Content uploaded by Lin Sophie Teng
Author content
All content in this area was uploaded by Lin Sophie Teng on May 02, 2024
Content may be subject to copyright.
A Questionnaire-Based Validation
of Multidimensional Models
of Self-Regulated Learning Strategies
LIN SOPHIE TENG
The University of Auckland
Faculty of Education and Social Work
School of Curriculum and Pedagogy
Private Bag 92601, Symonds Street
Auckland 1150
New Zealand
Email: lin.teng@auckland.ac.nz
LAWRENCE JUN ZHANG*
The University of Auckland
Faculty of Education and Social Work
School of Curriculum and Pedagogy
Private Bag 92601, Symonds Street
Auckland 1150
New Zealand
Email: lj.zhang@auckland.ac.nz
This study aimed to validate a newly-developed instrument, The Writing Strategies for Self-Regulated
Learning (SRL) Questionnaire, with respect to its multifaceted structure of SRL strategies in English as
a foreign language (EFL) writing. A total of 790 undergraduate students from 6 universities in Northeast
China volunteered to be participants. Conrmatory factor analyses (CFA) through structural equation
modeling (SEM) were applied to evaluate 3 hypothesized models. The results of the CFA validated a
9-factor correlated model of second language (L2) writing strategies for SRL with satisfactory psychome-
tric characteristics. Model comparisons conrmed a hierarchical, multidimensional structure of SRL as
the best model, in which self-regulation, as a higher order construct, accounted for the correlations of
the 9 lower-order writing strategies, pertaining to cognitive, metacognitive, social–behavioral, and moti-
vational regulation aspects. Multiple regression analysis revealed that 6 out of 9 SRL strategies had sig-
nicant predictive effects on EFL writing prociency. The empirical evidence lends preliminary support
to a transfer of SRL theory from educational psychology to the eld of L2/EFL education, particularly
L2/EFL writing. Implications of these ndings are discussed.
Keywords: self-regulated learning (SRL); writing strategies for SRL; Chinese university learners; question-
naire validation; language learning strategies (LLSs); multidimensional models of SRL
THERE IS GROWING RECOGNITION THAT
one of the fundamental goals of education is
to teach students to become self-regulated learn-
ers, who can actively and efciently manage their
own learning by deploying various self-regulated
learning (SRL) strategies in the learning process
(Zimmerman & Risemberg, 1997). In the edu-
cational psychology literature, SRL is dened as
“an active, constructive process whereby learners
set goals for their learning and then attempt to
monitor, regulate, and control their cognition,
The Modern Language Journal, 100, 0, (2016)
DOI: 10.1111/modl.12339
0026-7902/16/1–28 $1.50/0
C2016 The Modern Language Journal
motivation and behavior” (Pintrich, Wolters, &
Baxter, 2000, p. 453). In the eld of second and
foreign language (L2) education, studies reveal
the primary role of SRL strategies in fostering stu-
dents’ self-regulated capacity to improve their lan-
guage prociency (e.g., Oxford, 2013). It appears
that self-regulation is pivotal to student success in
L2 learning in academic settings.
Dörnyei (2005) posits that self-regulation is “a
multidimensional construct, including cognitive,
metacognitive, motivational, behavioral, and envi-
ronmental processes that learners can apply to en-
hance academic achievement” in different learn-
ing contexts (p. 101). To date, most empirical
studies have documented cognitive and metacog-
nitive strategies in rst language (L1) reading and
writing (e.g., Harris et al., 2011; Zimmerman &
*corresponding author.
2The Modern Language Journal 100 (2016)
Bandura, 1994) as well as L2 reading (N. J. Ander-
son, 1991; Zhang, 2008; Zhang, Gu, & Hu, 2008).
By contrast, there has been little explicit discus-
sion of L2 writing strategies based on SRL theo-
ries, especially in English as a foreign language
(EFL) settings, nor has the interplay of the multi-
ple dimensions of SRL strategies been empirically
examined in L2 contexts.
Furthermore, little attention has been given to
the development of instruments for evaluating
SRL strategies in EFL contexts. Despite an abun-
danceofresearchonevaluatingSRLstrategiesin
L1 settings (e.g., Pintrich et al., 1991; Weinstein,
Schulte, & Palmer, 1987), these instruments can-
not be generalized across settings, especially with
regard to EFL writing, which has been acknowl-
edged as a complex, situational, and multifaceted
process, uctuating in a wide range of contexts
(Byrnes, 2014a; Manchón, 2009; Zhang, 2013).
These lacunae point to a need to develop a theo-
retically robust instrument with vigorous psycho-
metric properties to evaluate students’ perceived
use of SRL strategies in EFL writing.
Our study attempts to conceptualize, develop,
and validate a new questionnaire framed within
the multidimensional structure of SRL theory for
evaluating EFL students’ perceived use of writing
strategies. We propose a multifaceted portrayal
of EFL writing strategies that includes cognition,
metacognition, social behavior, and motivational
regulation, all of which are essential to fostering
self-regulated writers. We also explore how stu-
dents’ reported use of writing strategies for SRL
affects their writing prociency.
SELF-REGULATED LEARNING (SRL)
SRL theory originates from educational psy-
chology and has exhibited exponential develop-
ment since the 1970s. As a construct, SRL refers to
the degree to which individuals are active partici-
pants in their own learning; it is a more dynamic
concept than learning strategies in that the learn-
ers’ “strategic efforts to manage their own achieve-
ment through specic beliefs and processes” are
pivotal to their success in learning (Zimmerman
& Risemberg, 1997, p. 105). SRL involves “cogni-
tive, affective, motivational, and behavioral com-
ponents that provide the individual with the ca-
pacity to adjust his or her actions and goals to
achieve desired results in light of changing en-
vironment conditions” (Dörnyei, 2005, p. 191).
Zimmerman and Schunk (2011) have argued that
in the self-regulating process, learners intention-
ally activate, sustain, and adjust cognitions, af-
fects, and behaviors to achieve their learning
goals through the effective deployment of learn-
ing strategies.
Researchers such as Oxford (2013) and Zim-
merman (2011) have acknowledged that SRL
strategies build on a multifaceted structure that
includes cognition, metacognition, social be-
havior, and motivational regulation. Learners
deploy different dimensions of SRL strategies to
help them actively control their internal mental
states, beliefs, observable behaviors, and environ-
ments in the learning process (Andrade & Evans,
2012; Zimmerman, 2013). During the past three
decades, extensive research on SRL strategies has
been conducted in order to nd ways for improv-
ing students’ academic achievement and fostering
their learning capacity. A rst step was to classify
the SRL strategies used in general learning con-
texts. Zimmerman and Martinez–Pons (1986),
for example, working with elementary and sec-
ondary students from both high achievement
and low achievement tracks, endeavored to iden-
tify different categories of SRL strategies. They
developed a semi-structured interview guide, the
Self-Regulated Learning Interview Scale (SRLIS),
to investigate students’ reported use of SRL strate-
gies in the contexts of class, homework, and self-
study. They elicited 14 categories of SRL strate-
gies, which included self-evaluation, organization
and transformation, goal setting and plan-
ning, information seeking, record keeping, self-
monitoring, environmental structuring, giving
self-consequences, rehearsing and memorizing,
seeking social assistance, and reviewing. Their
study revealed that of the three factors, gender,
social economic status, and SRL strategies, SRL
measures demonstrated the best prediction of
students’ standardized achievement test scores.
Another strand of SRL research gave primary
attention to how individual differences inuence
the deployment of SRL strategies. These stud-
ies on individual variations investigated constructs
such as motivations and goals (e.g., Zimmerman,
2008), attributions (e.g., Schunk & Rice, 1986),
self-efcacy (e.g., Schunk & Ertmer, 2000), and
emotions (e.g., Boekaerts, 2011). The empirical
ndings so far have conrmed that the use of SRL
strategies is inuenced by a range of individual
differences.
In sum, research in the eld of general ed-
ucation presents SRL strategies as multifaceted,
incorporating metacognition, affect, and diverse
social factors in general or specic learning con-
texts. On that basis it is now possible to explore,
both from theoretical and empirical perspectives,
how different SRL strategies inuence students’
learning outcomes in the context of EFL writing.
Lin Sophie Teng and Lawrence Jun Zhang 3
LANGUAGE LEARNING STRATEGIES (LLS)
AND SRL
When discussing SRL strategies for L2 learn-
ing it is important to refer to the already well-
developed eld of language learning strategies
(LLSs). Emanating from the seminal work of
Rubin (1975; see also Stern, 1975), a rich body of
studies has underlined the important relationship
between LLSs and students’ academic achieve-
ment (e.g., Cohen & Macaro, 2007; O’Malley &
Chamot, 1990). However, that research on LLSs
has encountered a barrage of criticisms, including
denitional fuzziness, contentious taxonomies,
insufcient theorizing, and a lack of a psycho-
metrically sound instrument for measuring LLSs
(Dörnyei, 2005; Tseng, Dörnyei, & Schmitt, 2006;
Woodrow, 2005).
Such criticisms would seem to indicate that LLS
research has run its course. However, researchers
have recently called for a more differentiated
analysis and evaluation of LLS research (e.g.,
Cohen & Grifths, 2015). It is worth recalling
that O’Malley and Chamot (1990) derived their
process-oriented theoretical understanding of
LLSs from J. R. Anderson’s (1983, 1985) well-
established principles for the adaptive control
of thought (ACT*). While their approach to
LLSs is well aligned with understanding about
learning strategies in both general and educa-
tional psychology, it may have overextended itself
by lumping together all strategies to cover all
language skill areas, thereby setting up numerous
denitional hurdles. Just as problematic is the fact
that, as Cohen and Macaro (2007), Gao (2007),
and Rose (2011), among others, have pointed
out, LLSs have actually not been well researched,
leading to inappropriate over-generalizations re-
garding their use (see also, Zhang, 2010). Indeed,
by examining LLS use with reference to specic
skill areas (e.g., listening, reading, vocabulary,
or writing), it has been possible to address the
criticisms of strategy taxonomies and to begin to
operationalize strategy use more rmly (Cohen
& Macaro, 2007; Macaro, 2006; Woodrow, 2005).
In an additional step, Dörnyei and associates
(e.g., Dörnyei, 2005; Tseng et al., 2006) have
suggested rejuvenating LLS research in terms
of a self-regulation mechanism, which concep-
tualizes learning strategies from the perspective
of students’ capacity to manage their own learn-
ing. In fact, some researchers (e.g., Cumming,
Busch, & Zhou, 2002) have proposed that en-
hancing learners’ self-regulatory capacity with a
goal orientation is the central tenet of the SRL
process. Their proposal has helped L2 scholars
reorient LLSs as goal-directed activities in rela-
tion to specic L2 contexts. Such a perspective
enables LLSs to be further explored within the
framework of SRL, which is interpreted 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. 453).
An SRL perspective would be especially con-
ducive to promoting active and productive
learning in specic L2 language skill areas and
learning contexts. Tseng et al. (2006), for ex-
ample, devised the Self-Regulation Capacity in
Vocabulary Learning Scale to assess the capacity
of strategic learning within an SRL framework.
Conrmatory factor analysis validated a 5-factor
model that included commitment control, emo-
tional control, metacognitive control, satiation
control, and environment control. Subsequent
exploratory factor analysis revealed the unidi-
mensionality of the 5 indicators in the instrument
with satisfactory psychometric properties. Their
ndings provided support for the appropriate-
ness of using the construct of self-regulation in
the specic case of vocabulary learning.
Encouraging as those ndings are, what re-
mains to be accounted for is the multidimension-
ality of the SRL construct, which also includes
metacognition, as Gao (2007) and Zhang (2010)
have argued, and its use in different educational
contexts. For example, Zhang et al.’s (2008) inves-
tigation of SRL strategies of Singaporean primary
school students in a bilingual/biliteracy learning
context showed a signicant relationship between
students’ use of SRL strategies and their language
prociency. However, their study only reported
cognitive strategies in bilingual reading contexts
without reference to the interplay of other SRL di-
mensions, such as metacognition, motivation, and
social behavior.
Based on this research evidence, we postu-
late that a particular strength of SRL, just like
for LLS research, is its heavy emphasis on the
learning process and learners’ pivotal role in it.
In addition, both SRL and LLS are overarch-
ing terms that include cognitive, metacognitive,
social–behavioral, and motivational components.
This makes it possible to incorporate the control
mechanism of cognition, behavior, environment,
and motivation and to begin to explore various
dimensions of learners’ development of SRL ca-
pacity in specic contexts, such as L2 writing.
We therefore suggest that L2 writing as a process
be usefully examined from a multidimensional
4The Modern Language Journal 100 (2016)
perspective that includes an understanding of
how learners set goals; attempt to monitor, regu-
late, and control their cognition, motivation, and
behavior in the learning process; and consider
how these aspects, in turn, are often guided and
constrained by learners’ goals and diverse contex-
tual features.
