisfaction. J. Educ . Psych ol. 102, 741–756. doi: 10.1037/a0019237
Kline, R. B. (2016). Principles and Practice of Structural Equation Modeling.
4th Edn. New York, NY: Guilford Press.
Kra, M. A., and Hill, H. C. (2020). Developing ambitious mathematics instruction
through web-based coaching: a randomized eld trial. Am. Educ. Res. J.
57, 2378–2414. doi: 10.3102/0002831220916840
Lara, L., and Saracostti, M. (2019). Eect of parental involvement on children’s
academic achievement in Chile. Front. Psychol. 10:1464. doi: 10.3389/
fpsyg.2019.01464
Lee, W., Lee, M. J., and Bong, M. (2014). Testing interest and self-ecacy as
predictors of academic self-regulation and achievement. Contemp. Educ.
Psychol. 39, 86–99. doi: 10.1016/j.cedpsych.2014.02.002
Liem, G. A. D., and Martin, A. J. (2015). Young people’s responses to environmental
issues: exploring the role of adaptability and personality. Person al. Indi vid.
Di ffer. 79, 91–97. doi: 10.1016/j.paid.2015.02.003
Liu, C., and Li, H. (2018). Stressors and stressor appraisals: the moderating
eect of task ecacy. J. Bus. Psy chol. 33, 141–154. doi: 10.1007/
s10869-016-9483-4
Ma, J., Peng, Y., and Wu, B. (2021). Challenging or hindering? e roles of
goal orientation and cognitive appraisal in stressor-performance relationships.
J. Organ. Beh av. 42, 388–406. doi: 10.1002/job.2503
MacKinnon, D. P., Lockwood, C. M., Homan, J. M., West, S. G., and Sheets, V.
(2002). A comparison of methods to test mediation and other intervening
variable eects. Psychol. Methods 7, 83–104. doi: 10.1037/1082-989X.7.1.83
Marsh, H. W., and Martin, A. J. (2011). Academic self-concept and academic
achievement: relations and causal ordering. Br. J. Edu c. Psychol. 81, 59–77.
doi: 10.1348/000709910X503501
Martin, A. J. (2007). Examining a multidimensional model of student motivation
and engagement using a construct validation approach. Br. J. Educ . Psych ol.
77, 413–440. doi: 10.1348/000709906X118036
Martin, A. J. (2009). Motivation and engagement across the academic lifespan:
a developmental construct validity study of elementary school, high school,
and university/college students. Educ. Psychol. Meas. 69, 794–824. doi:
10.1177/0013164409332214
Martin, A. J. (2014). Academic buoyancy and academic outcomes: towards a
further understanding of students with ADHD, students without ADHD,
and academic buoyancy itself. Br. J. Educ. Psychol. 84, 86–107. doi: 10.1111/
bjep.12007
Martin, A. J. (2020). Motivation and Engagement Scale. Sydney: Lifelong
Achievement Group.
Martin, A. J., and Burns, E. C. (2014). Academic buoyancy, resilience, and
adaptability in students with ADHD. ADHD Rep. 22, 1–9. doi: 10.1521/
adhd.2014.22.6.1
Martin, A. J., Durksen, T. L., Williamson, D., Kiss, J., and Ginns, P. (2016).
e role of a museum-based science education program in promoting content
knowledge and science motivation. J. Res. Sc i. Teach. 53, 1364–1384. doi:
10.1002/tea.21332
Martin, A. J., Kennett, R., Pearson, J., Mansour, M., Papworth, B., and
Malmberg, L.-E. (2021). Challenge and threat appraisals in high school
science: Investigating the roles of psychological and physiological factors.
