System 122 (2024) 103253

Contents lists available at ScienceDirect

System
journal homepage: www.elsevier.com/locate/system

Profiling L2 students’ writing self-efficacy and its relationship

with their writing strategies for self-regulated learning
Jianhua Zhang a, b, Lawrence Jun Zhang b, *
a
School of Foreign Languages, Sichuan University of Arts and Science, China
b
Faculty of Education and Social Work, University of Auckland, New Zealand

A R T I C L E I N F O A B S T R A C T

Keywords: The current study aimed to examine the relationship among writing self-efficacy, writing stra­
Profiles of writing self-efficacy tegies for self-regulated learning (SRL), and writing achievement in L2 students by adopting latent
SRL writing strategies profile analysis and path analysis. A sample of 391 L2 students from two universities in Western
L2 students
China was recruited to participate in the current study. They were required to respond to the
Latent profile analysis
Genre-Based L2 Writing Self-Efficacy Scale and the Writing Strategies for Self-Regulated Learning
Questionnaire and also write a given-prompt argumentative essay. Three profiles of writing self-
efficacy were identified through latent profile analyses: “Low on All Self-efficacy”, “Average on
All Self-efficacy”, and “High on All Self-efficacy”. Moreover, ANOVA and Welch’s Tests revealed
that those identified profiles were significantly distinct in writing self-efficacy, SRL writing
strategies, and writing achievement. Path analyses also demonstrated the profile differences in
the predictive effects of writing self-efficacy on SRL writing strategies and the predictive effects of
writing self-efficacy and SRL writing strategies on writing achievements. Methodological and
pedagogical implications were discussed.

1. Introduction

The importance of self-efficacy for acquiring writing competence and skills has been well-established (e.g., McCarthy et al., 1985;
Pajares et al., 1999; Pajares & Johnson, 1994; Pajares & Valiante, 1997, 1999). Writing self-efficacy concerns students’ self-confidence
in their capabilities to complete writing tasks. Given the differences between first language (L1) writing and second or foreign language
(hereafter referred to as L2) writing, Teng et al. (2018) put forward a model of writing self-efficacy covering three dimensions: lin­
guistic, self-regulatory, and performance.
Accumulated studies in the field of L2 writing have examined the predictive effect of writing self-efficacy on writing achievement
(e.g., Sun & Wang, 2020; Teng et al., 2018; Woodrow, 2011; Zabihi, 2018). However, it might be difficult to draw a definite conclusion
mainly because of some issues in measuring writing self-efficacy. Therefore, the conception of writing self-efficacy was elaborated to
address the task-specificity issue (Zhang et al., 2023).
Writing self-efficacy has been conceived to be closely related to writing self-regulation (e.g., Zimmerman, 2000; Zimmerman &
Risemberg, 1997). Students employ various techniques or strategies to manage their environmental, behavioural, and personal pro­
cesses to facilitate their self-regulation of writing (Glaser & Brunstein, 2007; Graham & Harris, 2000). SRL writing strategies attracted
scholars to concentrate their attention on factors influencing L2 students’ use of such strategies (e.g., Bai, 2015; Bai et al., 2014, 2020;

* Corresponding author. Faculty of Education and Social Work, University of Auckland, 10 Symonds Street, Auckland 1010, New Zealand.
E-mail addresses: [email protected], [email protected] (J. Zhang), [email protected] (L.J. Zhang).

https://doi.org/10.1016/j.system.2024.103253
Received 13 August 2023; Received in revised form 4 February 2024; Accepted 6 February 2024
Available online 21 February 2024
0346-251X/© 2024 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license
(http://creativecommons.org/licenses/by/4.0/).
J. Zhang and L.J. Zhang System 122 (2024) 103253

Teng & Zhang, 2018). The facilitative effect of SRL writing strategies on L2 students’ writing self-efficacy has been well examined (e.g.,
Bai & Guo, 2018; Guo & Bai, 2019). To our limited knowledge, only one study explicitly investigated the predictive effect of L2
students’ writing self-efficacy on SRL writing strategies (i.e., Teng, 2022). Therefore, it is imperative to conduct more studies to
confirm this predictive effect.
Studies have also demonstrated that L2 students’ writing self-efficacy and utilization of SRL writing strategies exerted significant
predictive impacts on their writing achievement or proficiencies (e.g., Sun & Wang, 2020; Teng & Huang, 2019; Teng & Zhang, 2016;
Zhang et al., 2023). To our limited knowledge, few scholars have explored the possible heterogeneity of the predictive effect of writing
self-efficacy and SRL writing strategies on L2 writing achievements by adopting the variable-centered approach, which often focuses
on the relationships between individual variables isolatedly. Compared with the variable-based approach, the person-centered
approach allows a researcher to investigate multiple variables simultaneously and identify subgroups or profiles of individuals
characterized by similar response patterns across these variables (Howard & Hoffman, 2018). Scholars adopting the person-centered
approach utilized different statistical methods, such as cluster analysis and latent profile analysis (LPA). By employing LPA, Kim et al.
(2015) detected three profiles of self-efficacy for learning English among L2 students: high, medium, and low self-efficacy profiles, thus
offering us insights into the group-level differences in self-efficacy. Therefore, it would be significant to examine whether the same
profiles may be shared by L2 students’ writing self-efficacy in that such profiling might enable us to understand the above-mentioned
potential heterogeneity, thus presenting a more detailed picture of the effects of writing self-efficacy and SRL writing strategies on L2
achievement.
To fill the research lacuna in the literature as the main objective of this study, we aimed to identify the various profiles of writing
self-efficacy as well as the heterogeneity in the relationship among writing self-efficacy, SRL writing strategies, and writing
achievement in the population of L2 students. Specifically, we intended to detect the latent profiles or subgroups of writing self-
efficacy, the subgroup differences in the predictive influence of writing self-efficacy on SRL writing strategies, and the combined
effects of writing self-efficacy and SRL writing strategies on L2 writing achievements.

2. Literature review

2.1. Self-efficacy in L2 writing

Self-efficacy is conceived as a crucial learning incentive that might exert effective predictive impacts on academic achievement, as
evidenced in educational psychology (e.g., Zimmerman, 2000). Specifically, it might determine what learners could plan to achieve,
how much effort and time they could invest and devote, how they might evaluate their goals, effort, and persistence, and how they
might adjust their emotions (e.g., enjoyment, anxiety, and boredom) (Pajares, 2003).
Scholars commonly acknowledged that self-efficacy plays a crucial role in developing writing competence and skills (e.g.,
McCarthy et al., 1985; Pajares et al., 1999; Pajares & Johnson, 1994; Pajares & Valiante, 1997, 1999). Writing self-efficacy relates to
how student writer might perceive their abilities to finish writing assignments (e.g., letters, essays, proposals, thesis, and dissertations)
(Pajares, 2003). Based on a synthetic review of previous studies, Bruning et al. (2013) reconceptualized writing self-efficacy by
attaching particular importance to linguistic and psychological factors involved in the writing process, thus entailing three focal di­
mensions: ideation, writing conventions, and self-regulation. Ideation self-efficacy was concerned with students’ self-confidence in
their abilities to come up with new ideas, which might impact other writing processes due to their cyclic nature. Self-efficacy in writing
conventions was concerned with students’ confidence in translating ideas within the constraints of generally accepted rules in a given
language. Self-efficacy for self-regulation was concerned with how confident students might be in successfully utilizing self-regulatory
strategies to respond to factors involved in the writing processes. Considering the differences between L1 and L2 writing, Teng et al.
(2018) reconceptualized writing self-efficacy from the cognitive-social perspective, entailing linguistic self-efficacy, self-regulatory
self-efficacy, and performance self-efficacy. Based on this conceptualization, they developed the Second Language Writer Self-Efficacy
Scale to gauge L2 students’ writing self-efficacy, which has been criticized for not tackling the task-specificity issue that was inherent
with other scales for writing self-efficacy in L1 and L2 fields (Sun & Wang, 2020). To solve the issue, Zhang et al. (2023) developed the
Genre-based Second Language Writing Self-Efficacy Scale by incorporating genre features of writing tasks, which examined four di­
mensions of writing self-efficacy: linguistic, classroom performance, genre-based performance, and self-regulatory self-efficacy. They
validated this scale against L2 students, demonstrating higher reliability and validity (i.e., construct validity, discriminative validity,
and predictive validity).
Scholars were interested in writing self-efficacy and devoted their efforts to examining its relationship with writing achievements in
the field of L2 writing. Woodrow (2011) initially investigated the relationship between writing self-efficacy and writing performance
in tertiary Chinese-speaking EFL learners. Structural equation modelling showed that writing self-efficacy as a whole exerted a sig­
nificant predictive effect on writing performance. Unfortunately, the drawback of this study might mainly lie in the design of the
measurement of writing self-efficacy, specifically for including items concerning translation and writing tests without any claimed
support, theoretical or other.
Later, Teng et al. (2018) also explored the relationship between writing self-efficacy and writing performance in tertiary
Chinese-speaking EFL learners. Correlational analyses showed that (1) linguistic self-efficacy correlated moderately with L2 writing
performance; (2) self-regulatory self-efficacy correlated weakly but positively with L2 writing performance; and (3) the correlation
strength of performance self-efficacy with L2 writing performance lay in between the former two self-efficacies. In the meantime,
Zabihi (2018) elaborated on the relationship between writing self-efficacy and L2 writing performance among Persian-speaking EFL
undergraduate students by focusing on complexity, accuracy and fluency. They reported that writing self-efficacy had moderate to