L2 WRITING RESEARCH
L2 writing research has been strongly inu-
enced by L1 writing theories with different foci
(Silva & Matsuda, 2010). The effects of cognitive
manipulation on the quality of written texts (e.g.,
Ong & Zhang, 2013), the inuence of the L1
on the L2 (e.g., Roca de Larios, Martin, & Mur-
phy, 2001), the function of written feedback (e.g.,
Hyland & Hyland, 2006; Lee, 2014), the role of
learner differences (e.g., Kormos, 2012), and the
effects of genre or task-based writing instruction
(e.g., Byrnes, 2014b; Hyland, 2007) have all been
explored. The next section discusses in greater de-
tail yet another prominent research area, that of
L2 writing strategies.
Classications of L2 Writing Strategies
Driven by the proliferation of research on pro-
cess writing in L1 English in the 1980s in North
America, L2 writing researchers (e.g., Leki, 1995;
Roca de Larios et al., 2008) turned to investigat-
ing and classifying specic strategies on the part
of L2 writers. However, their efforts were ham-
pered by the “shaky theoretical foundation of
research on learner strategies” (Manchón, Roca
de Larios, & Murphy, 2007, p. 230), well mani-
fested in the lack of an agreed-upon operational
classication of writing strategies. At present,
the classication of writing strategies is grounded
in an array of theoretical paradigms, among them
cognitive models of L1 writing with a focus on
process writing (Flower & Hayes, 1981), empiri-
cal research on language learning and language
use strategies in general contexts (O’Malley &
Chamot, 1990; Oxford, 1990), a sociocognitive ap-
proach (Leki, 1995), and goal theories in educa-
tional psychology (Cumming et al., 2002).
The process writing approach itself originated
from cognitive theory based on the three pri-
mary cognitive processes: planning, translating,
and reviewing (Flower & Hayes, 1981). Accord-
ingly, writing strategies were generally classied
in terms of three phases: pre-writing, drafting
(composing), and after writing (revising). A pro-
fusion of research studies followed, either ad-
dressing all three processes (e.g., Cumming, 1989;
Sasaki, 2000) or focusing on one specic type
such as planning (Victori, 1999), composing
(Chenoweth & Hayes, 2001), or revision (Berg,
1999). Some researchers explored specic ac-
tions related to writing processes, such as orga-
nizing the content for planning, translating in
the writing process, editing after draft completion
(Cumming, 1989; Victori, 1999); or they focused
on metacognitive strategies with the tripartite
distinction of planning, monitoring, and evalua-
tion (Wenden, 1991; Zhang, Aryadoust, & Zhang,
2016). Finally, yet other researchers applied learn-
ing strategy classications to L2 writing to ex-
plore multiple dimensions of L2 students’ writing
strategies (see Oxford et al., 2014, for more in-
formation). Not surprisingly, such a classication
of writing strategies was strongly inuenced by
two prominent taxonomies of general LLSs: (a)
O’Malley & Chamot’s (1990) three broad types
of learning strategies: cognitive, metacognitive,
and social-affective; and (b) Oxford’s (1990) six
factor strategies, which distinguish direct (mem-
ory, cognitive, and compensation) from indirect
(metacognitive, affective, and social) strategies.
While these classications have been widely used
(see, e.g., Oxford, 2013), they have not been
sufciently validated with specic populations or
learning contexts. Only Hsiao and Oxford (2002)
evaluated different functional classications of
LLSs with EFL students in Taiwan through conr-
matory factor analyses; they found that Oxford’s
(1990) six part taxonomy had the best t when
compared with other models (e.g., O’Malley &
Chamot, 1990).
As social dimensions have increasingly been
adduced in order to illuminate the composing
process, many researchers have reclassied writ-
ing strategies in terms of the sociocognitive na-
ture of writing activities. For example, Leki (1995)
reclassied writing strategies into 10 categories
that included clarifying and focusing strategies for
completing a task, using previous knowledge or
experience, making use of social contexts such
as using training or others’ feedback, taking a
stance toward teachers’ demands, and managing
time and learning efforts. She highlighted the
need to depict “the fullest range possible of strate-
gies employed” (Leki, 1995, p. 240), an emphasis
that was echoed by Riazi (1997), who concluded
that literacy production is “an interactive social-
cognitive process in that production of the texts
required extensive interaction between the indi-
vidual’s cognitive process and social/contextual
factors in different ways” (p. 105).
In another attempt to provide a more solid
ground for strategies research, researchers such
Lin Sophie Teng and Lawrence Jun Zhang 5
as Cumming et al. (2002) have used goal theories
in educational psychology as a frame of reference
and have dened writing strategies as “the level
of strategic operations for the activity of writing”
directed by goals (p. 193). They favored a re-
search path in which strategies “are analyzed in
reference to the goals people have to motivate
and guide their task performance as well as other
essential aspects of these activity structures and
the contexts in which they are embedded” (Cum-
ming et al., 2002, p. 193). They grouped writing
strategies into ve categories while integrating
self-regulation as an essential component that in-
cluded peer learning from others, self-regulation,
stimulation, and use of tools for resourcing.
Because this line of classication integrates social
and emotional regulation in order to capture
qualities of writing strategies, it shows important
similarities with the categories of SRL strategies
proposed by Zimmerman and Martinez–Pons
(1986). In other words, research on L2 writing
strategies has shifted from a purely cognitive
to a sociocognitive orientation, acknowledging
the cultural and contextual factors to reveal the
multidimensional nature of the writing process
(Silva & Matsuda, 2010). Even so, an empiri-
cal evaluation of an appropriate way to classify
learning strategies still remains to be presented
and validated for the specic context of L2
writing.
Research on L2 Writing Strategies
As with general learning strategies, research on
L2 writing strategies, too, has been strongly in-
uenced and directed by insights gained in L1
contexts. In line with the theoretical understand-
ing of L1 composing processes, numerous studies
(e.g., Raimes, 1985; Sasaki, 2000, 2004) have ex-
plored how learners’ language prociency (i.e.,
high prociency and low prociency) and learn-
ing experience (i.e., novice/expert, skilled/less
skilled, successful/less successful) inuence the
use of certain types of writing strategies (see also
Manchón et al., 2007, for a review). While ear-
lier studies evaluated different cognitive learning
strategies in the L2 writing process with a focus on
planning, composing, and revising, more recent
work has foregrounded the writing process as a
“socially situated, cognitive, communicative activ-
ity” (Manchón et al., 2007, p. 229). Then followed
a plethora of L2 studies that have taken a more
inclusive view of writing strategies and explored,
in particular, how writing prociency signicantly
inuences the use of a range of writing strategies
(e.g., Gordon, 2008; Zhang et al., 2016).
Another strand of research has investigated
how L1 strategies and language prociency in-
uenced the use of L2 writing strategies. Some
studies (Cumming, 1989; Silva, 1993) found that
L2 writing strategies are similar to L1 writing
strategies but strategy deployment is inuenced
by learners’ L1/L2 language prociency. More-
over, the transfer of L1 strategies to the L2 is in-
uenced by cultural and contextual factors and
writing processes (see Manchón et al., 2007, for a
summary).
A majority of these studies adopted qualitative
methods that drew on concurrent or retrospec-
tive introspection techniques (e.g., think-aloud
protocols and interviews) and worked with small
samples of participants in order to solicit situa-
tional and individualized writing strategies (e.g.,
Raimes, 1985; Sasaki, 2000, see also Grifths &
Oxford, 2014, for a review). However, only a few
studies (e.g., Cumming, 1989) reported the inter-
rater and intra-rater reliabilities of the qualita-
tive coding and analysis, a fact that makes it dif-
cult to interpret the data (see, e.g., Manchón
et al. 2007). Some studies used self-developed or
modied questionnaires to depict the pattern of
writing strategies used by learners (e.g., Gordon,
2008). Unfortunately, such questionnaires were
largely ad hoc creations, without being theoreti-
cally validated in relation to specic sociocultural
contexts. One way to overcome some of these
shortcomings is through the use of validated ques-
tionnaires with clear reliabilities, an approach ad-
vocated by Petri´
c and Czárl (2003) for being able
to provide holistic information on a large scale for
researchers to depict the pattern of students’ use
of writing strategies in a specic learning context.
EXISTING QUESTIONNAIRES FOR
EVALUATING LLS AND SRL STRATEGIES
O’Malley and Chamot (1990) pointed out that
questionnaires are a useful method for solicit-
ing individual perceptions and interpretations of
students’ own learning experience that can pro-
vide explanations for behavior (e.g., trait-like fea-
tures). Large-scale questionnaires, in particular,
enable the use of quantitative methodologies as
a way of developing a model for understanding
how writing strategies interact with other factors
such as social and psychological variables in L2
contexts (Hsiao & Oxford, 2002; Petri´
c & Czárl,
2003; Zhang et al., 2016).
In the literature, three questionnaires devel-
oped to evaluate LLSs in L2 settings and SRL
strategies in general learning contexts stand out:
the Strategy Inventory of Language Learning
6The Modern Language Journal 100 (2016)
(SILL; Oxford, 1990), the Learning and Study
Strategies Inventory (LASSI; Weinstein et al.,
1987), and the Motivated Strategies for Learning
Questionnaire (MSLQ; Pintrich et al., 1991). The
rst, the SILL (Oxford, 1990), has been widely
used to investigate learning strategies in L2 con-
texts. The SILL is a 5-point scale questionnaire
ranging from 1 (never use it) to 5 (often use it).
The instrument has been used to address specic
strategic behavior, including six types of strate-
gies broken down into: (a) memory strategies,
(b) cognitive strategies, (c) compensation strate-
gies, (d) metacognitive strategies, (e) affective
strategies, and (f) social strategies. Some re-
searchers have developed their versions of inven-
tories based on the SILL to investigate writing
strategies (e.g., Petri´
c & Czárl, 2003). Although
this instrument has been extensively used in vari-
ous contexts, especially in general language learn-
ing environments, it is not above criticism. As
Tseng et al. (2006) argued, the scales of specic
strategies in the SILL are not cumulative, which
makes it impossible to “assume a linear relation-
ship between individual item scores and the to-
tal item scores” (p. 83). That shortcoming has
marred the application of the SILL to specic con-
texts with robust psychometric properties.
The second instrument, the LASSI, was de-
veloped by Weinstein et al. (1987) as a diag-
nostic and prescriptive instrument within SRL
theory to evaluate students’ awareness about
and use of learning strategies in terms of skill,
will, and self-regulation. It is an 80-item self-
report inventory with a 5-point scale ranging
from 1 (not at all typical of me) to 5 (very
much typical of me). Of the 10 scale strategies,
the Skill categories include concentration, se-
lecting main ideas, and information processing;
the Will categories include motivation, attitudes,
and anxiety; the Self-regulation categories com-
prise time management, study aids, self-testing,
and test strategies. Coefcient alphas of the 10
scale strategies range from.76 to .87. The inter-
nal correlations between these scales range from
.13 (between information process and anxiety)
to .59 (between time management and motiva-
tion), revealing the discriminant validity of the
measurement.
Pintrich et al. (1991) developed a self-report in-
strument, the MSLQ, with 7-point Likert scales
ranging from 1 (not at all true of me) to 7
(very true of me). The MSLQ was designed
to evaluate two distinct constructs: motivation
(31 items) and learning strategies (50 items)
of college students in classroom environments.
The motivational scales include the assessment
of what Pintrich et al. referred to as students’
value (extrinsic and intrinsic goal orientation
and task value), expectancy (control belief and
self-efcacy), and affect (test anxiety); the learn-
ing strategy scales include students’ use of cog-
nitive strategies (rehearsal, elaboration, orga-
nization, and critical thinking), metacognitive
strategies (planning, monitoring, and regulat-
ing strategies), and resource management (ef-
fort management, time and environment man-
agement, and help-seeking). The motivation and
learning strategy sections correspond to the three
elements in the denition of SRL: motivation,
metacognition, and behavior. The instrument is
intended as a coherent framework in which the
15 different subscales on the MSLQ could be
utilized together or singly by taking the mean
of the items corresponding to each factor. The
MSLQ has undergone extensive psychometric
testing and the overall internal consistency reli-
ability (Cronbach’s alpha) was found to be ade-
quate for a motivation scale (α=.78) and a learn-
ing strategy scale (α=.71), respectively. How-
ever, conrmatory factor analysis did not gener-
ate satisfactory goodness-of-t indices for the mo-
tivation sections (GFI =.77; AGFI =.73; RMR
=.07), or for the learning strategy subscales
(GFI =.78; AGFI =.75; RMR =.08) (Pintrich
et al., 1991). The goodness-of-t indices were
lower than the recommended benchmark values
(Hu & Bentler, 1999), indicating that the MSLQ’s
construct validity might not be reliable enough,
particularly if it is applied in other learning
contexts.