Educ. Psych ol. 41, 618–639. doi: 10.1080/01443410.2021.1887456
Martin, A. J., Mansour, M., and Malmberg, L.-E. (2020). What factors inuence
students’ real-time motivation and engagement? An experience sampling
study of high school students using mobile technology. Educ . Psych ol. 40,
1113–1135. doi: 10.1080/01443410.2018.1545997
Martin, A. J., and Marsh, H. W. (2009). Academic resilience and academic
buoyancy: multidimensional and hierarchical conceptual framing of causes,
correlates, and cognate constructs. Ox f. Rev. Ed uc. 35, 353–370. doi:
10.1080/03054980902934639
Martin, A. J., Nejad, H. G., Colmar, S., and Liem, G. A. D. (2012). Adaptability:
conceptual and empirical perspectives on responses to change, novelty and
uncertainty. Aust. J. Gui d. Couns. 22, 58–81. doi: 10.1017/jgc.2012.8
Martin, A. J., Nejad, H. G., Colmar, S., and Liem, G. A. D. (2013). Adaptability:
how students’ responses to uncertainty and novelty predict their academic
and non-academic outcomes. J. Educ. Ps ychol . 105, 728–746. doi: 10.1037/
a0032794
Martin, A. J., Nejad, H. G., Colmar, S., Liem, G. A. D., and Collie, R. (2015).
e role of adaptability in promoting control and reducing failure dynamics:
a mediation model. Learn. Indiv id. Differ. 38, 36–43. doi: 10.1016/j.
lindif.2015.02.004
Martin, A. J., Yu, K., Ginns, P., and Papworth, B. (2017). Young people’s academic
buoyancy and adaptability: a cross-cultural comparison of China with North
America and the United Kingdom. Educ. Psychol. 37, 930–946. doi:
10.1080/01443410.2016.1202904
Means, B., Toyama, Y., Murphy, R., Bakia, M., and Jones, K. (2009). Evaluation
of Evidence-Based Practices in Online Learning: A Meta-Analysis and Review
of Online Learning Studies. Washington DC: U.S. Department of Education,
Oce of Planning, Evaluation, and Policy Development.
Martin et al. Adaptability, Online Learning, and COVID-19
Frontiers in Psychology | www.frontiersin.org 15 August 2021 | Volume 12 | Article 702163
Morales, E. E. (2000). A contextual understanding of the process of educational
resilience: high achieving Dominican American students and the ‘resilience
cycle’. Innov. High. E duc. 25, 7–22. doi: 10.1023/A:1007580217973
Murayama, K., Pekrun, R., Suzuki, M., Marsh, H. W., and Lichtenfeld, S. (2016).
Don’t aim too high for your kids: parental overaspiration undermines students’
learning in mathematics. J. Pers. S oc. Psychol. 111, 766–779. doi: 10.1037/
pspp0000079
Muthén, L. K., and Muthén, B. O. (2017). Mplus User’s Guide. 7th Edn. Los
Angeles, CA: Muthén & Muthén.
Neubauer, A., Schmidt, A., Kramer, A., and Schmiedek, F. (2020). A little
autonomy support goes a long way: daily autonomy-supportive parenting,
child well-being, parental need fulllment, and change in child, family, and
parent adjustment across the adaptation to the COVID-19 pandemic. Child
De v. doi: 10.1111/cdev.13515 [Epub ahead of print]
Nguyen, T. (2015). e eectiveness of online learning: beyond no signicant
dierence and future horizons. MERLOT J. Online Learn. Teach. 11, 309–
319.
Nicholas, J., Poladian, L., Mack, J., and Wilson, R. (2015). Mathematics preparation
for university: entry, pathways and impact on performance in rst year science
and mathematics subjects. Int. J. Innov at. Sci. Math. E duc. 23, 37–59.
OECD (2018). PISA 2015 Results in Focus. Paris: OECD.
Patrick, S., and Powell, A. (2009). A Summary of Research on the Effectiveness
of K-12 Online Learning. Vienna, VA: International Association for K-12
Online Learning.
Peña-López, I. (2015). Students, Computers and Learning: Making the Connection.
Paris: OECD.
Quesada-Pallarès, C., Sánchez-Martí, A., Ciraso-Calí, A., and Pineda-Herrero, P.
(2019). Online vs. classroom learning: examining motivational and self-
regulated learning strategies among vocational education and training students.
Front. Psychol. 10:2795. doi: 10.3389/fpsyg.2019.02795
Rutter, M. (1987). Psychosocial resilience and protective mechanisms. Am. J.