2
J. Zhang and L.J. Zhang System 122 (2024) 103253

strong correlations with three dimensions of L2 writing performance. Path analysis showed that (1) writing self-efficacy significantly
predicted complexity, accuracy, and fluency of L2 writing performance, and (2) compared with working memory capacity, writing
self-efficacy demonstrated different effects on dimensions of L2 writing performance. However, these findings might be mainly
undermined by adopting the Self-Efficacy for Writing Scale in the L2 writing context without any modification because the scale was
originally designed to assess L1 writers’ writing self-efficacy.
Recently, Sun and Wang (2020) probed the relationship between writing self-efficacy and L2 writing performance in
Chinese-speaking EFL learners. The regression analysis revealed that writing self-efficacy could significantly predict L2 writing per­
formance after controlling for socioeconomic status. However, the finding might be jeopardized by the measurement of L2 writing
performance because the composite scores of both writing and translation tasks in the College English Test (CET), a national-wide
English test in China, could be considered inappropriate for evaluating writing performance.
Therefore, based on these accumulated studies reviewed above, we can see that scholars might not draw an irrefutable conclusion
regarding the predictive effects of writing self-efficacy on writing achievement among L2 students. Thus, it is imperative to conduct
more studies to delineate the predictive effects. Moreover, scholars in previous studies mainly adopted the variable-based approach.
Although the variable-based approach could help us understand the relationship between writing self-efficacy and writing achieve­
ment, the person-based approach might offer us a more nuanced grasp of that relationship by utilizing cluster analysis or LPA. Kim
et al. (2015) initially examined the patterns of self-efficacy for learning English among Korean-speaking EFL learners by adopting LPA.
Through LPAs, they identified three groups of learners with high, medium, and low self-efficacy profiles. Later, Chen et al. (2022)
examined profiles of L2 student writers based on their SRL writing strategies and writing self-efficacy using LPA and found three
profiles: Efficacious self-regulators, moderate strategists, and unmotivated learners. However, their use of the z-standardized mean of
the subscales for gauging SRL writing strategies and writing self-efficacy might obscure essential nuances and compromise the results
concerning the profiles of L2 student writers. Thus, it is significant to investigate whether the same pattern of self-efficacy found by
Kim et al. (2015) could exist among L2 student writers and, furthermore, whether L2 students in different profiles of writing
self-efficacy might share the same predictive effect of writing self-efficacy on writing achievement.

2.2. SRL writing strategies in L2 writing

SRL is essential to learners’ perception of the orchestration of the cognitive, motivational, and emotional aspects of learning. It is
“an active, constructive process whereby learners set goals for their learning and then attempt to monitor, regulate, and control their
cognition, motivation, and behaviour, guided and constrained by their goals and the contextual features in the environment” (Pintrich,
2000, p. 453). Effective SRL processes require learners’ active deployment of a range of strategies to help them intentionally activate,
sustain, and adjust cognition, affect, and behaviour to achieve their learning goals (Zimmerman & Schunk, 2011).
Self-regulation could enable writers to manipulate the syntactic and semantic dimensions of writing (Bruning et al., 2013; Zim­
merman & Bandura, 1994; Zimmerman & Kitsantas, 1999, 2002) and to improve their utilization of writing strategies (Brunstein &
Glaser, 2011; Graham & Harris, 2000). Scholars identified a great number of SRL strategies that writers utilized to manipulate
environmental, behavioural, and personal processes identified in the self-regulation of writing (Glaser & Brunstein, 2007; Graham &
Harris, 2000; Zimmerman & Risemberg, 1997).
Research on the role of SRL writing strategies in the context of L2 writing mainly focused on factors influencing L2 students’
utilization of SRL writing strategies, for instance, proficiency (Bai et al., 2014), instruction (Bai, 2015), motivational regulation
strategies (Teng & Zhang, 2018), gender and grade level (Bai et al., 2020), and writing self-efficacy, growth mindset, and task value
(Bai et al., 2021). For instance, Bai et al. (2014) explored the relationship between SRL writing strategies and English proficiency in
Singapore primary pupils. Analysis of variance (ANOVA) results revealed that students with different proficiency levels showed
systematic differences in their use of SRL strategies. It was found that such strategies as planning, text-generating, monitoring and
evaluating, revising, and resourcing contributed to participants’ English proficiency and writing development. Later, Bai (2015)
revealed that strategy-based writing instruction could enhance pupils’ writing competence and SRL strategy use.
Moreover, Teng and Zhang (2018) further explored the relationship between SRL writing strategies and English writing proficiency
by delineating the role of motivation regulation strategies among English majors in China. They asked the participants to write a
given-topic argumentative essay, which was evaluated using the ESL Composition Profile. Mediation models revealed (1) motivation
regulation strategies had a weak direct effect on writing performance; (2) motivation regulation strategies had an indirect effect on
writing performance through cognitive and metacognitive strategies; and (3) motivation regulation strategies had significant, direct
effects on cognitive, metacognitive, and social behavioural strategies. These findings implied that motivation regulation acted as a
precursor and mediator of SRL writing strategies and had close interactions with cognitive, behavioural, and contextual factors
involved in the SRL process. Recently, Bai et al. (2020) investigated the effect of gender, writing proficiency, and grade level on SRL
writing strategies in upper graders in Hong Kong primary schools. The results of a three-way multivariate analysis of variance revealed
that gender, writing proficiency, and grade level had impacts on students’ use of SRL writing strategies. Meanwhile, Bai et al. (2021)
examined the relationships between motivational variables, SRL writing strategies, and writing competence in Hong Kong’s primary
pupils. Structural equation modelling revealed that writing self-efficacy and growth mindset had a modest predictive effect on stu­
dents’ use of SRL writing strategies, while task values showed complex predictive effects.
Only a limited number of studies focused on the predictive effect of SRL writing strategies on writing performance among L2
students. Teng and Zhang (2016) investigated the relationship between SRL writing strategies and writing proficiency in tertiary
Chinese EFL learners. Multiple regression analyses revealed that text processing, idea planning, goal-oriented monitoring and eval­
uating, feedback handling, motivational self-talk and emotional control were significant predictors of EFL learners’ English writing

3
J. Zhang and L.J. Zhang System 122 (2024) 103253

proficiency. Recently, Teng and Huang (2019) investigated the predictive effect of SRL writing strategies on English writing profi­
ciency in secondary EFL learners. Results of multiple regression analyses revealed that (1) nine SRL writing strategies as a whole acted
as a strong predictor of secondary EFL learners’ writing proficiency, and (2) goal-oriented monitoring had the most significant pre­
dictive effect on writing proficiency, which was followed by motivational self-talk, text processing, idea planning, interest enhance­
ment, and emotional control in descending order. Meanwhile, Sun and Wang (2020) also investigated the predictive effect of the SRL
writing strategy on writing proficiency in tertiary Chinese EFL learners. Hierarchical linear regression analysis revealed that SRL
writing strategies had a predictive effect on EFL learners’ writing proficiency. However, to our limited knowledge, few scholars have
devoted their attention to the possible group-level heterogeneity of the predictive effect of SRL writing strategies on L2 writing
achievements.