Findings from these instruments clearly sup-
port utilizing questionnaires to evaluate learning
strategies. Also, the popularity of the MSLQ and
the LASSI produced clear evidence that SRL was
an important construct that merits further re-
search. However, the utility of these instruments
has been subject to criticism. As noted earlier,
critics initially argued against the broad-brush in-
vestigation of general learning strategies, which
intrinsically marred the conceptual framework
and classication of these instruments. In addi-
tion, contextual and cultural differences have not
been fully considered and explored. For exam-
ple, the LASSI and the MSLQ inventories were
widely applied in educational contexts, such as
sports, music, and L1 reading and writing, with-
out sufcient empirical validity in L2 specic
contexts.
In sum, how to evaluate writing strategies for
SRL in EFL settings is far from resolved and
needs to be addressed in relation to specic
learning contexts. Instruments that were initially
Lin Sophie Teng and Lawrence Jun Zhang 7
developed for L1 or L2 learners in general learn-
ing contexts cannot be directly applied to EFL
writing contexts, nor can their data be treated as
reliable for providing insights into specic writ-
ing issues. Such a lacuna in the research literature
calls for the development of a theoretically more
robust instrument with strong psychometric prop-
erties to evaluate the use of writing strategies for
SRL in EFL environments.
WRITING INSTRUCTION IN CHINA
Although writing is a required component in
the English language teaching curriculum in Chi-
nese universities, it is commonly regarded as the
most challenging language skill and one in which
Chinese students nd it difcult to develop com-
petence above low performance levels (Lei, 2008;
Zhang, 2013). Despite over six years of learning
English in schools and even with continuous prac-
tice in university, Chinese students’ improvement
in writing tends to be limited (Wang, 2014; Zhao,
2010). Students’ slow progress in writing skills is
also related to how English writing is taught. Be-
cause the teaching of EFL writing is test-driven
and product-oriented, teachers pay little attention
to cultivating students’ own interest in and moti-
vation for writing; nor do they have time for fos-
tering different learning strategies to help them
to learn to write (Teng & Zhang, 2016).
The pedagogy for undergraduates favors mas-
tery of comprehensive knowledge of English in
terms of grammar, vocabulary, and reading, while
speaking and writing courses are usually offered
as electives. In turn, writing instruction in univer-
sities aims to develop students’ declarative knowl-
edge ranging from “micro-skills such as orthogra-
phy and sentence level writing to macro skills such
as paragraph and whole text writing” (Woodrow,
2011, p. 511). Although teachers do ask students
to practice writing in order to pass the writing
test as required in the national examinations such
as the CET-Band 4 or the CET-Band 6 in the
classroom, they rarely give writing assignments for
students to complete after class. Scholars such
as Zhao (2010) and Teng (2016) have pointed
out that the curricula, teaching syllabuses, in-
structional materials, and forms of evaluation are
prescribed by authorities such as the faculty or
administrative committees who are in charge of
a particular specialty. This product-oriented and
teacher-centered approach allows for little in-
structional variety or student involvement in the
learning process and is itself a consequence of
teachers’ poor training in the teaching of L2 writ-
ing (Lee, 2014; Zhang, 2016).
RESEARCH ON WRITING STRATEGIES
IN CHINA
Incorporating aspects of process-oriented writ-
ing instruction as Western pedagogies, research
on writing strategies in China has focused on de-
scriptive studies, the transfer of L1 writing strate-
gies to L2 writing, individual differences in the
use of strategies, and the effect of strategies-based
writing instruction (e.g., Lei, 2008; Nguyen & Gu,
2013; Wang & Wen, 2002). Previous studies have
foregrounded individual differences in terms of
age, motivation, self-efcacy, and language pro-
ciency and have found that Chinese students re-
ported using lower level writing strategies. Also,
they have tended to explore writing strategies un-
der the optic of a typical process approach with
multiple drafting, revising, and teacher or peer
feedback. In other cases they have focused on
some cognitive and/or metacognitive strategies
without systematic instructional guidance or the-
oretical foundation. By comparison, little atten-
tion has been given to exploring other categories
of writing strategies (e.g., social and affective
strategies) from SRL theory in EFL writing (Teng
& Zhang, 2016). In addition, as Zhang (2016)
observed, EFL practitioners often have limited
knowledge of their students’ strategy repertoires,
the role of individual differences inuencing stu-
dents’ use of writing strategies, and the ways to
integrate task-based writing strategies into their
teaching.
Given the essential role of writing strategies
in inuencing students’ learning outcomes (see
Plonsky, 2011, for a meta-analysis), it is critical
to know what strategies EFL Chinese students ac-
tually do use and which strategies help improve
their writing performance. Assuming that a large-
scale inquiry is preferable, a reliable and valid in-
strument for targeting specic EFL writing con-
texts with a sound theoretical framework needs to
be developed.
A MULTIDIMENSIONAL MODEL OF L2
WRITING STRATEGIES FOR SRL
Informed by Oxford’s (2013) denition of
LLSs, we redene L2 writing strategies for SRL as
deliberate, goal-directed attempts to make writing
enjoyable, less challenging, and more effective.
According to Zimmerman & Risemberg (1997),
self-regulation of writing refers to self-initiated
thoughts, feelings, and actions that writers use
to attain various goals, including improving their
writing skills and knowledge; enhancing the qual-
ity of the text they create; and sustaining learning
8The Modern Language Journal 100 (2016)
effects in either learning-to-write or composing
processes. While applied linguistics often makes
a functional distinction between strategies for
learning and strategies for using the language
(Wenden, 1991, see also A. D. Cohen, 2014), our
study takes an inclusive and comprehensive view
of writing strategies for SRL in that it encompasses
a range of strategies used by L2 writers for devel-
oping linguistic knowledge, for producing writ-
ten language, or for completing a task. To us,
both purposes play a critical role in fostering self-
regulated learners who “can approach challeng-
ing tasks and problems by choosing from a reper-
toire of tactics those they believe best suited to
the situation, and applying those tactics appropri-
ately” (Winne & Perry, 2000, pp. 533–534).
For the purposes of this study, we follow Zim-
merman’s (1989) understanding of SRL as a
dynamic, multidimensional process in which
learners are “metacognitively, motivationally, and
behaviorally” active in regulating their own learn-
ing (p. 329). Informed by sociocognitive theory,
our approach to self-regulation emphasizes the
reciprocal determinism of the environment on
the person, mediated through behavior (Ban-
dura, 1991). In writing contexts, environmental
processes refer to writers’ self-regulation of the
social setting in which they write, behavioral pro-
cesses pertain to writers’ self-regulation of overt
motoric activities associated with writing, and
personal processes involve writers’ self-regulation
of cognitive beliefs and affective states associated
with writing (Dinsmore, Alexander, & Loughlin,
2008). Following Zimmerman and Risemberg
(1997), we hold that “each of these triadic forms
of self-regulation interacts reciprocally via a cyclic
feedback loop through which writers self-monitor
and self-react to feedback about the effective-
ness of specic self-regulatory techniques or
processes” (p. 73). Therefore, explanations for
self-regulation development will need to address
the role of cognitive and metacognitive, social–
behavioral, and motivational processes (see also
Alexander, Graham, & Harris, 1998; Andrade &
Evans, 2012).
Cognitive and Metacognitive Strategies
Cognitive strategies refer to skills students
use to process the information or knowledge in
completing a task (Pintrich et al., 1991). They
help learners construct, transform, and apply
L2 knowledge (Oxford, 2013). A plethora of
research has conrmed the essential role of
text processing, organization, and rehearsal in
fostering active engagement in learning and high
levels of academic achievement (e.g., Pintrich &
De Groot, 1990; Winne, 2011; Zhang et al., 2008).
In turn, metacognitive strategies refer to the
skills used to control and regulate learners’ own
cognition and the cognitive resources they can
apply to meet the demands of particular tasks
(Winne, 2011). Many studies have conrmed
the signicant role of metacognitive strategies
such as goal setting, planning, monitoring, and
evaluating in cultivating L2 prociency (e.g.,
Zhang, 2010). According to Oxford (2013), “the
cognitive and metacognitive strategies facilitate
understanding, increase meaningful mental
associations, and are the most useful strategies
for long-term retention of information” (p. 30);
all contribute to deep processing. By including
cognitive and metacognitive strategies it should
be possible to ascertain students’ active role in
learning to write; in turn, having this kind of eval-
uative instrument might raise students’ awareness
of what cognitive/metacognitive strategies can
facilitate their learning, thus contributing to
better academic performance.
Social Behavioral Strategies
As a key aspect of self-regulation, social–
behavioral strategies involve individuals’ attempts
to control their learning behavior under the in-
uence of contextual and environmental aspects
(Zimmerman, 1989, 2011). Although there is
a wide range of strategies subsumed by such a
construct, we only focus on two subcategories
involved in feedback-handling and peer learning.
SRL models emphasize the importance of feed-
back loops in which learners monitor, evaluate,
and adjust strategies, goals, and motivational
factors in a given task (Zimmerman, 2013).
Therefore, how learners handle others’ feedback
mediates the use and adjustment of other strate-
gies, and, in turn, affects learning outcomes.
Also, the development of SRL relies on the social
mediation and interactive support from teach-
ers and peers, which is benecial to learners’
active learning and enhancement of motivation
(Schunk & Rice, 1986). Self-regulated learners do
not work in isolation. When faced with a complex
task, they either seek help from others who are
knowledgeable such as their peers, family mem-
bers, and teachers, or consult written resources
(Zimmerman & Risemberg, 1997).
Although previous studies have conrmed
the essential role of social behavior strategies
(e.g., Zimmerman & Martinez–Pons, 1986), little
attention has been paid to the evaluation of
the construct in EFL writing through empirical
Lin Sophie Teng and Lawrence Jun Zhang 9
research. In this study, we propose that social
behavior strategies, as a distinct construct of
the SRL mechanism, interact with cognition,
metacognition, and motivational regulation,
which, taken together, contribute to EFL writing
performance. Ferris and Robert’s (2001) study
has shown how important teachers’ feedback
on earlier drafts of work can be in inuencing
learners’ writing processes. Peer response is also
valuable in providing feedback and extending
the writer’s audience beyond just the teacher.
Indeed, it has been argued that developing self-
regulated learners in classroom activities requires
peer interaction; an additional benet is that it
also contributes to constructing a cooperative
learning environment (Zimmerman, 2013).
Motivational Regulation Strategies
Motivational regulation strategies are dened
as the procedure or thoughts that students
apply purposefully to sustain or increase their
willingness to engage in a task (Wolters, 1999).
According to Zimmerman & Risemberg (1997),
writing “is a social cognitive process wherein
writers must be aware of readers’ expectations
and must be willing to devote the personal time
and effort necessary to revise text drafts until they
communicate effectively” (p. 76). In other words,
writing achievement is contingent on the degree
of individuals’ motivational control to use the
strategies to regulate their writing performance.
Previous studies, albeit limited, have found that
motivational regulation strategies interact with
cognitive, behavioral, and contextual variables in
SRL models (Pintrich, 2004). For that reason, mo-
tivational regulation strategies play a mediating
role in inuencing students’ choice, effort, cogni-
tive engagement, and academic performance in
educational contexts, such as psychology, math,
science, and L1 English (e.g., Wolters & Mueller,
2010). However, whether motivational regulation
strategies also constitute a key component of SRL
strategies in EFL writing as yet awaits empirical
exploration. Based on the tenets of SRL and
existing research, for this study we propose that
motivational regulation strategies be included
in the spectrum of SRL strategies so as to better
characterize the writing process.
To sum up, on the one hand, the choice of the
four dimensional strategies corresponds to the
sociocognitive view of SRL, which acknowledges
the regulation of behavior, person, and environ-
ment. On the other hand, this model is in line
with the interpretation of writing from cognitive
theory, which postulates that writing behavior is
inuenced by “cognitive, affective, and social con-
ditions” (Hayes, 1996, p. 5). Harris et al. (2011),
among other scholars, have regarded writing as
the most complex challenge in developing lan-
guage prociency due to its “recursive, strategic,
and multi-dimensional” characteristics (p. 188).
This view is echoed by scholars in L2/EFL writing
(e.g., Byrnes, 2014a; Manchón, 2009; Silva &
Matsuda, 2010; Zhang, 2013).
THE STUDY
This study aims to validate the Writing Strate-
gies for Self-Regulated Learning Questionnaire
(WSSRLQ) with regard to the proposed multi-
faceted structure of SRL strategies and the
predictive effect of the elicited SRL strategies
on EFL writing prociency. Framed within the
SRL theory, the WSSRLQ was designed to be
multidimensional, measuring the constructs
of cognition, metacognition, social behavior,
and motivational regulation. The instrument
was subjected to a series of conrmatory factor
analyses (CFA) through structural equation mod-
eling (SEM) in order to examine its factorial
structure. We rst hypothesized that the writing
strategies for SRL would be reclassied into 9
correlated factors probing the four dimensions of
self-regulation: cognition, metacognition, social
behavior, and motivational regulation. We then
explored whether there would be a higher order
construct explaining the variance of the 9 sub-
strategies. For evaluating predictive validity, the
study further examined how the elicited 9 SRL
strategies predicted EFL students’ writing pro-
ciency. Two research questions motivated the
study:
RQ1. What structural model best represents
the dimensions of EFL writing strategies
for SRL?