Orthopsychiatry 37, 317–331. doi: 10.1111/j.1939-0025.1987.tb03541.x
Schaufeli, W. B., and Bakker, A. B. (2004). Job demands, job resources, and
their relationship with burnout and engagement: a multi-sample study. J.
Organ. Behav. 25, 293–315. doi: 10.1002/job.248
Schaufeli, W. B., and Taris, T. W. (2014). “A critical review of the job demands-
resources model: implications for improving work and health,” in Bridging
Occupational, Organizational and Public Health. eds. G. Bauer and O. Hemming
(New York, NY: Springer).
Schneider, B., Krajcik, J., Lavonen, J., Salmela-Aro, K., Broda, M., Spicer, J.,
et al. (2016). Investigating optimal learning moments in US and Finnish
science classes. J. Res. Sci. Teach. 53, 400–421. doi: 10.1002/tea.21306
Schunk, D. H., and DiBenedetto, M. K. (2014). “Academic self-ecacy,” in
Handbook of Positive Psychology in Schools. eds. M. J. Furlong, R. Gilman
and E. S. Huebner (New York, NY: Elsevier), 115–521.
Shrout, P. E., and Bolger, N. (2002). Mediation in experimental and
nonexperimental studies: new procedures and recommendations. Psychol.
Methods 7, 422–445. doi: 10.1037/1082-989X.7.4.422
Skaalvik, E. M., and Skaalvik, S. (2018). Job demands and job resources as
predictors of teacher motivation and well-being. Soc. Psychol. Educ. 21,
1251–1275. doi: 10.1007/s11218-018-9464-8
Sung, Y. T., Yang, J. M., and Lee, H. Y. (2017). e eects of mobile-computer-
supported collaborative learning: meta-analysis and critical synthesis. Re v.
Educ. Res. 87, 768–805. doi: 10.3102/0034654317704307
alheimer, W. (2017). Does eLearning Work? What the Scientific Research Says.
Somerville, MA: Work-Learning Research, Inc.
omson, S., De Bortoli, L., and Underwood, C. (2016). PISA 2015: A First
Look at Australia’s Results. Melbourne: ACER.
Tinmaz, H., and Lee, J. H. (2020). An analysis of users’ preferences on learning
management systems. Smart Learn. Environ. 7, 1–17. doi: 10.1186/
s40561-020-00141-8
Xanthopoulou, D., Bakker, A. B., Demerouti, E., and Schaufeli, W. B. (2007).
e role of personal resources in the job demands-resources model. Int. J.
Stress. Manag. 14, 121–141. doi: 10.1037/1072-5245.14.2.121
Yin, H., Huang, S., and Lv, L. (2018). A multilevel analysis of job characteristics,
emotion regulation, and teacher well-being: a job demands-resources model.
Front. Psychol. 9:2395. doi: 10.3389/fpsyg.2018.02395
Yukselturk, E., and Bulut, S. (2007). Predictors for student success in an online
course. J. Educ . Technol. S oc. 10, 71–83.
Yuwono, K. T., and Sujono, H. D. (2018). e eectiveness of E-learning: a
meta-analysis. J. Phys. Conf. Ser. 1140:012024. doi: 10.1088/1742-
6596/1140/1/012024
Conflict of Interest: e authors declare that the research was conducted in
the absence of any commercial or nancial relationships that could beconstrued
as a potential conict of interest.
Publisher’s Note: All claims expressed in this article are solely those of the
authors and do not necessarily represent those of their aliated organizations,
or those of the publisher, the editors and the reviewers. Any product that may
be evaluated in this article, or claim that may be made by its manufacturer, is
not guaranteed or endorsed by the publisher.
Copyright © 2021 Martin, Collie and Nagy. is is an open-access article distributed
under the terms of the Creative Commons Attribution License (CC BY). e use ,
distribution or reproduction in other forums is permitted, provided the
original author(s) and the copyright owner(s) are credited and that the original
publication in this journal is cited, in accordance with accepted academic practice.
No use, distribution or reproduction is permitted which does not comply with
these terms.