2.3. The relationships between writing self-efficacy and SRL writing strategies

Zimmerman (2000) argued that motivation might play a crucial role in determining how students engage in self-regulated learning.
As mentioned in the above section, writing self-efficacy could act as an essential motivational factor. Therefore, writing self-regulation
was postulated to be closely linked with writing self-efficacy (Zimmerman & Risemberg, 1997).
Researchers have concentrated on how SRL writing techniques helped L2 students develop their writing self-efficacy (e.g., Bai &
Guo, 2018; Guo & Bai, 2019). Bai and Guo (2018) investigated the effect of SRL writing strategies on writing self-efficacy in EFL
learners in Hong Kong primary schools. Multiple ANOVAs revealed significant differences in writing self-efficacy among students with
high, medium, and low use of SRL writing strategies. Structural equation modelling demonstrated that SRL writing strategies such as
planning and self-monitoring exerted a more significant impact on EFL learners’ writing self-efficacy than other strategies. Further­
more, Guo and Bai (2019) extended the effect of SRL writing strategies on writing self-efficacy in EFL learners in Hong Kong primary
schools by adding another variable of intrinsic motivation. T-tests showed that high achievers reported more frequent use of SRL
writing strategies than low achievers. In contrast, zero-order correlations indicated that SRL writing strategies positively correlated
with intrinsic motivation and writing self-efficacy in both cohorts. Structural equation modelling revealed that SRL writing strategy
use had different effects on high and low achievers’ writing self-efficacy and intrinsic motivation. For high achievers, the use of
planning strategy had an impact on writing self-efficacy and intrinsic motivation, while the use of self-monitoring strategies exerted an
impact only on intrinsic motivation. In contrast, for lower achievers, the use of self-motivating strategies and revising strategies
impacted writing self-efficacy and intrinsic motivation. The finding suggested students with different writing achievements adopted
different SRL writing strategies to enhance their motivation in writing.
However, the predictive effect of writing self-efficacy on the employment of SRL strategies among L2 students has seldom been
explored. To our limited knowledge, only two studies explicitly examined such a predictive effect (e.g., Bai et al., 2022; Teng, 2022).
Bai et al. (2022) found a positive correlation between self-efficacy and SRL strategy use among EFL students from Hong Kong primary
schools. By employing multiple regression analyses, Teng (2022) revealed that self-efficacy showed significant predictive effects on all
kinds of self-regulated learning strategies (i.e., cognition, metacognition, and motivational regulation) except for strategies for social
behaviour in the population of tertiary L2 student writers. Therefore, the result concerning the predictive impacts of writing
self-efficacy on SRL writing strategies might be tentative, and therefore, it is necessary to conduct more studies to verify these effects.
Several issues in the literature concerning the relationship between writing self-efficacy, SRL writing strategy, and writing
achievement have not been addressed appropriately. Firstly, the variable-centered approach adopted in the available studies assumes
homogeneity across the entire sample; in other words, the participants were characterized by a single normal population distribution
(Howard & Hoffman, 2018). This assumption might cover up the fact that the participants come from a population with different
distributions. Secondly, the variable-centered approach focuses on the linear relationship between the above-mentioned variables
across the entire sample and may overlook subgroup differences, which would not be valid or applicable in real situations, as
demonstrated by Guo and Bai (2019). Thirdly, the results concerning the profiles of L2 student writers found in Chen et al. (2022)
might be tentative and a little obscure, offering limited information about subgroups of L2 learners in terms of writing self-efficacy.
Therefore, to address the aforementioned issues, the current study was designed to examine the relationship among writing self-
efficacy, SRL writing strategies, and writing achievement in L2 students by employing LPA. Different from the variable-centered
approach, LPA recognizes and accommodates heterogeneity by identifying latent subgroups within the population, allows for the
simultaneous examination of multiple variables, and identifies subgroups of individuals characterized by similar response patterns
across these variables (i.e., Berlin et al., 2014; Peugh & Fan, 2013; Spurk et al., 2020). Accordingly, LPA acknowledges that different
subgroups may exist with distinct response patterns and may facilitate us to uncover subgroup-specific relationships, capturing the
complexity of relationships among variables and providing a more nuanced understanding of these relationships. As a result, LPA could
facilitate us to examine the latent subgroups of L2 students in writing self-efficacy and then explore the nuanced subgroup differences
in the predictive effects of writing self-efficacy on SRL writing strategies and the predictive effects of writing self-efficacy and SRL
writing strategies on writing achievement, thus offering an elaborated understanding of these relationships and shedding insights on
SRL-based instructional practices. Specifically, the current study intended to answer the following research questions:

RQ1 What are the profiles of L2 students’ writing self-efficacy?

RQ2 Are there differences in SRL writing strategies and L2 writing achievement across those profiles?
RQ3 Are there differences in the predictive effects of writing self-efficacy on the use of SRL writing strategies across those profiles?
RQ4 Are there differences in the predictive effects of writing self-efficacy and SRL writing strategies on L2 writing achievement
across those profiles?

4
J. Zhang and L.J. Zhang System 122 (2024) 103253

3. Methods

3.1. Participants

After agreeing to participate and signing formal consent, a total of 396 EFL students from two medium-ranking universities in
Western China were voluntarily participated in the current study. They had been studying English as a foreign language for at least six
years while also becoming fluent in Mandarin Chinese. They enrolled in a two-year integrated English course during their college,
which concentrated on teaching five language skills (i.e., listening, speaking, reading, writing, and translating) in an integrated
manner. They did not attend any courses specialized in English. Among the participants, 250 were females, accounting for 63.13 per
cent, and 146 were males, accounting for 36.87 per cent. Their average age was 20.4, and they varied from 18 to 22. They were
involved in different grades, of which 47.25 per cent (n = 206) were freshmen, 25.46 per cent (n = 111) were sophomores, and 27.29
per cent (n = 119) were juniors. Their English proficiency ranged from pre-intermediate to intermediate.

3.2. Instruments

3.2.1. The genre-based L2 Writing Self-Efficacy Scale

Zhang et al.’s (2023) Genre-Based L2 Writing Self-Efficacy Scale was directly adopted to measure the participants’ writing
self-efficacy. It was specifically designed to measure L2 learners’ self-perceived confidence for producing written essays where the
participants could score their confidence on a 5-point Likert scale ranging from 1 (does not describe me) to 5 (describes me perfectly). As
reported in Zhang et al. (2023), this scale demonstrated sound psychometric characteristics when used with Chinese-speaking EFL
tertiary students.
Four facets of writing self-efficacy were examined in the current study: linguistic self-efficacy, self-regulatory self-efficacy, class­
room performance self-efficacy, and genre-based performance self-efficacy. When the scale items were presented to the participants,
they were all translated into Chinese, ensuring that they could comprehend each item entirely and preventing any potential mis­
understandings. Similarly, translation and back translation were used to confirm and support the accuracy and equivalence of the
translation. The Cronbach’s alpha coefficient, which indicated the scale’s strong internal consistency with a value of 0.922, was also
used to assess the reliability of this instrument. Confirmatory factor analysis demonstrated the scale fitted well with the participants
involved in this study (x2 = 240.913; df = 96; p < .001; x2/df = 2.51; CFI = 0.941; TLI = 0.926; RMSEA = 0.062 [0.052, 0.072]; SRMR
= 0.056).