RQ2. Do the EFL writing strategies for SRL
predict EFL writing prociency?
In order to address them, we proposed three
structural models to evaluate the dimensions of
SRL strategies in EFL writing:
Model 1: A 9-Factor Correlated Model of EFL Writ-
ing Strategies for SRL. This model specied 40 items
into nine distinct but correlated writing strategies
framed within SRL theory;
Model 2: A 4-Factor Second-Order Model of EFL Writ-
ing Strategies for SRL. Grounded in SRL theory, we
proposed Model 2 to examine whether the nine
SRL strategies exhibited a hierarchical structure.
We hypothesized that the 9 SRL strategies were
10 The Modern Language Journal 100 (2016)
conceptualized into four second-order correlated
factors, including cognition, metacognition,
social behavior, and motivational regulation.
Model 3: A One-Factor Second-Order Model of EFL
Writing Strategies for SRL. As a competing hierar-
chy model we postulated that a single higher or-
der common factor, self-regulation, might be suf-
cient to account for the correlations of the 9
lower order strategies.
Participants
A total number of 790 undergraduate students
were recruited from six universities in Northeast
China. This was a convenience sample, with par-
ticipants selected from the rst year and the sec-
ond year (rst year N=427, 54%; second year N=
363, 46%). All participants were volunteers from
six majors: Economics (15%, N=119), Civil En-
gineering (16%, N=126), Visual Arts (17%, N=
134), Psychology (14%, N=111), English (25%,
N=197), and Law (13%, N=103). They were
aged between 18 and 22 (M=20.04, SD =1.21);
40% were male (N=316), 60% female. They re-
ported an average of 9 years of formal English lan-
guagelearning(M=9.12, SD =1.04).
Questionnaire Development
The WSSRLQ was developed through a three-
phase process: item generating, initial piloting,
and psychometric evaluation. The process of item
generation for the WSSRLQ began with focus
group interviews. As Dörnyei (2010) argued, in-
volving targeted learners in the item-generating
process adds to the credibility and quality of the
items used in the questionnaire. To this end, 10
undergraduate students were assigned to each fo-
cus group, with each group showing diversity in
terms of gender, year level, and disciplinary major.
During the focus group interviews, participants
were invited to describe what strategies they used
while performing a writing task or learning writ-
ing knowledge inside and outside the classroom.
The analysis of the focus group data helped gen-
erate the initial items (see Appendix A for a list of
guided interview questions).
We also consulted the relevant literature and
critically examined established instruments for
evaluating SRL strategies (e.g., Pintrich et al.,
1991; Wolters, 1999; Zimmerman & Martinez–
Pons, 1986). As argued by many researchers (e.g.,
Dörnyei, 2010; Petri´
c & Czárl, 2003), such an
item-generation procedure lends construct valid-
ity to a questionnaire.
Given the wide range of SRL strategies in the
proposed four dimensions, the EFL writing strate-
gies in our questionnaire were selective and syn-
thetic, rather than fully inclusive based on the
following criteria: (a) EFL students in the focus
group interviews mentioned that they had used
these writing strategies in completing a writing
task inside and outside the classroom, (b) the
writing strategies reected any of the four dimen-
sions of cognition, metacognition, social behav-
ior, and motivational regulation, and (c) the writ-
ing strategies were essential to promoting SRL, as
supported by the research literature.
A list of 45 items pertaining to EFL writing
strategies was generated. Three experts in the
eld of LLSs and SRL were invited to examine
the initial list. Specically, they scrutinized the
theoretical rationale, checked the questions for
the constructs being measured, and rated the de-
gree to which the survey questions matched the
constructs as dened in the study. This procedure
resulted in the elimination of items that had re-
ceived the lowest rating. The revised list was then
given to 15 EFL students who checked the items
for clarity and readability. As a result, irrelevant
and double-barreled items were eliminated and
related statements were combined. The nal in-
strument containing 40 items was sequenced log-
ically and organized according to the clusters of
subcategories. In line with Pintrich et al. (1991)
and Tseng et al. (2006), a 7-point Likert scale with
gradation rating from 1 (not at all true of me) to
7 (very true of me) was adopted to explore the
trait features of SRL strategies; the mean scores
of these items were made cumulative.
Given the level of English language prociency
of these participants, the English questionnaire
was translated into Chinese by the rst author and
then veried by the second author, both of whom
are L1 Chinese speakers. Back translation was ap-
plied to ensure parallelism of bilingual versions
of the instrument. The questionnaire was then
subjected to more statistical scrutiny to determine
construct validity using CFA.
Writing Test
In order to evaluate students’ writing pro-
ciency, we adopted a given-topic argumentative
writing test selected from the International En-
glish Language Testing System (IELTS) Task 2,
which is one of the most widely used English-as-
a-second-language (ESL) tests around the world.
The writing task requires an extended compo-
sition in response to a proposition or question.
The argumentative writing test has been proven
Lin Sophie Teng and Lawrence Jun Zhang 11
to be effective in evaluating students’ academic
achievements in light of linguistic compe-
tence, critical thinking, and articulation of ideas
(Varghese & Abraham, 1998). The production
of an argumentative essay requires students’
active use of a range of strategies such as using
knowledge to generate ideas, monitoring and
evaluating their progress, revising the written es-
say, or actively regulating their emotion to sustain
their learning efforts throughout the composing
process. Active use of these writing strategies in
the composing process reects students’ proac-
tive engagement with a writing task (Zimmerman
& Risemberg, 1997). The topics or contexts of
language in the IELTS are designed to avoid bias
against any group of candidates of a particular
background.
In our study, all participants were required to
produce a written argument on a given topic
with at least 250 words within an hour in class-
room environments. Then their writing scripts
were marked based on the standard IELTS Task
2 analytic scales, which focus on task response,
coherence and cohesion, lexical resource, and
grammatical range and accuracy. Two evaluators
who were certied IELTS examiners were invited
to evaluate these essays. The interrater reliabil-
ity and intrarater reliability were at .88 and .92,
respectively.
Data Analysis
The data were subjected to CFAs using SEM
through the IBM SPSS AMOS computer program,
Version 22 (Arbuckle, 2013). CFA is a useful sta-
tistical measure to test a theoretical model and a
tighter specication of multiple hierarchies by uti-
lizing the factor, correlation, and covariance pat-
terns, and residual or error values within a data
matrix (Kline, 2011). In this study, the maximum
likelihood (ML) estimation method was applied
to evaluate the three hypothesized models. While
there are no golden rules for assessing model t,
we interpreted the CFA data based on the sev-
eral omnibus t statistics to test whether a spe-
cic model of interest adequately ts the data be-
cause different indices reect different aspects of
model t. As suggested by Kline (2011), we rst
reported a chi-square statistic, along with its de-
grees of freedom (df) and associated pvalue. The
ratio of chi-square x2divided by the df <.30 indi-
cates the best t with the nonsignicant pvalue.
However, due to the sample-size dependency of
the chi-square test statistic, signicant values of
chi-square are usually found when large sample
sizes are involved in CFA.
Therefore, we enlisted other t indices that in-
cluded the Goodness of Fit Index (GFI), the Root
Mean Square Error of Approximation (RMSEA)
with its corresponding 90% condence interval,
the Standardized Root Mean Square Residual
(SRMR), the Comparative Fit Index (CFI), and
the Tucker–Lewis Index (TLI). The GFI is a
commonly reported t index to measure the hy-
pothesized model and the observed covariance,
with values of over .90 generally indicating an ac-
ceptable model t. The RMSEA is a critical value
in this study because it assesses the model t while
taking into the account the complexity of the
model structure. A value of .05 or less is indicative
of good model t. Unlike the chi-square statistic,
both the SRMR and the CFI are not sensitive to
sample size. Values for the SRMR less than .08
and values of CFI larger than .90 are generally
indicative of acceptable model t. TLI analyzes
the discrepancy between the chi-squared value
of the hypothesized model and the chi-squared
value of the null model. It is not sensitive to
sample size but depends on the average size of
the correlations in the data. The recommended
values of TLI are equal to or more than .90
indicating an acceptable level of model t (see
Hu & Bentler, 1999, for more information).
In this study, chi-square statistics were also used
to help us select the appropriate structural model
when comparing the hypothesized nested mod-
els. The difference in chi-square as a ratio of the
difference in df was examined with the signicant
p-value indicating that the reference model is a
better t for the data (Kline, 2011).
To evaluate the predictive validity of the instru-
ment, a simultaneous multiple regression analysis
was used to investigate how the 9 SRL strategies
predicted EFL learners’ writing performance. As-
sumptions of multiple regressions were rst exam-
ined. Given that the variables (9 SRL strategies) in
this set of analysis were correlated, a Bonferroni
adjustment was adopted to set the alpha level for
the overall test of each regression to be a more
conservative value.
Procedures
The questionnaire was given to participants af-
ter a writing course to elicit authentic context-
based strategies. Instructions were reviewed and
claried rst, and any doubts and comments were
recorded and addressed. Participants spent on av-
erage 10–15 minutes completing the Chinese ver-
sion of the WSSRLQ. On the following day, the
students were given an in-class writing test, which
was used to elicit their writing prociency.
12 The Modern Language Journal 100 (2016)
Data collected from the WSSRLQ were
screened and cleaned rst. Missing responses,
normality, and homogeneity for multivariate anal-
yses were examined thoroughly. Data collected
from the completed WSSRLQ were subjected
to a series of CFAs to evaluate the hypothesized
dimensions of EFL writing strategies for SRL.
They were then correlated with the participants’
writing scores in a regression model to check how
the elicited SRL strategies affected students’ EFL
writing prociency.
RESULTS
Descriptive Statistics and Normality Check
Descriptive statistical analyses showed that the
mean scores of the 40 items ranged from 3.13 to
5.72 with standard deviations ranging from .94 to
1.73. The values for skewness were between −1.23
and .35 and the values for kurtosis were between
−1.03 and 2.17. They were far less than the cutoff
values of ±3.0 and ±8.0 for skewness and kur-
tosis, respectively, indicating the univariate nor-
mality of the response (Kline, 2011). Appendix B
shows the means, standard deviations, skewness,
and kurtosis of the 40–item questionnaire.
Initially, four cases with systematic response
bias (e.g., same response for the entire question-
naire) were eliminated. In addition, six cases
with missing values were removed without impu-
tation because the total proportion of missing
values was far less than the cutoff value of 5%
(Enders, 2010). Thus, a nal sample size of
780 participants for a 40-item scale met the de-
sired cases-to-variables ratio (5:1) analysis (Field,
2009). The assumptions of linearity, singularity,
and homogeneity of the sample were satised
and no outlying cases were detected.
Multivariate normality was examined using
Mardia’s normalized multivariate kurtosis value
(equivalent to a zscore). Normalized coefcients
greater than 3.00 are indicative of nonnormality
(Field, 2009). In the AMOS software program, the
multivariate kurtosis value of 355.42 represented
Mardia’s coefcient of multivariate kurtosis, the
critical ratio of which was 33.72, so these data
were multivariate nonnormal. In order to test
whether nonnormality inated the signicance
of the regression paths, Bootstrap ML estimates
were performed to provide bias-corrected con-
dence intervals for each bootstrap estimate.
Findings showed that all of the signicant paths
kept the same signicance as the original re-
sults, indicating that the nonnormality of these
variables was not able to affect the proposed
paths’ signicance.
Evaluating A Nine-Factor Correlated Model of EFL
Writing Strategies for SRL
Based on the theoretical design, we rst tested
a 9-Factor Correlated Model (Model 1) as we had
hypothesized (see Figure 1).
This model specied 40 items into 9 distinct
but correlated writing strategies framed within
SRL theory. In Figure 1, rectangles represent
observed variables (items in a questionnaire)
and ovals indicate unobserved variables (latent
variable/factor). A one-headed arrow indicates a
hypothesized one-way direction, whereas a two-
headed arrow indicates a correlation between
two variables. In this model, the 9 latent factors’
loadings were each xed to be one and did not
have to be estimated. The 40 observed variables
had measurement errors. Each indicator was
constrained to load only on the factor it was
designed to measure. Factor covariances were
free to be estimated and error terms associated
with each indicator were uncorrelated. Each item
pair measure had a nonzero loading on a specic
writing strategy that the questionnaire was de-
signed to measure, and a zero loading on all other
factors.
Results of a CFA revealed an acceptable model
t (x2780 =1676; df =743; p<.001; x2/df =
2.25; GFI =.91; TLI =.91; CFI =.92; RMSEA
=.045 [.042, .049], SRMR =.056), with 40 items
loading on 9 correlated factors as designed. Fig-
ure 2 shows the standardized results for the 9-
factor correlated model. The parameter estimates
presented here are all standardized as this facili-
tates the interpretation of parameters along with
standard errors. Standardized estimates loadings
(factor loading) from the factors to the observed
variables are higher than the benchmark value
.50, thereby suggesting an acceptable effect size
(Raykov & Marcoulides, 2008). In this model, all
40-item parameter estimates were statistically sig-
nicant at p<.001.
Inter-Correlations of the 9 EFL Writing Strategies for
SRL. The CFA results also conrmed the dis-
criminant validity of the 9 EFL writing strategies as
evidenced by the small to moderately strong cor-
relations of the 9 factors. As depicted in Table 1,
inter-correlation coefcients ranged from r=
.12 between peer learning (PL) and emotional
control (EC) to r=.56 between goal-oriented
monitoring and evaluating (GME) and peer
learning (PL). All 9 factors were signicantly
Lin Sophie Teng and Lawrence Jun Zhang 13
FIGURE 1
9-Factor Correlated Model of EFL Writing Strategies for SRL (N=780).
Note. GME =Goal-Oriented Monitoring and Evaluating; IP =Idea Planning; PL =Peer Learning; FH =Feedback
Handling; IE =Interest Enhancement; EC =Emotional Control; MST =Motivational Self-Talk; TP =Text Processing;
Course Memory =CM.
correlatedwitheachotheratp<.01. Our re-
sults revealed that these 9 factors of EFL writing
strategies for SRL were clearly correlated but also
distinct constructs.
As shown in Table 1, goal-oriented monitoring
and evaluating strategies (GME) of the metacog-
nitive component were strongly correlated with
interest enhancement (IE, r=.52) and peer
learning (PL, r=.56). As regards the cognitive
dimension, text processing strategies (TP) only
had a small correlation with course memory
(CM, r=.29) but were moderately correlated
with idea planning (IP, r=.35) and goal-oriented
monitoring and evaluating (GME, r=.38). In the
motivational regulation dimension, motivational
self-talk (MST) had medium correlations with
7 SRL strategies and a small correlation with
peer learning. All these correlation coefcients
indicated that cross-loadings of the 9 factors were
likely to exist.
14 The Modern Language Journal 100 (2016)
FIGURE 2
A 9-Factor Correlated Model of EFL Writing Strategies for SRL With Standardized Regression Weight
(N=780).
Note. All 40-item parameter estimates were statistically signicant at p<.001. GME =Goal-Oriented Monitoring and
Evaluating; IP =Idea Planning; PL =Peer Learning; FH =Feedback Handling; IE =Interest Enhancement; EC =
Emotional Control; MST =Motivational Self-Talk; TP =Text Processing; Course Memory =CM.
Lin Sophie Teng and Lawrence Jun Zhang 15
TABLE 1
Inter-Correlations for the 9-Factor Correlated Model of EFL Writing Strategies for SRL
Dimensions Strategies TP CM IP GME PL FH IE MST EC
Cognition TP 1
CM .29** 1
Metacognition IP .35** .25** 1
GME .38** .44** .39** 1
Social Behavior PL .25** .27** .25** .56** 1
FH .20** .17** .22** .13** .33** 1
Motivational Regulation IE .35** .21** .32** .52** .25** .24** 1
MST .31** .44** .31** .41** .21** .31** .43** 1
EC .23** .26** .26** .19** .12** .37** .33** .49** 1
Note. TP =Text Processing; CM =Course Memory; IP =Idea Planning; GME =Goal-Oriented Monitoring and
Evaluating; PL =Peer Learning; FH =Feedback Handling; IE =Interest Enhancement; MST =Motivational Self-
Tal k ; E C =Emotional Control; ** =All correlations are signicant at p<.01.
TABLE 2
Means, Standard Deviations, and Internal
Reliabilities of the 9 EFL Writing Strategies for SRL
(N=780)
Dimensions Writing Strategies MSDα
Cognition TP (6 items) 4.67 1.50 .80
CM (3 items) 4.47 1.12 .76
Metacognition IP (3 items) 4.61 0.91 .73
GME (6 items) 3.76 1.26 .86
Social behavior PL (3 items) 3.60 1.11 .80
FH (4 items) 5.61 1.17 .79
Motivational
regulation
IE (4 items) 4.61 1.02 .84
MST (8 items) 4.97 1.43 .87
EC (3 items) 5.17 1.33 .75
Note. TP =Text Processing; CM =Course Memory; IP =
Idea Planning; GME =Goal-Oriented Monitoring and
Evaluating; PL =Peer Learning; FH =Feedback Han-
dling; IE =Interest Enhancement; MST =Motivational
Self-Talk; EC =Emotional Control.
Internal Reliability. Scale reliability tests (Cron-
bach’s alpha) were computed for each of the
9 SRL scales. Table 2 shows that Cronbach’s al-
pha coefcient of the 9 SRL strategies was much
higher than the benchmark value .70, suggesting
a robust internal reliability of each scale. Table 2
presents the means, standard deviations, and
internal reliabilities of the 9 factors.
Model Comparisons
In order to evaluate the hierarchical structure
underlying the 9 writing strategies, we proposed
two high-order models based on the theoretical
framework of SRL. First, we postulated that the
9 SRL strategies conceptualized into 4 higher
order correlated factors, including cognition,
metacognition, social behavior, and motivational
regulation. Figure 3 shows the 4-Factor Second-
Order Model of EFL Writing Strategies for SRL
(Model 2).
Given the cross-loading of the 9 SRL strategies,
we further proposed a One-Factor Second-Order
Model (Model 3), which postulated that a single
common factor, self-regulation, as a higher order,
was sufcient to account for the correlations of
the 9 lower order strategies (see Figure 4).
In both Model 2 and Model 3, each indicator
was constrained to load only on the factor it was
designed to measure, factor covariances were free
to be estimated, and error terms associated with
each indicator were uncorrelated. Table 3 shows
the t indices of the competing models.
Model comparisons showed that both Model 2
and Model 3 had acceptable model t indices. We
conducted further comparisons between Model 1
and Model 2 (x2
M1−x2
M2=18; df
M1−df
M2
=13; p=.16). There was no signicant im-
provement between the two models. However,
the indices of Model 3 (one high order factor
model) improved signicantly in t over Model 2
(x2
M2−x2
M3=14; df
M2−df
M3=4; p=.007)
and Model 1 (x2
M1−x2
M3=32; df
M1−df
M3=
17; p=.015). The signicant difference between
chi-square values suggests that Model 3 was signif-
icantly better than the other two competing mod-
els. On this basis, we retained Model 3 with self-
regulation as a hierarchical construct explaining
the 9 SRL strategies as the model of best t in this
study. For this model, all item parameter estimates
were statistically signicant at p<.001. The 9 SRL
strategies had loaded on the hypothesized high
16 The Modern Language Journal 100 (2016)
FIGURE 3
A 4-Factor Second-Order Model of EFL Writing Strategies for SRL
Note. GME =Goal-Oriented Monitoring and Evaluating; IP =Idea Planning; PL =Peer Learning; FH =Feedback
Handling; IE =Interest Enhancement; MST =Motivational Self-Talk; EC =Emotional Control; TP =Text Processing;
Course Memory =CM.
order latent construct with standardized estimates
over the recommended value of .50 (Raykov &
Marcoulides, 2008). The structure coefcients of
the 9 subcategories ranged between .53 and .79
on the construct of self-regulation, supporting
convergent validity (Kline, 2011). Figure 5 shows
the standardized regression weight of Model 3.
As shown in Figure 5, idea planning (IP), goal-
oriented monitoring and evaluating (GME), mo-
tivational self-talk (MST), and interest enhance-
ment (IE) had large loadings on self-regulation
compared with other writing strategies, revealing
the essential role of metacognition and motiva-
tional regulation in the SRL process.
Lin Sophie Teng and Lawrence Jun Zhang 17
FIGURE 4
A 1-Factor Second-Order Model of EFL Writing Strategies for SRL
Note. GME =Goal-Oriented Monitoring and Evaluating; IP =Idea Planning; PL =Peer Learning; FH =Feedback
Handling; CM =Course Memory; IE =Interest Enhancement; EC =Emotional Control; TP =Text Processing; MST
=Motivational Self-Talk.
Predictive Effect of SRL Strategies on Writing
Prociency
A simultaneous multiple regression analysis was
conducted, in which the 9 SRL strategies were en-
tered as a group in one step. Adjusted Bonfer-
roni value was at .006. Results showed that the
9 SRL strategies as a whole, explained approxi-
mately 37% of the variance in students’ writing
scores, F(9,745) =45.251, p<.001, R2=.37,
adjusted R2=.36. The effect size (R2=.37)
indicates that the 9 SRL strategies as a whole
were a strong factor predicting students’ writing
18 The Modern Language Journal 100 (2016)
TABLE 3
Goodness-of-Fit Indices for Competing Models
Model x2df x2/df CFI TLI GFI RMSEA RMSEA 90% CI SRMR
Model 1 1676*743 2.25 .92 .91 .91 .045 .042–.049 .056
Model 2 1658*730 2.27 .91 .90 .91 .047 .043–.051 .055
Model 3 1644*726 2.26 .93 .92 .92 .044 .039–.048 .054
Note. df =degree of freedom; *=p<.001; CFI =Comparative Fit Index; TLI =Tucker–Lewis Index; GFI =Goodness-
of-Fit Index; RMSEA =Root Mean Square Error of Approximation; RMSEA 90% CI =RMSEA 90% condence
interval, SRMR =Standardized Root Mean Square Residual.
performance (strong effect size =R2>. 35;
Cohen, 1992). Table 4 presents the standardized
regression coefcient (β), t-statistic, p-value, and
condence interval (95% CI).
As shown in Table 4, individual predictors of
text processing (TP), idea planning (IP), goal-
oriented monitoring and evaluating (GME), feed-
back handling (FH), motivational self-talk (MST),
and emotional control (EC) yielded signicant
predictions for students’ writing prociency. In-
terest enhancement (IE), course memory (CM),
and peer learning (PL) were not identied as
signicant predictors of students’ writing pro-
ciency.
DISCUSSION
The overarching aim of this study was to vali-
date a self-report instrument, the WSSRLQ, for
evaluating the multifaceted structure of writing
strategies used by EFL students for SRL. The
40 items in the questionnaire were measured
using a 7-point scale, ranging from 1 (not at all
true of me) to 7 (very true of me) to probe into
the trait features of self-regulation. The results
of CFAs provided substantial evidence for the
factorial structure of the instrument, entailing
motivational self-talk, interest enhancement,
emotional control, goal-oriented monitoring and
evaluating, idea planning, text processing, course
memory, feedback handling, and peer learning.
In general, not only did the ndings support the
utility of the WSSRLQ as a measure of EFL writing
strategies for SRL with satisfactory psychometric
properties, but they also revealed that the 9 SRL
strategies were reliably distinguished on both
conceptual and empirical grounds.
Model comparisons revealed that the one-
factor second order model (Model 3) had the
best model t over the other two competing mod-
els (Model 1 and Model 2). This means that self-
regulation, as an integrated construct, is sufcient
to account for the correlations of the lower order
writing strategies. Our results lend support to the
sociocognitive view of the SRL process, requiring
the deployment of a range of writing strategies
in relation to cognition, metacognition, social
behavior, and motivational regulation (Zimmer-
man, 2013). In addition, these four dimensions
are not completely distinct but interact with one
another in the SRL process. As Bandura (1986)
explained, human function of self-regulation is a
cyclical, triadic, and reciprocal process, in which
“behavior, cognitive and other personal factors,
and environmental events all operate as interact-
ing determinants of each other” (p. 18). That
might help explain the cross-loading of the 9 SRL
strategies across the four dimensions. Thus, scores
of the 9 subcategories can be calculated separately
to reveal the level of students’ perceived use of
each of the writing strategies. The averaged sums
of the 9 substrategies collectively represent stu-
dents’ overall level of self-regulation capacity.
Multiple Dimensions of EFL Writing Strategies for SRL
In this study, the 9 factors conrmed by CFA
were conceptually interpreted with reference to
the four core paradigms of SRL, namely cogni-
tion, metacognition, social behavior, and motiva-
tional regulation.
The cognitive dimension of writing strategies
for SRL comprised text processing and course
memory. Text processing with six items reected
students’ use of linguistic, rhetorical, and dis-
course knowledge to generate a written text (e.g.,
When writing, I check the structure for logical coher-
ence). Course memory with three items referred
to students’ actively remembering writing knowl-
edge taught in the writing course (e.g., I write use-
ful words and expressions taught in writing courses to
help me remember them).
The metacognitive dimension included idea
planning and goal-oriented monitoring and
evaluating, reecting the triadic aspects of the
metacognitive regulation (Pintrich et al., 2000).
Idea planning with three items referred to spe-
cic idea-generating behavior before writing
Lin Sophie Teng and Lawrence Jun Zhang 19
FIGURE 5
One-Factor Second-Order Factor Model of EFL Writing Strategies for SRL (N=780)
Note. All parameter estimates were statistically signicant at p<.001. GME =Goal-Oriented Monitoring and Eval-
uating; IP =Idea Planning; PL =Peer Learning; FH =Feedback Handling; CM =Course Memory; IE =Interest
Enhancement; EC =Emotional Control; TP =Text Processing; MST =Motivational Self-Talk.