3.2.2. The Writing Strategies for Self-Regulated Learning Questionnaire

The Writing Strategies for Self-Regulated Learning Questionnaire designed by Teng and Zhang (2016) was utilized to gauge the
participants’ employment of SRL writing strategies, including text revising, peer learning, feedback handling, interest enhancement,
and motivational self-talk. We modified the original questionnaire by deleting the items with factor loadings lower than 0.5, with 23
items (items 1, 7, 8, 9, 10, 12, 11, 13, 15,16, 17, 18, 19, 23, 26, 30, 31, 32, 33, 38, 39, 40 on the initial list) being removed. The
modified questionnaire focused on five specific SRL writing strategies: Text revising, peer learning, feedback handling, interest
enhancement, and motivational self-talk. With the help of the revised questionnaire, participants could report their use of SRL writing
strategies on a 7-point Likert scale ranging from 1 (not at all true of me) to 7 (very much true of me). The reliability of the revised
questionnaire was also assessed by Cronbach’s alpha coefficient, whose value was 0.895, suggesting the sound internal consistency of
this scale. The result of confirmatory factor analysis showed that the revised questionnaire fitted well with the participants involved in
this study (x2 = 248.297; df = 109; p < .001; x2/df = 2.278; CFI = 0.928; TLI = 0.910; RMSEA = 0.057 [0.048, 0.067]; SRMR = 0.053).

3.2.3. English writing test

Participants’ writing achievements were evaluated by a writing test. The participants were required to complete a given-prompt
argumentative writing task with at least 150 words. They were also asked to complete the writing task online within 40 min due to
the constraints of the COVID-19 pandemic. The writing task was selected partially because the participants were more familiar with
argumentative writing than other writing tasks and partially because argumentative writing was frequently tested in CET and thus was
sufficiently practised among the participants. Therefore, one topic regarding the use of the Internet in education from the item pool of
CET 4 was selected (see Appendix A for details).
Jacobs et al.’s (1981) ESL Composition Profile, one of the extensively used analytical scoring rubrics in the field of L2 writing, was
employed to gauge participants’ writing achievement (e.g., Chen & Zhang, 2019; Huang & Zhang, 2020; Teng, 2022; Xu et al., 2022,
2023a, 2023b). This scoring rubric covers five dimensions of writing performance: content, organization, language, vocabulary, and
mechanics. Compared with the 0–15 holistic scoring scale employed for the CET writing, this 0-100 rubric assigns different weights to
the aforementioned dimensions: 0.3 for content, 0.2 for organization, 0.25 for language, 0.2 for vocabulary, and 0.05 for mechanics
and would provide more details and distinctions about the participants’ writing achievement.

3.3. Procedure

The current study incorporated two sessions, which were implemented in a day. In session one, the Genre-Based L2 Writing Self-
Efficacy Scale and the Writing Strategies for Self-Regulated Learning Questionnaire were given to the participants to elicit their self-
perception of writing self-efficacy and the use of SRL writing strategies. After signing the online consent form, the participants

5
J. Zhang and L.J. Zhang System 122 (2024) 103253

provided their responses to these scales online, specifically through an online survey website named Wenjuanxing in China, which
shares common features with Qualtrics. Before responding, they were told that their responses might neither be scored nor exert any
impact on their academic performance. If they could respond truthfully according to their actual situation, their responses might be
highly valued. We took note of and responded to both of their queries and comments concerning but not limited to responding to these
scales throughout the whole session. They completed the items of these scales within 4–7 min.
In session two, the participants were given 40 min to complete the assigned argumentative writing task, and then their written
essays were assembled to assess their EFL writing achievement. Two experienced CET4 writing raters were invited to score the
gathered essays using Jacobs et al.’s grading criteria mentioned above. The raters’ inter-rater reliability was rAB = 0.939, indicating
reliable scoring.

3.4. Data analysis

Before any analysis was done on the assembled data, the data screening and cleaning recommended by Dörnyei and Taguchi (2010)
were conducted. Five responses were removed from the data because they showed a lack of effort, intentional wrongdoing, or inac­
curate answers. Meanwhile, the missing data were also inspected and examined using SPSS 25. Therefore, the data concerning five
participants were eliminated from the current study, leaving the finalist list of 391 participants for the analysis. The multivariate
normality of the involved variables was examined by Mardia’s skewness and kurtosis test, the results of which indicated that the
variables were multivariate non-normally distributed (Mardia’s skewness = 239.477 (χ2 = 15732.526) > 1.96; Mardia’s kurtosis =
1560.288 (χ2 = 4515.724) > 1.96)
LPAs were employed to explore L2 students’ writing self-efficacy profiles with Mplus 8.3 (Muthen & Muthen, 2017), where we used
the maximum likelihood estimation with robust standard errors to estimate the parameters of these latent profile solutions.
Furthermore, whether the constructed latent profiles fitted well with the collected data would be determined based on the estimates
proposed by Jung and Wickrama (2008): the Akaike information criterion (AIC), the Bayes Information criterion (BIC), the adjusted
Bayes Information criterion (aBIC), and entropy. Besides, the Lo–Mendell–Rubin likelihood ratio test (LMR-LRT) and the bootstrapped
likelihood ratio test (BLRT) were employed to detect whether there are significant differences between k and k− 1 profiles.
Moreover, Levene’s Test for Equality of Variance were employed to examine the equality of variance of the collected data before
conducting the statistical analyses on the differences across the profiles of writing self-efficacy, the results of which might determine
the methods to detect these differences, such as ANOVA or Welch’ Test and Least Significant Difference (LSD) analyses or Tamhane’s
T2 Tests. ANOVA or Welch’s Tests were employed to inspect the differences across the best-fitted profiles of writing self-efficacy in
writing self-efficacy and SRL writing strategies, and post hoc LSD analyses or Tamhane’s T2 Tests were used to detect the differences
between each profile pair. Furthermore, the same statistical methods were also utilized to examine the difference across these profiles
in L2 students’ writing achievements. In addition, path analyses were employed to scrutinize the differences across these profiles in the
following predictive effects: the predictive effects of writing self-efficacy on SRL writing strategies and the predictive effects of writing
self-efficacy and SRL writing strategies on writing achievement. The models constructed for the path analyses were assessed according
to the goodness-of-fit indices suggested by Kline (2016): the χ2 test statistic with its level of significance, the comparative fit index
(CFI), the Tucker-Lewis index (TLI), the root mean square error of approximation (RMSEA), and the standardized root mean square
residual (SRMR).

4. Results

4.1. Profiles of L2 students’ writing self-efficacy

The profiles of L2 students’ writing self-efficacy were determined using the exploratory LPA. Three solutions, specifically the two-
profile, three-profile, and four-profile, were examined. The latent profile that fits the data the best was selected according to all of the
aforementioned estimations, specifically, higher entropy, lower AIC, BIC, and aBIC, and the significant p values for LMR and BLRT. The
identified profiles could also be checked to see if they differ conceptually from one another in terms of the indicator variables. The LPA
results concerning three solutions are presented in Table 1.
As Table 1 shows, compared with the four-profile solution with an insignificant p-value for LMR (pprofile-4 = 0.2648), the p-values
for the two-profile and three-profile solutions were significant (pprofile-2 = 0.0004, pprofile-3 = 0.0050). These findings might point to
statistical discrepancies between the comparisons of one profile versus two profiles and two profiles versus three profiles rather than
three profiles versus four profiles. Meanwhile, in contrast to the entropies of the two-profile and four-profile solutions, the three-profile
solution showed a greater entropy value (entropyprofile-3 = 0.919 > entropyprofile-2 = 0.891> entropyprofile-4 = 0.813), suggesting that
the three-profile solution might be the best-fitted one for the profiles of L2 students’ writing self-efficacy. The subtlety of the three

Table 1
The LPA results.
C K AIC BIC aBIC Entropy LMR BLRT

2 49 19461.677 19656.144 19500.660 .891 .0004 .0000

3 66 18785.451 19047.386 18837.971 .919 .0050 .0000
4 83 18613.804 18943.207 18678.852 .886 .2648 .0000

6
J. Zhang and L.J. Zhang System 122 (2024) 103253

profiles is displayed in Fig. 1.