20 The Modern Language Journal 100 (2016)
TABLE 4
Simultaneous Multiple Regression Model of the 9 SRL Strategies on EFL Writing Scores (N=745)
95% CI
SRL Dimensions Predictor βtpLLUL
Cognition CM .12 .78 .67 −1.74 2.65
TP .31 3.64 <.001 −.23 5.27
Metacognition IP .21 2.35 .004 −1.88 2.28
GME .25 2.98 .001 −5.96 .38
Social Behavior PL .11 −.71 .472 −3.10 1.45
FH .17 1.75 .003 −1.43 4.09
Motivational Regulation IE .12 .71 .223 −.92 3.86
MST .24 2.87 <.001 −1.91 5.12
EC −.16 −1.99 .002 −5.93 .16
Note. CM =Course Memory; TP =Text Processing; IP =Idea Planning; GME =Goal-Oriented Monitoring and Eval-
uating; PL =Peer Learning; FH =Feedback Handling; IE =Interest Enhancement; MST =Motivational Self-Talk;
EC =Emotional Control. The dependent variable was writing score. R2=.37, Adjusted R2=.36; CI =condence
interval; LL =lower limit; UL =upper limit.
(e.g., Before writing, I use the Internet to search for
related information to help me plan). Goal-oriented
monitoring and evaluating with six items in-
cluded an arsenal of strategies such as setting up
goals to direct writing activities (e.g., When learn-
ing to write, I set up goals for myself in order to direct my
learning activities) or monitoring and evaluating
the mastery of knowledge and performance in
writing courses (e.g., I monitor my learning process in
writing courses; I evaluate the mastery of the knowledge
or skills learned in writing courses).
Feedback handling and peer learning were in-
terpreted together to reect how learners uti-
lized social–behavioral factors to promote their
self-regulatory capability. Feedback handling with
four items embraced students’ attitude toward
teacher and peer feedback (e.g., I try to improve
my English writing based on teachers’ feedback). Peer
learning with three items involved the procure-
ment of help from peers in the learning environ-
ment, and as such it is also a social interaction
(e.g., I discuss with my peers to have more ideas to
write).
The motivational regulation dimension in-
cluded three types of writing strategies: motiva-
tional self-talk, interest enhancement, and emo-
tional control. Motivational self-talk with eight
items included the self-encouragement in knowl-
edge mastery (e.g., I persuade myself to work hard in
writing courses to improve my writing skills and knowl-
edge) and academic performance (e.g., I remind
myself about how important it is to get good grades in
writing courses). Interest enhancement with four
items reected students’ tendency to make learn-
ing more enjoyable (e.g., I look for ways to bring
more fun to the learning of writing). Emotional con-
trol with three items measured learners’ efforts to
reduce distraction when completing a writing task
or learning to write (e.g., I nd ways to regulate my
mood when I want to give up writing).
Correlations of the 9 EFL Writing Strategies for SRL
Results of the inter-correlation coefcient pro-
vided a clear picture of how the 9 EFL writing
strategies correlated with each other in the SRL
process. The signicant and moderate correla-
tionsindicatethatthe9SRLstrategiesaredis-
tinct but interrelated and interwoven during the
learning-to-write process.
Goal-oriented monitoring and evaluating
strategies were strongly correlated with peer
learning of the social dimension and interest
enhancement of the motivational regulation
dimension. The signicant correlations reect
the essential role of metacognition in the SRL
process (Winne & Hadwin, 2010). This means
that students who have awareness about realiz-
ing and monitoring their task goals might also
exert effort to regulate their social behavior and
intrinsic motivation to maintain or increase their
engagement with tasks.
Also notable are the moderate correlations of
motivational regulation strategies, particularly
motivational self-talk and interest enhancement
with the other dimensions of SRL strategies.
This means that EFL students who are active in
regulating their motivation tend to deploy a rich
repertoire of cognitive, metacognitive, and social
strategies. Our ndings are consistent with some
previous studies, which reported small to medium
correlations of motivational regulation strategies
Lin Sophie Teng and Lawrence Jun Zhang 21
with cognitive and metacognitive strategies in L1
contexts (Cooper & Corpus, 2009; Wolters, 1999).
As Wolters (1999) explained, motivational regu-
lation strategies are helpful to “increase students’
level of cognitive engagement, overall level of
effort, and subsequent achievement within an aca-
demic setting” (p. 285). The positive correlations
between motivational regulation strategies and
social behavior strategies (e.g., peer learning and
feedback handling) lend support to the argument
that “all aspects of self-regulation, including moti-
vational regulation, are developed through social
and cultural interaction/inuences” (Wolters
& Mueller, 2010, p. 633). Our study theoreti-
cally supports an increasingly strong argument
for viewing motivational regulation as a promi-
nent facet for developing self-regulated learners
(Zimmerman, 2008).
Predictive Effect of SRL Strategies on Writing
Prociency
The 9 writing strategies for SRL as a whole pro-
duced a large effect on students’ EFL writing pro-
ciency, supporting the validity of the one high
order model, in which self-regulation as an inte-
grated construct affects students’ learning perfor-
mance. Although only six types of writing strate-
gies for SRL had signicant predictive effects on
writing prociency, our ndings provide some
preliminary support for the claim that L2 writing
achievement is contingent upon the use of dif-
ferent dimensions of SRL strategies, which play
an essential role in mobilizing, directing, and sus-
taining learning efforts, therefore affecting stu-
dents’ academic performance (Manchón et al.,
2007).
Within the cognitive dimension, only text pro-
cessing strategies signicantly predicted writing
scores with a large effect size. This suggests that
learners’ use of linguistic and writing knowledge
is a critical factor affecting their writing perfor-
mance, as revealed in many other studies (e.g.,
Winne, 2011; Zhang et al., 2008). The result also
supports studies that have argued for the essen-
tial role of cognitive processes in fostering active
engagement and enhancing students’ writing out-
comes (e.g., Flower & Hayes, 1981; Ong & Zhang,
2013). Course memory, however, was not a signif-
icant predictor of writing scores. This means that
remembering learning certain materials or course
knowledge, as kinds of surface strategies, did not
have a direct effect on individuals’ writing scores.
Of the metacognitive strategies, both idea plan-
ning and goal-oriented monitoring and evaluat-
ing were signicant predictors of writing scores.
ResultssuggestthatEFLlearnerswhoreliedon
deeper processing strategies were more likely to
perform better in writing tests. As Harris et al.
(2011) explained, metacognitive strategies con-
tributed to learners’ cognitive maturity, which in
turn had a positive effect on their academic per-
formance. Our study, along with the previous nd-
ings, corroborates the arguments for the essential
role of metacognitive strategies in optimizing writ-
ing performance (e.g., Pintrich et al, 2000; Zhang
& Zhang, 2013).
Of the motivational regulation strategies, both
emotional control and motivational self-talk were
signicant predictors of students’ writing pro-
ciency. As discussed earlier, the statements on
emotional control strategies in the questionnaire
were directly related to how students actively reg-
ulated their negative emotions such as anxiety or
worries in task-based environments (e.g., taking a
test). Our ndings revealed that positive emotions
about the learning situation encouraged students
to become more committed to the task, thus con-
tributing to better writing outcomes. This corrob-
orates Boekaerts’s (2011) argument that emotion
control in the service of one’s goal is “a promi-
nent capacity that is key to having success in all
areas of life” (p. 409). In addition, the signi-
cant prediction of motivational self-talk on writing
scores indicates that students who used intrinsic
and extrinsic reasons (performance and mastery
self-talk) to motivate them to learn and/or sus-
tain their learning efforts tended to perform bet-
ter in writing tests. Our nding mirrors some pre-
vious studies conducted in other contexts such as
in the United States (Wolters, 1999) and Germany
(Schwinger, Steinmayr, & Spinath, 2009). For ex-
ample, Wolters (1999) found that performance
self-talk produced a weak signicant prediction
on students’ grade-point average. Schwinger et al.
(2009) reported that mastery self-talk had an
indirect effect on examination performance in
German contexts. Although no consistent results
have been found, these empirical studies together
provide evidence in support of using motiva-
tional self-talk strategies for improving learners’
task performance. However, interest enhance-
ment did not generate any signicant effects on
writing prociency. This might be due to the test-
like environment, in which EFL students did not
have time to use interest enhancement strategies
to facilitate their task performance.
Of the social behavior strategies, peer learn-
ing did not generate any predictive effect on writ-
ing performance. Similarly, Pintrich et al.’s (1991)
study found that peer learning and help-seeking
were not signicantly related to course grades.
22 The Modern Language Journal 100 (2016)
However, feedback handling yielded a signicant,
positive effect on students’ writing scores. As Zim-
merman and Schunk (2011) have asserted, within
the triadic cyclical model of SRL, feedback aids
individuals’ monitoring and self-evaluation in the
SRL process, contributing to positive academic
outcomes.
Taken together, our ndings conrm the direct
effect of some SRL strategies in promoting EFL
students’ writing performance. The results lend
support to some earlier studies on LLSs, for
positive predictive relationships between writ-
ing strategies and language learning outcomes
in general or in specic writing settings (see
Manchón et al., 2007; Plonsky, 2011, for more
information). A less consistent but perhaps more
interesting pattern was the varying predictions
of motivational regulation strategies and social
behavior strategies on students’ writing scores. In
line with Pintrich’s (2004) study, our data lend
preliminary support to the view that individuals’
learning achievement is directly inuenced by
their self-regulation of cognition, motivation, and
behavior, which mediates the relations among the
person, the environment, and the achievement.
The salient role of SRL strategies also reveals
that in promoting active and efcient learning,
students need strategies-based instruction from
a multidimensional perspective when they learn
how to write (see e.g., Zhang et al., 2016). As Zim-
merman and Bandura (1994) argued, “students
needed to be taught skills and strategies for man-
aging not only the cognitive aspects of managing
learning but also methods in which to motivate
themselves for academic pursuits in the face of
difculties or attractive alternatives” (p. 857).
CONCLUSION AND IMPLICATIONS
This article has reported the validation of a
self-report questionnaire, the WSSRLQ, to evalu-
ate the perceived use of writing strategies for SRL
in EFL learning environments. The CFA results
conrmed that the nine EFL writing strategies for
SRL represented reliably distinguishable but cor-
related aspects under an overarching construct of
self-regulation. The moderate correlations of the
nine writing strategies across the four conceptual
dimensions reveal that during the self-regulating
process EFL students’ use of writing strategies
relating to cognition, metacognition, social be-
havior, and motivational regulation is interwoven.
Theoretically, our results render preliminary
evidence for transferring educational psychology
theory to the eld of L2/EFL education, particu-
larly EFL writing. The study lends some support to
the social cognitive view of SRL, which emphasizes
the proactive engagement of SRL processes under
the triadic interplay of individuals, behaviors, and
environments (Zimmerman, 2013). The ndings
collectively reveal how human behavior and the
regulation of cognition and motivation as inter-
acting determinants of each other inuence stu-
dents’ academic performance (Bandura, 1991).
Practically, the newly developed questionnaire
(WSSRLQ) might be applied as a self-evaluation
tool for students to appraise the degree of aware-
ness of SRL writing strategies and cultivate a habit
of using them in developing writing skills in EFL
contexts. Although it does not provide an exhaus-
tive illustration of writing strategies across the four
dimensions, the evaluating process may neverthe-
less give learners a sense of the utility of the nine
elicited SRL strategies from cognitive, metacog-
nitive, social–behavioral, and motivational regula-
tion aspects. Students may further adjust or try to
deploy other SRL strategies to achieve their learn-
ing goals.
In addition, the instrument might be useful in
classrooms as a pedagogical tool for evaluating
students’ preferences for using different dimen-
sions of SRL strategies when teachers are inter-
ested in engaging their students with strategies-
based writing instruction for effective learning in
authentic contexts. As Wolters and Benzon (2013)
posited, “knowing what strategies are preferred or
used most often by students within more authen-
tic academic contexts provides insight into which
onesmightbestbeusedasthetargetofinstruc-
tional interventions” (p. 201).
The signicant predictive power of some writ-
ing strategies suggests that teachers might want to
provide guidance on which strategies should be
taught to the targeted EFL students in classroom
environments. For example, if some motivation
regulation strategies continue to be linked posi-
tively to students’ academic outcomes, then one
basic implication is that these strategies should be
taught directly to students.
LIMITATIONS
Like many studies, our study is not exempt from
limitations. First, the single method of strategy
evaluation through self-report data might fail to
provide rich and accurate information of what
writing strategies learners use in reality, as argued
by some researchers (Cohen & Macaro, 2007; Ox-
ford et al., 2014). For example, participants may
have forgotten some strategies they have used in
the past or they may have reported using some
strategies that they have never used before, or they
Lin Sophie Teng and Lawrence Jun Zhang 23
may have misunderstood the items in the ques-
tionnaire. Therefore, we recommend that multi-
methods for data collection (e.g., stimulated re-
call after completing a task, reection journals,
among others) be used in future studies to offer
more comprehensive evaluations of SRL strate-
gies. However, we should be aware of the inherent
shortcomings of all these self-report measures, as
they can only access part of the writing process, of
which learners are consciously aware.
Second, we recruited only university students in
China. This prevents generalization of our nd-
ings to other populations, such as younger stu-
dents from schools. Further studies are needed
for expanding the sampling methods and partic-
ipant pool. Students of different age groups or
other ethnicities should be included.
Third, although this study investigated stu-
dents’ reported use of SRL strategies from four
dimensions, our instrument solicited only nine
specic strategies, without providing all possible
strategies used in the learning-to-write or com-
posing processes in L2 settings. More studies are
needed to specically tap into other essential as-
pects of SRL (e.g., self-efcacy and environmental
control).
Fourth, this study did not distinguish writing
strategies for using and strategies for learning L2
writing, but instead included both broad cate-
gories for promoting SRL. Based on the predic-
tive effects, it might be interesting to separate the
questionnaire in future research with a focus on
learning to write or using writing for learning con-
tent area knowledge.
Finally, given that our study only found six types
of SRL strategies having direct effects on EFL writ-
ing prociency, it might be interesting to explore
whether other types of writing strategies such
as peer learning and interest enhancement have
indirect effects on EFL writing scores.
REFERENCES
Alexander, P. A., Graham, S., & Harris, K. R. (1998).
A perspective on strategy research: Progress and
prospects. Educational Psychology Review,10, 129–
154.
Anderson, J. R. (1983). The architecture of cognition.Cam-
bridge, MA: Harvard University Press.
Anderson, J. R. (1985). Cognitive psychology and its impli-
cations (2nd ed.). New York: Freeman.
Anderson, N. J. (1991). Individual differences in strat-
egy use in second language reading and testing.
Modern Language Journal,75, 460–472.
Andrade, M. S., & Evans, N. W. (2012). Principles
and practices for response in second language writing:
Developing self-regulated learners.NewYork:Rout-
ledge/Taylor & Francis.
Arbuckle, J. L. (2013). IBM SPSS Amos 22 user’s guide.
Crawfordville, FL: Amos Development Corpora-
tion.
Bandura, A. (1986). Social foundations of thought and ac-
tion: A social cognitive theory. Englewood Cliffs, NJ:
Prentice–Hall.
Bandura, A. (1991). Social cognitive theory of self-
regulation. Organizational Behavior and Human
Decision Processes,50, 248–287.
Berg, E. C. (1999). The effects of trained peer response
on ESL students’ revision types and writing quality.
Journal of Second Language Writing,8, 215–241.
Boekaerts, M. (2011). Emotions, emotion regulation,
and self-regulation of learning. In B. J. Zimmer-
man & D. H. Schunk (Eds.), Handbook of self-
regulation of learning and performance (pp. 408–425).
New York: Routledge/Taylor & Francis.
Byrnes, H. (2014a). Theorizing language development
at the intersection of ‘task’ and L2 writing: Re-
considering complexity. In H. Byrnes & R. M.
Manchón (Eds.), Task-based language learning: In-
sights from and for L2 writing (pp. 79–103). Philadel-
phia/Amsterdam: John Benjamins.
Byrnes, H. (2014b). Linking task and writing for
language development: Evidence from a genre-
based curricular approach. In H. Byrnes & R.
M. Manchón (Eds.), Task-based language learning:
Insights from and for L2 writing (pp. 237–263).
Philadelphia/Amsterdam: John Benjamins.
Chenoweth, N. A., & Hayes, J. R. (2001). Fluency in writ-
ing: Generating text in L1 and L2. Written Commu-
nication,18, 80–98.
Cohen, A. D. (2014). Strategies in learning and using
a second language (2nd ed.). New York: Rout-
ledge/Taylor & Francis.
Cohen, A. D., & Grifths, C. (2015). Revisiting LLS re-
search 40 years later. TESOL Quarterly,49, 414–
429.
Cohen, A. D., & Macaro, E. (Eds.). (2007). Language
learner strategies: Thirty years of research and practice.
Oxford: Oxford University Press.
Cohen, J. (1992). A power primer. Psychological Bulletin,
112, 155–159.
Cooper, C. A., & Corpus, J. H. (2009). Learners’ devel-
oping knowledge of strategies for regulating mo-
tivation. Journal of Applied Developmental Psychology,
30, 525–536.
Cumming, A. (1989). Writing expertise and second-
language prociency. Language Learning,39, 81–
135.
Cumming, A., Busch, M., & Zhou, A. (2002). Inves-
tigating learners’ goals in the context of adult
second-language writing. In S. Ransdell & M. L.
Barbier (Eds.), New directions for research in L2 writ-
ing (pp. 189–208). Dordrecht, the Netherlands:
Kluwer.
24 The Modern Language Journal 100 (2016)
Dinsmore, D. L., Alexander, P. A., & Loughlin, S.
M. (2008). Focusing the conceptual lens on
metacognition, self-regulation, and self-regulated
learning. Educational Psychology Review,20, 391–
409.
Dörnyei, Z. (2005). Language learning strategies and
student self-regulation. In Z. Dörnyei (Ed.), The
psychology of the language learner: Individual dffer-
ences in second language acquisition (pp. 162–196).
Mahwah, NJ: Lawrence Erlbaum.
Dörnyei, Z. (2010). Questionnaires in second language re-
search: Construction, administration, and processing.
New York: Routledge/Taylor & Francis.
Enders, C. K. (2010). Applied missing data analysis.New
York: Guilford.
Ferris, D., & Roberts, B. (2001). Error feedback in
L2 writing classes: How explicit does it need to
be? Journal of Second Language Writing,10, 161–
184.
Field, A. P. (2009). Discovering statistics using SPSS (and
sex, drugs and rock ’n’ roll) (3rd ed.). London:
SAGE.
Flower, L., & Hayes, J. R. (1981). A cognitive process the-
ory of writing. College Composition and Communica-
tion,32, 365–387.
Gao, X. (2007). Has language learning strategy research
come to an end? A response to Tseng et al. Applied
Linguistics,28, 615–620.
Gordon, L. (2008). Writing and good language learn-
ers. In C. Grifths (Ed.), Lessons from good lan-
guage learners (pp. 244–254). Cambridge: Cam-
bridge University Press.
Grifths, C., & Oxford, R. L. (2014). The twenty-rst
century landscape of language learning strate-
gies: Introduction to this special issue. System,43,
1–10.
Harris, K. R., Graham, S., MacArthur, C., Reid, R., &
Mason, L. H. (2011). Self-regulated learning pro-
cesses and children’s writing. In B. J. Zimmerman
&D.H.Schunk(Eds.),Handbook of self-regulation of
learning and performance (pp. 187–202). New York:
Routledge/Taylor & Francis.
Hayes, J. R. (1996). A new framework for understand-
ing cognition and affect in writing. In C. M. Levy
& S. Randsdell (Eds.), The science of writing: Theo-
ries, methods, individual differences, and applications
(pp. 1–27). Mahwah, NJ: Lawrence Erlbaum.
Hsiao, T. Y., & Oxford, R. L. (2002). Comparing theo-
ries of language learning strategies: A conrma-
tory factor analysis. Modern Language Journal,86,
368–383.
Hu, L. T., & Bentler, P. M. (1999). Cutoff criteria for t
indexes in covariance structure analysis: Conven-
tional criteria versus new alternatives. Structural
Equation Modeling,6, 1–55.
Hyland, K. (2007). Genre pedagogy: Language, literacy
and L2 writing instruction. Journal of Second Lan-
guage Writing,16, 148–164.
Hyland, K., & Hyland, F. (2006). Feedback on second
language students’ writing. Language Teaching,39,
83–101.
Kline, R. B. (2011). Principles and practice of structural
equation modeling (3rd ed.). New York: Guilford.
Kormos, J. (2012). The role of individual differences in
L2 writing. Journal of Second Language Writing,21,
390–403.
Lee, I. (2014). Revisiting teacher feedback in EFL writ-
ing from sociocultural perspectives. TESOL Quar-
terly,48, 201–213.
Lei, X. (2008). Exploring a sociocultural approach to
writing strategy research: Mediated actions in writ-
ing activities. Journal of Second Language Writing,17,
217–236.
Leki, I. (1995). Coping strategies of ESL students in writ-
ing tasks across the curriculum. TESOL Quarterly,
29, 235–260.
Macaro, E. (2006). Strategies for language learning and
for language use: Revising the theoretical frame-
work. Modern Language Journal,90, 320–337.
Manchón, R. (2009). Introduction: Broadening the per-
spective of L2 writing scholarship: The contribu-
tion of research on foreign language writing. In R.
Manchón (Ed.), Writing in foreign language contexts:
Learning, teaching, and research (pp. 1–19). Cleve-
don, UK: Multilingual Matters.
Manchón, R., Roca de Larios, J., & Murphy, L. (2007).
A review of writing strategies: Focus on conceptu-
alization and impact of rst language. In A. D. Co-
hen & E. M. Macaro (Eds.), Language learner strate-
gies: Thirty years of research and practice (pp. 229–
250). Oxford: Oxford University Press.
Nguyen, T. C. L., & Gu, Y. (2013). Strategy-based in-
struction: A learner-focused approach to develop-
ing learner autonomy. Language Teaching Research,
17, 9–30.
O’Malley, J. M., & Chamot, A. U. (1990). Learning strate-
gies in second language acquisition. Cambridge: Cam-
bridge University Press.
Ong, J., & Zhang, L. J. (2013). Effects of the manipu-
lation of cognitive processes on EFL writers’ text
quality. TESOL Quarterly,47, 375–398.
Oxford, R. L. (1990). Language learning strategies: What
every teacher should know. Boston, MA: Heinle &
Heinle.
Oxford, R. L. (2013). Teaching and researching lan-
guage learning strategies (2nd ed.). Harlow, UK:
Pearson.
Oxford,R.L.,Rubin,J.,Chamot,A.U.,Schramm,K.,
Lavine, R., Gunning, P., & Nel, C. (2014). The
learning strategy prism: Perspectives of learning
strategy experts. System,43, 30–49.
Petri´
c, B., & Czárl, B. (2003). Validating a writing strat-
egy questionnaire. System,31, 187–215.
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: Aca-
demic Press.
Pintrich, P. R. (2004). A conceptual framework for as-
sessing motivation and self-regulated learning in
college students. Educational Psychology Review,16,
385–407.
Lin Sophie Teng and Lawrence Jun Zhang 25
Pintrich, P. R., & De Groot, E. V. (1990). Motiva-
tional and self-regulated learning components of
classroom academic performance. Journal of Edu-
cational Psychology,82, 33–40.
Pintrich, P. R., Smith, D. A., Garcia, T., & McKeachie,
W. J. (1991). A manual for the use of the Motivated
Strategies for Learning Questionnaire (MSLQ).Ann
Arbor, MI: The University of Michigan Press.
Pintrich, P. R., Wolters, C. A., & Baxter, G. P. (2000). As-
sessing metacognition and self-regulated learning.
In G. Schraw & J. C. Impara (Eds.), Issues in the mea-
surement of metacognition (pp. 43–97). Lincoln, NE:
Buro Institute of Mental Measurements.
Plonsky, L. (2011). The effectiveness of second lan-
guage strategy instruction: A meta-analysis. Lan-
guage Learning,61, 993–1038.
Raimes, A. (1985). What unskilled ESL students do
as they write: A classroom study of composing.
TESOL Quarterly,19, 229–258.
Raykov, T., & Marcoulides, G. A. (2008). An introduc-
tion to applied multivariate analysis.NewYork:Rout-
ledge.
Riazi, A. (1997). Acquiring disciplinary literacy: A social-
cognitive analysis of text production and learning
among Iranian graduate students of education.
Journal of Second Language Writing,6, 105–137.
Roca de Larios, J., Manchón, R., Murphy, L., & Martín,
J. (2008). The foreign language writer’s strategic
behaviour in the allocation of time to writing pro-
cesses. Journal of Second Language Writing,17, 30–
47.
Roca de Larios, J., Martin, J., & Murphy, L. (2001). A
temporal analysis of formulation processes in L1
and L2 writing. Language Learning,51, 497–538.
Rose, H. (2011). Reconceptualizing strategic learning
in the face of self-regulation: Throwing language
learning strategies out with the bathwater. Applied
Linguistics,33, 92–98.
Rubin, J. (1975). What the “good language learner” can
teach us. TESOL Quarterly,9, 41–51.
Sasaki, M. (2000). Toward an empirical model of EFL
writing processes: An exploratory study. Journal of
Second Language Writing,9, 259–291.
Sasaki, M. (2004). A multiple-data analysis of the 3.5-
year development of EFL student writers. Lan-
guage Learning,54, 525–582.