According to Fig. 1, participants assigned to Profile 1 were typified by low scores on all the indicator items concerning linguistic
self-efficacy, self-regulatory self-efficacy, classroom performance self-efficacy, and genre-based performance self-efficacy; those
assigned to Profile 2 by average scores on all the indicator items; those assigned to Profile 3 by high scores on all the items. Overall, the
participants in the three profiles scored relatively lower on classroom performance self-efficacy than on linguistic self-efficacy, self-
regulatory self-efficacy, and genre-based performance self-efficacy. We labelled Profiles 1 to 3 as “Low on All Self-efficacy”, “Average
on All Self-Efficacy”, And “High on All Self-Efficacy”, respectively (see Fig. 2).
ANOVA (or Welch’s Tests) and post hoc LSD analyses (or Tamhane’s T2 Tests) were utilized to detect the differences across three
identified profiles in four dimensions of writing self-efficacy. The results of ANOVA (or Welch’s Tests) in Table 2 reveal that three
profiles were significantly distinctive in linguistic self-efficacy (F = 130.166, p = .000, η2 = 0.461), self-regulatory self-efficacy (F =
123.688, p = .000, η2 = 0.423), classroom performance self-efficacy (F = 246.889, p = .000, η2 = 0.56), and genre-based performance
self-efficacy (F = 314.519, p = .000, η2 = 0.67). According to the results of LSD analyses (or Tamhane’s T2 Tests) in Table 3, all profile
pairs were significantly different in linguistic self-efficacy (pprofile1-2 = .018, pprofile1-3 = .003), self-regulatory self-efficacy (pprofile1-2 =
.000, pprofile1-3 = .000, pprofile2-3 = .000), class performance self-efficacy (pprofile1-2 = .000, pprofile1-3 = .000, pprofile2-3 = .000), and genre-
based performance self-efficacy (pprofile1-2 = .000, pprofile1-3 = .000, pprofile2-3 = .000) except the following single pair: the Profile 2 versus
Profile 3 pair in linguistic self-efficacy (p = .242).
In addition, the participants in our samples across three profiles were distributed as follows: sixty-seven participants were allocated
to Profile 1, accounting for 17.135%; two hundred twenty-five to Profile 2, accounting for 57.545%; and ninety-nine students to Profile
3, accounting for 25.320%.

4.2. Profile differences in SRL writing strategies and writing achievement

In the same vein, ANOVA (or Welch’s Tests) and LSD analyses (or Tamhane’s T2 Tests) were conducted to discover L2 students’
differences in SRL writing strategies and writing achievement across the identified latent profiles of writing self-efficacy, the results of
which are shown in Tables 4 and 5.
According to the results of ANOVA (or Welch’s Tests) in Table 4, participants assigned to three profiles were significantly different
in the following SRL writing strategies: text revising (F = 31.654, p = .000, η2 = 0.14), peer learning (F = 24.758, p = .000, η2 = 0.113),
feedback handling (F = 6.088, p = .003, η2 = 0.028), interest enhancement (F = 23.35, p = .000, η2 = 0.107), and motivational self-talk
(F = 33.587, p = .000, η2 = 0.135). Statistically significant differences across the three profiles were also found in L2 students’ writing
achievement (F = 4.536, p = .011, η2 = 0.023).
Furthermore, post hoc LSD analyses (or Tamhane’s T2 Tests) were conducted on SRL writing strategies and writing achievement in
order to uncover the subtle differences between the profile pairs, the results of which are presented in Table 5.
As the results of LSD analyses (or Tamhane’s T2 Tests) in Table 5 reveal, all profile pairs were significantly distinctive in the
following SRL writing strategies: text revising (pprofile1-2 = .000, pprofile1-3 = .000, pprofile2-3 = .000), peer learning (pprofile1-2 = .000,
pprofile1-3 = .000, pprofile2-3 = .000), feedback handling (pprofile1-2 = .021, pprofile1-3 = .000, pprofile2-3 = .000), interest enhancement (pprofile1-
3 = .003, pprofile2-3 = .003) and motivational self-talk (pprofile1-2 = .034, pprofile1-3 = .000, pprofile2-3 = .000) except the following single
pair: the Profile 1 versus Profile 2 pair in interest enhancement (p = .430). Moreover, the statistically significant differences in writing
achievement were found in the Profile 1 versus Profile 2 pair (pprofile1-2 = .018) and the Profile 1 versus Profile 3 pair (pprofile1-3 = 0.003)

Fig. 1. Three profiles of writing self-efficacy.

Note: LS = linguistic self-efficacy, SRS = self-regulatory self-efficacy, CPS = classroom performance self-efficacy, and GPS = genre-based perfor­
mance self-efficacy.

7
J. Zhang and L.J. Zhang System 122 (2024) 103253

Fig. 2. Path models of writing self-efficacy on SRL writing strategies.

8
J. Zhang and L.J. Zhang System 122 (2024) 103253

Table 2
Results of differences in writing self-efficacy across latent profiles.
Factors Mean Levene (df) F Sig. η2
Profile1 Profile2 Profile3

LS 2.949 4.183 5.226 7.703 (2, 388) Welch’s F (2142.796) = 130.166 p < .001 .461
SRS 3.453 4.317 5.394 3.504 (2, 388) Welch’s F (2148.240) = 123.688 p < .001 .423
CPS 2.168 3.287 4.871 1.144 (2, 388) ANOVA F (2,388) = 246.889 p < .001 .56
GPS 2.605 4.080 5.308 8.515 (2, 388) Welch’s F (2143.840) = 314.519 p < .001 .67

Note. LS = linguistic self-efficacy, SRS = self-regulatory self-efficacy, CPS = classroom performance self-efficacy, GPS = genre-based self-efficacy.

Table 3
Results of differences in writing self-efficacy across latent profiles.
Factors Profile pair Statistics

MD SE Sig.

LS 1–2 − 2.267 .958 .018

1–3 − 3.240 1.089 .003
2–3 − .972 .830 .242
SRS 1–2 − 1.234 .110 .000
1–3 − 2.277 .125 .000
2–3 − 1.043 .096 .000
CPS 1–2 − .864 .104 .000
1–3 − 1.941 .118 .000
2–3 − 1.077 .090 .000
GPS 1–2 − 1.119 .111 .000
1–3 − 2.703 .126 .000
2–3 − 1.585 .096 .000

Note. LS = linguistic self-efficacy, SRS = self-regulatory self-efficacy, CPS = classroom performance self-efficacy, GPS = genre-based self-efficacy.