Schunk, D. H., & Ertmer, P. A. (2000). Self-regulation
and academic learning: Self-efcacy enhancing
interventions. In M. Boekaerts, P. R. Pintrich, &
M. Zeidner (Eds.), Handbook of self-regulation
(pp. 631–649). San Diego, CA: Academic
Press.
Schunk, D. H., & Rice, J. M. (1986). Extended attribu-
tional feedback: Sequence effects during remedial
reading instruction. Journal of Early Adolescence,6,
55–66.
Schwinger, M., Steinmayr, R., & Spinath, B. (2009).
How do motivational regulation strategies affect
achievement: Mediated by effort management
and moderated by intelligence. Learning and Indi-
vidual Differences,19, 621–627.
Silva, T. (1993). Toward an understanding of the distinct
nature of L2 writing: The ESL research and its im-
plications. TESOL Quarterly,27, 657–677.
Silva, T., & Matsuda, P. K. (Eds.). (2010). Practicing theory
in second language writing. West Lafayette, IN: Par-
lor Press.
Stern, H. H. (1975). What can we learn from the good
language learner. The Canadian Modern Language
Review,31, 304–318.
Teng, L. S. (2016). Changes in teachers’ beliefs after
a professional development project for teaching
writing: Two Chinese cases. Journal of Education for
Teaching,42, 106–109.
Teng, L. S., & Zhang, J. L. (2016). Fostering strate-
gic learning: The development and validation of
the Writing Strategies for Motivational Regulation
Questionniare (WSMRQ). The Asia-Pacic Educa-
tion Researcher,25, 123–134.
Tseng, W. T., Dörnyei, Z., & Schmitt, N. (2006). A new
approach to assessing strategic learning: The case
of self-regulation in vocabulary acquisition. Ap-
plied Linguistics,27, 78–102.
Varghese, S. A., & Abraham, S. A. (1998). Undergradu-
ates arguing a case. Journal of Second Language Writ-
ing,7, 287–306.
Victori, M. (1999). An analysis of writing knowledge
in EFL composing: A case study of two effec-
tive and two less effective writers. System,27, 537–
555.
Wang, J. (2014). A synthesis of the studies on L2 writ-
ing in the Chinese context. In J. Wang & D. Lin
(Eds.), Teaching and researching EFL writing from
psycho-cognitive and socio-cultural perspectives: Proceed-
ings of the 8th International Conference on Teaching
and Researching EFL Writing in China (pp. 37–69).
Beijing: FLTRP.
Wang, W., & Wen, Q. (2002). L1 use in the L2 compos-
ing process: An exploratory study of 16 Chinese
EFL writers. Journal of Second Language Writing,11,
225–246.
Weinstein, C. E., Schulte, A., & Palmer, D. R. (1987).
LASSI: Learning and Study Strategies Inventory.Clear-
water, FL: Heinle and Heinle.
Wenden, A. (1991). Learner strategies for learner autonomy.
New York: Prentice Hall.
Winne, P. H. (2011). A cognitive and metacognitive anal-
ysis of self-regulated learning. In B. J. Zimmerman
& D. H. Schunk (Eds.), Handbook of self-regulation
of learning and performance (pp. 15–32). New York:
Routledge/Taylor & Francis.
Winne, P. H., & Hadwin, A. F. (2010). Self-regulated
learning and socio-cognitive theory. In P. Pene-
lope, B. Eva, & B. McGaw (Eds.), International en-
cyclopedia of education (pp. 503–508). Oxford, UK:
Elsevier.
Winne, P. H., & Perry, N. E. (2000). Measuring self-
regulated learning. In M. Boekaerts, P. R. Pintrich,
& M. Zeidner (Eds.), Handbook of self-regulation
(pp. 531–566). San Diego, CA: Academic Press.
Wolters, C. A. (1999). The relation between high school
students’ motivational regulation and their use of
26 The Modern Language Journal 100 (2016)
learning strategies, effort, and classroom perfor-
mance. Learning & Individual Differences,11, 281–
299.
Wolters, C. A., & Benzon, M. B. (2013). Assessing and
predicting college students’ use of strategies for
the self-regulation of motivation. Journal of Experi-
mental Education,81, 199–221.
Wolters, C. A., & Mueller, S. A. (2010). Motivation regu-
lation. In P. P. B. McGaw (Ed.), International encyclo-
pedia of education (3rd ed.; pp. 631–635). Oxford,
UK: Elsevier.
Woodrow, L. (2005). The challenge of measuring lan-
guage learning strategies. Foreign Language Annals,
38, 90–98.
Woodrow, L. (2011). College English writing affect: Self-
efcacy and anxiety. System,39, 510–522.
Zhang, L. J. (2008). Constructivist pedagogy in strate-
gic reading instruction: Exploring pathways to
learner development in the English as a second
language (ESL) classroom. Instructional Science,36,
89–116.
Zhang, L. J. (2010). A dynamic metacognitive systems ac-
count of Chinese university students’ knowledge
about EFL reading. TESOL Quarterly,44, 320–353.
Zhang, L. J. (2013). Second language writing as and for
second language learning. Journal of Second Lan-
guage Writing,22, 446–447.
Zhang, L. J. (2016). Reections on pedagogical imports
of Western practices for professionalizing second-
language writing and writing-teacher education.
Writing & Pedagogy,8.
Zhang, L. J., Aryadoust, V., & Zhang, D. (2016). Tak-
ing stock of the effects of strategies-based instruc-
tion on writing in Chinese and English in Sin-
gapore primary schools. In R. E. Silver & W.
Bokhorst–Heng (Eds.), Quadrilingual education in
Singapore: Pedagogical innovation in language educa-
tion (pp. 103–126). New York: Springer.
Zhang, L. J., Gu, Y., & Hu, G. (2008). A cognitive per-
spective on Singaporean primary school pupils’
use of reading strategies in learning to read in En-
glish. British Journal of Educational Psychology,78,
245–271.
Zhang, L. J., & Zhang, D. (2013). Thinking metacog-
nitively about metacognition in second and for-
eign language learning, teaching, and research:
Toward a dynamic metacognitive systems perspec-
tive. Contemporary Foreign Languages Studies,369,
111–121.
Zhao, H. (2010). Investigating learners’ use and under-
standing of peer and teacher feedback on writing:
A comparative study in a Chinese English writing
classroom. Assessing Writing,15, 3–17.
Zimmerman, B. J. (1989). A social cognitive view of self-
regulated academic learning. Journal of Educational
Psychology,81, 329–339.
Zimmerman, B. J. (2008). Investigating self-regulation
and motivation: Historical background, method-
ological developments, and future prospects.
American Educational Research Journal,45, 166–
183.
Zimmerman, B. J. (2011). Motivational sources and
outcomes of self-regulated learning and perfor-
mance. In B. J. Zimmerman & D. H. Schunk
(Eds.), Handbook of self-regulation of learning and per-
formance (pp. 49–64). Mahwah, NJ: Lawrence Erl-
baum.
Zimmerman, B. J. (2013). From cognitive modeling to
self-regulation: A social cognitive career path. Ed-
ucational Psychologists,48, 135–147.
Zimmerman, B. J., & Bandura, A. (1994). Impact of
self-regulatory inuences on writing course attain-
ment. American Educational Research Journal,31,
845–862.
Zimmerman, B. J., & Martinez–Pons, M. (1986). De-
velopment of a structured interview for assess-
ing student use of self-regulated learning strate-
gies. American Educational Research Journal,23, 614–
628.
Zimmerman, B. J., & Risemberg, R. (1997). Becoming
a self-regulated writer: A social cognitive perspec-
tive. Contemporary Educational Psychology,22, 73–
101.
Zimmerman, B. J., & Schunk, D. H. (2011). Self-
regulated learning and performance: An intro-
duction and an overview. In B. J. Zimmerman
& D. H. Schunk (Eds.), Handbook of self-regulation
of learning and performance (pp. 1–12). New York:
Routledge/Taylor & Francis.
APPENDIX A
Guided Interview Questions
1. What strategies do you use in the writing
course to help you learn?
2. What strategies do you use pre, during and
after the writing process?
3. Do you plan before writing? If yes, please
explain the process.
4. Do you revise after writing? If yes, please ex-
plain the process.
5. Can you monitor and evaluate your writ-
ing process and performance in the writing
course?
6. How do you motivate yourself in the com-
posing or learning-to-write process?
7. How do you solve learning challenges in the
writing course?
8. Would you seek help from others in the
learning-to-write process?
Lin Sophie Teng and Lawrence Jun Zhang 27
APPENDIX B
Descriptive Statistics of the Writing Strategies for Self-Regulated Learning Questionnaires (WSSRLQ)
(40 items, N=780)
Items MSDSkewness Kurtosis
Text Processing (TP)
1. When writing, I use some literary devices to make the
composition more interesting.
3.39 1.58 0.29 –0.83
2. When writing, I check grammar mistakes. 5.05 1.59 –0.84 –0.01
3. When writing, I check spelling and punctuation. 4.32 1.65 –0.28 –0.97
4. When writing, I check the structure for logical coherence. 4.25 1.63 –0.28 –0.83
5. When writing, I check the cohesiveness or connection among
sentences.
4.68 1.53 –0.58 –0.45
6. When writing, I check whether the topic and the content
have been clearly expressed.
5.12 1.42 –0.86 0.21
Course Memory (CM)
1. I write useful words and expressions taught in writing courses
to help me remember them.
4.40 1.73 –0.28 –0.98
2. I speak out useful words and expressions taught in writing
courses to help me remember them.
4.70 1.58 –0.54 –0.59
3. I read my class notes and the course material over and over
again to help me remember them.
4.31 1.54 –0.21 –0.85
Idea Planning (IP)
1. Before writing, I read related articles to help me plan. 4.66 1.63 –0.58 –0.61
2. Before writing, I use the internet to search for related
information to help me plan.
4.68 1.65 –0.60 –0.54
3. Before writing, I think about the core elements of a good
composition I have learned to help me plan.
4.50 1.54 –0.39 –0.66
Goal-Oriented Monitoring and Evaluating (GME)
1. When learning to write, I set up goals for myself in order to
direct my learning activities.
3.32 1.58 0.34 –0.67
2. When learning to write, I check my progress to make sure I
achieve my goal.
3.76 1.52 0.01 –0.77
3. I evaluate my mastery of the knowledge and skills learned in
writing courses.
4.01 1.52 –0.06 –0.71
4. I monitor my learning process in writing courses. 3.80 1.53 –0.02 –0.72
5. When writing, I tell myself to follow my plan. 4.16 1.61 –0.28 –0.70
6. When learning to write, I set up a learning goal to improve
my writing.
Peer Learning (PL)
1. I brainstorm with my peers to help me write. 3.13 1.51 0.35 –0.83
2. I discuss with my peers to have more ideas to write with. 3.78 1.62 0.03 –1.03
3. I work with my peers to complete a writing task. 3.79 1.70 0.09 -0.93
Feedback Handling (FH)
1. I am open to peer feedback on my writing. 5.52 1.24 –1.12 1.41
2. I am open to teacher feedback on my writing. 5.64 0.94 –1.23 1.60
3. I try to improve my English writing based on peer feedback. 5.31 1.26 –0.95 1.06
4. I try to improve my English writing based on teacher
feedback.
5.72 1.10 –1.21 2.17
Interest Enhancement (IE)
1. I look for ways to bring more fun to the learning of writing. 4.62 1.54 –0.49 –0.48
2. I choose interesting topics to practice writing. 4.72 1.68 –0.56 –0.60
3. I connect the writing task with my real life to intrigue me. 4.45 1.65 –0.32 –0.75
4. I try to connect the writing task with my personal interest. 4.66 1.61 –0.51 –0.51
28 The Modern Language Journal 100 (2016)
Items MSDSkewness Kurtosis
Motivational Self-Talk (MST)
1. I remind myself about how important it is to get good grades
in writing courses.
4.88 1.59 –0.68 –0.30
2. I tell myself that it is important to practice writing to
outperform my peers.
5.19 1.39 –0.86 0.28
3. I compete with other students and challenge myself to do
better than them in writing courses.
4.94 1.40 –0.67 –0.14
4. I tell myself to practice writing to get good grades. 4.90 1.48 –0.64 –0.26
5. I tell myself that I need to keep studying to improve my
writing competence.
4.78 1.49 –0.60 –0.27
6. I persuade myself to work hard in writing courses to improve
my writing skills and knowledge.
4.97 1.39 –0.68 –0.13
7. I persuade myself to keep on learning in writing courses to
nd out how much I can learn.
4.78 1.37 –0.54 –0.20
8. I tell myself that I should keep on learning in writing courses
to become good at writing.
5.34 1.31 –0.86 0.37
Emotional Control (EC)
1. I tell myself not to worry when taking a writing test or
answering questions in writing courses.
5.15 1.43 –0.85 0.36
2. I tell myself to keep on writing when I want to give it up. 5.04 1.37 –0.71 0.25
3. I nd ways to regulate my mood when I want to give up
writing.
5.33 1.20 –0.82 0.98