Table 4
Results of differences in SRL writing strategies and writing achievement across latent profiles.
Factors Mean Levene (df) F Sig. η2
Profile1 Profile2 Profile3

TR 4.202 4.807 5.429 1.814 (2, 388) ANOVA F (2,388) = 31.654 p < .001 .14
PL 3.846 4.161 4.845 1.173 (2, 388) ANOVA F (2,388) = 24.758 p < .001 .113
FH 5.169 5.277 5.626 2.109 (2, 388) Welch’s F (2155.264) = 6.088 p < .005 .028
IE 4.418 4.736 5.475 1.656 (2, 388) ANOVA F (2,388) = 23.35 p < .001 .107
MST 4.549 4.998 5.631 3.248 (2, 388) Welch’s F (2157.733) = 33.587 p < .001 .135
WA 73.993 76.260 77.232 .466 (2, 388) ANOVA F (2,388) = 4.536 p < .05 .023

Note. TR = text revising, PL = peer learning, FH = feedback handling, IE = interest enhancement, MST = motivational self-talk, WA = writing
achievement.

rather than the Profile 2 versus Profile 3 pair (pprofile2-3 = .242).

4.3. Profile differences in the predictive effects of writing self-efficacy on SRL writing strategies

Path analyses were utilized to explore the predictive effects of L2 students’ writing self-efficacy on their use of SRL writing stra­
tegies. Table 6 provides the details about the model fit indices of the models for path analyses, showing that these models are fully
saturated. The results of the path analyses are displayed in Table 7.
According to the results of path analyses in Table 7, in terms of text revising, linguistic self-efficacy and self-regulatory self-efficacy
demonstrated significant predictive effects for all the participants (βLS = 0.247, p = .000; βSRS = 0.244, p = .000); no facets of writing
self-efficacy exerted significant predictive effects for those in Profile 1; linguistic self-efficacy and self-regulatory self-efficacy also
showed significant predictive effects for those in Profile 2 (βLS = 0.284, p = .001; βSRS = 0.185, p = .015); self-regulatory self-efficacy
also showed significant predictive effects for those in Profile 3 (β = 0.28, p = .01). In terms of peer learning, self-regulatory self-efficacy
and classroom performance self-efficacy demonstrated significant predictive effects for all the participants (βSRS = 0.263, p = .000;
βCPS = .21, p = .002); self-regulatory self-efficacy also showed a significant predictive effect for those in Profile 1 (β = 0.34, p = .001);
self-regulatory self-efficacy and classroom performance self-efficacy also showed significant predictive effects for those in Profile 2
(βSRS = 0.196, p = .01; βCPS = .174, p = .01); no facets of writing self-efficacy exerted significant predictive effects for those in Profile 3.
In terms of feedback handling, linguistic self-efficacy, self-regulatory self-efficacy, and classroom performance self-efficacy demon­
strated significant predictive effects for all the participants (βLS = 0.135, p = .034; βSRS = 0.375, p = .000; βCPS = − 0.312, p = .002);

9
J. Zhang and L.J. Zhang System 122 (2024) 103253

Table 5
Results of differences in SRL writing strategies and writing achievement across latent profiles.
Factors Profile pair Statistics

MD SE Sig.

TR 1–2 − 1.476 .085 .000

1–3 − 2.704 .097 .000
2–3 − 1.228 .074 .000
PL 1–2 − .605 .137 .000
1–3 − 1.228 .156 .000
2–3 − .623 .119 .000
FH 1–2 − .316 .136 .021
1–3 − .999 .155 .000
2–3 − .684 .118 .000
IE 1–2 − .108 .136 .430
1–3 − .457 .155 .003
2–3 − .349 .118 .003
MST 1–2 − .318 .150 .034
1–3 − 1.057 .170 .000
2–3 − .738 .130 .000
WA 1–2 − 2.267 .958 .018
1–3 − 3.240 1.089 .003
2–3 − .972 .830 .242

Note. TR = text revising, PL = peer learning, FH = feedback handling, IE = interest enhancement, MST = motivational self-talk, WA = writing
achievement.

Table 6
Model fit indices for the models for path analyses.
χ2 df CFI TLI RMSEA SRMR AIC BIC

TR 0 0 1 1 0 0 1078.585 1150.022
PL 0 0 1 1 0 0 1090.945 1162.382
FH 0 0 1 1 0 0 1067.462 1138.898
IE 0 0 1 1 0 0 1155.886 1227.322
MST 0 0 1 1 0 0 980.015 1051.452
Overall 0 0 1 1 0 0 4989.702 5148.45

Note. TR = text revising, PL = peer learning, FH = feedback handling, IE = interest enhancement, MST = motivational self-talk.

Table 7
Differences in predictiveness of writing self-efficacy on SRL writing strategies.
Factors overall Profile 1 Profile 2 Profile 3

β p β p β p β p

TR LS .247 .000 .089 .533 .284 .001 .113 .337

SRS .244 .000 .119 .364 .185 .015 .28 .01
CPS − .087 .129 .032 .777 − .121 .072 .016 .876
GPS .13 .052 .189 .101 .082 .381 − .127 .18
PL LS .053 .381 .004 .976 .094 .223 .012 .932
SRS .263 .000 .34 .001 .196 .01 .096 .392
CPS .21 .002 − .055 .672 .174 .01 .222 .076
GPS − .054 .440 − .012 .917 − .026 .778 − .088 .521
FH LS .135 .034 .067 .623 .145 .104 .041 .771
SRS .375 .000 .427 .000 .198 .003 .338 .001
CPS − .209 .004 − .312 .002 − .097 .163 − .209 .016
GPS − .041 .555 − .002 .991 − .043 .651 − .045 .66
IE LS .04 .464 .051 .663 − .01 .895 .113 .393
SRS .371 .000 .196 .122 .343 .000 .253 .022
CPS .081 .186 − .006 .951 .057 .343 .081 .43
GPS − .024 .723 − .015 .904 .032 .736 − .123 .327
MST LS .102 .052 .016 .906 .108 .086 .182 .144
SRS .472 .000 .462 .000 .419 .000 .241 .024
CPS − .038 .526 − .086 .42 − .016 .815 .111 .195
GPS .04 .520 − .045 .721 .089 .313 .035 .756

Note. LS = linguistic self-efficacy, SRS = self-regulatory self-efficacy, CPS = classroom performance self-efficacy, GPS = genre-based self-efficacy, TR
= text revising, PL = peer learning, FH = feedback handling, IE = interest enhancement, MST = motivational self-talk.

10
J. Zhang and L.J. Zhang System 122 (2024) 103253

self-regulatory self-efficacy and classroom performance self-efficacy also showed a significant predictive effect for those in Profile 1
(βSRS = 0.427, p = .000; βCPS = − 0.312, p = .002); self-regulatory self-efficacy also showed significant predictive effects for those in
Profile 2 (β = 0.198, p = .003); self-regulatory self-efficacy and classroom performance self-efficacy exerted significant predictive
effects for those in Profile 3 (βSRS = 0.338, p = .001; βCPS = − 0.209, p = .016). In terms of interest enhancement, self-regulatory self-
efficacy showed significant predictive effects for all the participants (β = 0.371, p = .003), those in Profile 2 (β = 0.343, p = .000) and
those in Profile 3 (β = 0.253, p = .022) rather than those in Profile 1. In terms of motivational self-talk, self-regulatory self-efficacy
showed significant predictive effects for all the participants (β = 0.472, p = .000) and those in Profiles 1 to 3 (βprofile1 = 0.462, p =
.000; βprofile2 = 0.419, p = .000; βprofile3 = 0.241, p = .000).

4.4. Profile differences in predictiveness of writing self-efficacy and SRL writing strategies on writing achievement

Path analyses were utilized to determine the differences across profiles of writing self-efficacy in predictive effects of writing self-
efficacy and SRL writing strategies on writing achievements. Table 8 provides the details about the model fit indices of the models for
path analyses, showing that these models are fully saturated. The results of the path analyses are displayed in Table 9.
As the results of path analyses in Table 9 show, for the overall participants in the current study, linguistic self-efficacy and genre-
based performance self-efficacy exposed a significant predictive effect on writing achievements (βLS = 0.133, p = .041; βGPS = .179, p =
.013). In contrast, for the participants in Profile 1, motivational self-talk showed a significant negative predictive impact on writing
achievements (β = − 0.46, p = .011); for those in Profile 2, genre-based performance self-efficacy could significantly predict writing
achievements (β = 0.184, p = .036); for those in Profile 3, neither facets of writing self-efficacy nor SRL writing strategies demonstrated
significant predictive effect on writing achievements.

5. Discussion

The current study aimed to examine profiles of writing self-efficacy and the heterogeneity of its relationship with SRL writing
strategies among L2 students. According to the LPAs, three profiles of writing self-efficacy among the participants in the current study
came into sight: “Low on All Self-Efficacy”, “Average on All Self-Efficacy”, and “High on All Self-Efficacy”. Moreover, ANOVA and
Welch’s Tests showed that there were significant differences across those identified profiles of writing self-efficacy in terms of writing
self-efficacy (i.e., linguistic self-efficacy, self-regulatory self-efficacy, classroom performance self-efficacy, and genre-based perfor­
mance self-efficacy), SRL writing strategies (text revising, peer learning, feedback handling, interest enhancement, and motivational
self-talk), and writing achievements. Furthermore, the results of the post hoc LSD analyses and Tamhane’s T2 Tests showed that all the
profile pairs were significantly distinctive in the aforementioned variables except the following pairs: The Profile 1 versus Profile 2 pair
in interest enhancement and the Profile 2 versus Profile 3 pair in writing achievement. Besides, path analyses revealed differences
across the profiles of writing self-efficacy in the predictive effect of writing self-efficacy on SRL writing strategies and the effects of
writing self-efficacy and SRL writing strategies on writing achievement.

5.1. Profiles of writing self-efficacy

Three profiles of writing self-efficacy emerged among the sample of L2 students involved in the current study: “Low on All Self-
Efficacy”, “Average on All Self-Efficacy”, and “High on All Self-Efficacy”. This finding is similar to that of Kim et al. (2015) found three
profiles of self-efficacy for learning English among Korean-speaking EFL learners: high, medium, and low self-efficacy. What makes the
current study distinct from Kim et al. (2015) is the different conceptualizations of self-efficacy adopted. Kim et al. (2015) adopted the
unitary conceptualization of self-efficacy and used the mean score of all the items measuring self-efficacy as the basis of LPAs. In
contrast, self-efficacy was treated as a multidimensional construct in the current study, and the LPAs were based on all the items, as
illustrated in Fig. 1. Therefore, the findings in the current study revealed that three profiles differed significantly in facets of writing
self-efficacy, thus contributing to our elaborate grasp of the heterogeneity of L2 students in writing self-efficacy.
The current study also demonstrated significant differences across the identified profiles of writing self-efficacy in terms of SRL
writing strategies and writing achievements. L2 students assigned to the High on All Self-efficacy profile reported higher use of SRL
writing strategies than those assigned to the Average on All Self-efficacy profile, followed by those assigned to the Average on All Self-
efficacy profile. The finding might imply different patterns of integrating writing self-efficacy and SRL writing strategies among L2
students: The Low on All Self-Efficacy profile + low use of SRL writing strategies, the Average on All Self-Efficacy profile + mediate use
of SRL writing strategies, the High on All Self-Efficacy profile + high use of SRL writing strategies. These patterns might suggest a
positive correlation between facets of writing self-efficacy and the use of SRL writing strategies.
Theoretically, three profiles of writing self-efficacy found in this study might help us identify previously unrecognized latent
subgroups in the population of L2 students, each characterized by distinctive levels of writing self-efficacy (i.e., high, medium, and

Table 8
Model fit indices for the models for path analyses.
χ2 df CFI TLI RMSEA SRMR AIC BIC

Profiles 0 0 1 1 0 0 2652.535 2783.052

Overall 0 0 1 1 0 0 2624.378 2668.034

11
J. Zhang and L.J. Zhang System 122 (2024) 103253

Table 9
Differences in predictiveness of writing self-efficacy and SRL writing strategies on writing achievements.
Factors overall Profile 1 Profile 2 Profile 3

β p β p β p β p

LS .133 .041 .012 .943 .117 .19 .165 .158

SRS − .033 .612 .016 .917 .023 .769 − .078 .507
CPS − .004 .962 .11 .558 .000 .999 .069 .519
GPS .179 .013 .126 .371 .184 .036 − .071 .613
TR − .016 .78 − .038 .846 − .034 .641 − .04 .741
PL − .017 .761 .101 .532 .035 .682 − .176 .169
FH .034 .59 .247 .156 .025 .77 − .034 .862
IE − .002 .975 .106 .543 − .044 .572 − .032 .845
MST − .129 .06 − .46 .011 − .115 .194 .059 .768

Note. LS = linguistic self-efficacy, SRS = self-regulatory self-efficacy, CPS = classroom performance self-efficacy, GPS = genre-based self-efficacy, TR
= text revising, PL = peer learning, FH = feedback handling, IE = interest enhancement, MST = motivational self-talk.

low). Furthermore, the findings of three latent profiles of writing self-efficacy might suggest that L2 students’ writing self-efficacy
progresses on the continuum from low through medium to high, with its dimensions simultaneously interactong with each other,
contributing to our theoretical exploration of self-efficacy development in L2 writing. Besides, significant differences across these
profiles in the SRL writing strategies and writing achievements might provide further evidence of concurrent validity for identifying
latent profiles of writing self-efficacy in L2 students.

5.2. Profile differences in the predictive effects of writing self-efficacy on SRL writing strategies

The results of path analyses revealed that for the whole sample, their linguistic self-efficacy could predict their employment of text
revising and feedback handling; their classroom performance self-efficacy could predict their use of peer learning and feedback
handling; their self-regulatory self-efficacy could predict their utilization of all the SRL writing strategies.
These findings contributed to our understanding of the nuances of the predictive effects of writing self-efficacy on SRL writing by
discerning differences among L2 students involved in the profiles of writing self-efficacy. For instance, the predictive impact of lin­
guistic self-efficacy on text revising occurred only in the Average on All Self-efficacy Profile; the predictive effect of linguistic self-
efficacy on feedback handling did not appear in any of the identified profiles. In the same vein, the predictive influence of class­
room performance self-efficacy on peer learning appeared only in the Average on All Self-efficacy Profile, whereas the predictive
influence of classroom performance self-efficacy on feedback handling showed the reverse results. Besides, the predictive impacts of
self-regulatory self-efficacy on text revising and interest enhancement repeated in the Average on All Self-efficacy Profile and the High
on All Self-efficacy Profile rather than in the Low on All Self-efficacy Profile. In contrast, the predictive impacts of self-regulatory self-
efficacy on peer learning repeated in the Low on All Self-efficacy Profile and the Average on All Self-efficacy Profile rather than in the
High on All Self-efficacy Profile. However, the current study also found that the predictive effect of self-regulatory self-efficacy on
feedback handling and motivational self-talk was reiterated across the identified profiles.
The differences in the predictive effects of writing self-efficacy on SRL writing strategies between the profiles and the overall sample
might suggest that the traditional variable-centered approach might conceal essential subgroup differences in these effects.
Furthermore, the differences in the predictive effects across the profiles of writing self-efficacy may imply that the predictive effects of
writing self-efficacy on SRL writing strategies might be gradient; in other words, these predictive effects vary as a function of the
development of writing self-efficacy in L2 students. On the other hand, the consistency of the predictive effect of self-regulatory self-
efficacy feedback handling and motivational self-talk across the profiles may provide more robust and valid evidence that self-
regulatory self-efficacy is essential for L2 students to employ certain SRL writing strategies, such as feedback handling and motiva­
tional self-talk (Cheng & Zhang, 2024; Zhang, 2022).

5.3. Profile differences in predictive effects of writing self-efficacy and SRL writing strategies on L2 writing achievement

Path analyses revealed that linguistic and genre performance self-efficacy acted as significant predictors of writing achievement for
all the participants, which might reinforce the importance of linguistic and genre performance self-efficacy (e.g., Teng et al., 2018;
Zhang et al., 2023). However, the path analyses discovered the differences across the profiles of writing self-efficacy in terms of the
predictive effect of writing self-efficacy and SRL writing strategies. Specifically, for the Low on All Self-efficacy profile, only their
motivational self-talk could negatively predict their writing achievement; for the Average on All Self-efficacy profile, only their
genre-based performance self-efficacy exerted a significant impact on their writing achievement. In contrast, for the High on All
Self-efficacy profile, neither facets of writing self-efficacy nor SRL writing strategies served as a significant predictor of their writing
achievement. The profile differences in the predictive effects of writing self-efficacy and SRL writing strategies on writing achievement
may further our understanding of these effects by providing more details about how these effects vary across different subgroups.
Moreover, the profile differences in these aforementioned effects might suggest that as L2 students develop and enhance their writing
self-efficacy, their writing achievement will not be improved correspondingly. Besides, it should be cautious when utilizing L2 stu­
dents’ writing self-efficacy to predict their writing achievement.

12
J. Zhang and L.J. Zhang System 122 (2024) 103253

What is unexpected in the current study is the finding that only motivational self-talk exerted a negative predictive effect on writing
achievement in the Low on All Self-efficacy profile. This result was in great contrast to earlier findings demonstrating the predictive
effects of most SRL writing strategies on writing achievement (e.g., Sun & Wang, 2020; Teng & Huang, 2019; Teng & Zhang, 2016). As
stated before, the Low on ALL Self-efficacy profile was characterized by low levels of writing self-efficacy dimensions. For them,
writing an argumentative essay in a limited time would be challenging and strenuous and thus impose a large burden on them. Thus,
they may frequently employ motivational self-talk to encourage themselves to complete the assigned writing task. Frequent
employment of motivational self-talk in the writing process might consume a certain amount of time that should be utilized for
improving writing performance and thus severely impede the quality of the assigned writing task.

6. Conclusion

The current study explored the relationship among writing self-efficacy, SRL writing strategies, and writing achievement among L2
students by employing LPA. Based on the LPA results, three profiles of writing self-efficacy were identified among L2 students in the
current study: “Low on All Self-efficacy”, “Average on All Self-efficacy”, and “High on All Self-efficacy”. Moreover, ANOVA and
Welch’s Tests uncovered that L2 students belonging to these three profiles were significantly distinct in their writing self-efficacy and
reported use of SRL writing strategies. In addition, path analyses unveiled that there were differences among these three profiles in
terms of the predictive effects of writing self-efficacy on SRL writing strategies and those of writing self-efficacy and SRL writing
strategies on writing achievement.
A few implications might emerge from the findings for L2 writing research and pedagogical practices. Methodologically, using the
LPAs, we may be able to identify various profiles that would help us discern the group-level differences among L2 students regarding
writing self-efficacy, thus offering us a base for examining the heterogeneity in the predictive effect of writing self-efficacy on SRL
writing strategies and their effects on L2 writing achievement. Pedagogically, the identification of profiles of writing self-efficacy may
inform teachers of the delicacy of L2 students’ writing self-efficacy. Moreover, the findings of profile differences in SRL writing
strategies and in the predictive effects of writing self-efficacy on SRL writing strategies and those of writing self-efficacy and SRL
writing strategies on writing achievement might alert instructors of the inappropriateness of one-fit-all perception of improving
students’ writing self-efficacy to enhance their use of SRL writing strategies. These findings may further suggest that teachers should
tailor their teaching techniques to specifically target the particular learning problems that students in different profiles may encounter.
For instance, for students in the Low on All Self-efficacy profile, teachers should help them strengthen their self-regulatory self-efficacy
and classroom performance self-efficacy in order to enable them to increase the use of text revising, interest enhancement, and peer
learning. In contrast, for those in the Average on All Self-efficacy Average and High on All Self-efficacy profiles, teachers could try to
assist them in maintaining their current level of writing self-efficacy to facilitate their use of SRL writing strategies. Besides, additional
advice may be offered to those in the Low on All Self-efficacy and High on All Self-efficacy profiles: They may be recommended to
calibrate their classroom performance self-efficacy because a too low or high level of that self-efficacy could exert a significant negative
influence on feedback handling.
Undoubtedly, there may be some limitations that deserve attention. Firstly, it should be noted that the current study mainly
involved beginning and intermediate L2 students as participants rather than advanced ones. As a result, the conclusion of the current
study cannot be extrapolated to those advanced learners. Thus, scholars might be advised to include advanced L2 students in future
studies to confirm the results of the current study. Secondly, to evaluate the participants’ writing achievement, only a writing task (i.e.,
argumentative writing) was utilized, and thus, it could not capture the complete picture of writing achievement. Accordingly, it might
be recommended that scholars include more writing tasks (i.e., narrative and expositive writing) to test whether the profile differences
in the predictive effects of writing self-efficacy and SRL writing strategies on writing achievement are stable across various writing
tasks. Thirdly, because of the inherent shortcoming of self-reported questionnaires, we suggest that scholars incorporate the data from
other sources (e.g., stimulated interviews) to verify the profiles of writing self-efficacy identified in the current study.

Ethic statement

This study is approved by the University of Auckland Human Participants Ethics Committee (Protocol no: UAHPEC21883).

CRediT authorship contribution statement

Jianhua Zhang: Writing – review & editing, Writing – original draft, Visualization, Software, Methodology, Investigation, Formal
analysis, Data curation, Conceptualization. Lawrence Jun Zhang: Writing – review & editing, Supervision, Project administration,
Methodology, Conceptualization.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

13
J. Zhang and L.J. Zhang System 122 (2024) 103253

Appendix B. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.system.2024.1032532.

References

Bai, B. (2015). The effects of strategy-based writing instruction in Singapore primary schools. System, 53, 96–106. https://doi.org/10.1016/j.system.2015.05.009
Bai, B., & Guo, W. (2018). Influences of self-regulated learning strategy use on self-efficacy in primary school students’ English writing in Hong Kong. Reading &
Writing Quarterly, 34(6), 523–536. https://doi.org/10.1080/10573569.2018.1499058
Bai, B., Guo, W., & Wang, C. (2022). Relationships between struggling EFL writers’ motivation, self-regulated learning (SRL), and writing competence in Hong Kong
primary schools. Applied Linguistics Review. https://doi.org/10.1515/applirev-2020-0131. online first.
Bai, R., Hu, G., & Gu, P. Y. (2014). The relationship between use of writing strategies and English proficiency in Singapore primary schools. Asia-Pacific Education
Researcher, 23(3), 355–365. https://doi.org/10.1007/s40299-013-0110-0
Bai, B., Shen, B., & Mei, H. (2020). Hong Kong primary students’ self-regulated writing strategy use: Influences of gender, writing proficiency, and grade level. Studies
In Educational Evaluation, 65, Article 100839. https://doi.org/10.1016/j.stueduc.2020.100839
Bai, B., Wang, J., & Nie, Y. (2021). Self-efficacy, task values and growth mindset: What has the most predictive power for primary school students’ self-regulated
learning in English writing and writing competence in an Asian Confucian cultural context? Cambridge Journal of Education, 51(1), 65–84. https://doi.org/
10.1080/0305764X.2020.1778639
Berlin, K. S., Williams, N. A., & Parra, G. R. (2014). An introduction to latent variable mixture modeling (part 1): Overview and cross-sectional latent class and latent
profile analyses. Journal of Pediatric Psychology, 39, 174–187.
Bruning, R., Dempsey, M. S., Kauffman, D. F., McKim, C., & Zumbrunn, S. (2013). Examining dimensions of self-efficacy for writing. Journal of Educational Psychology,
105(1), 25–38. https://doi.org/10.1037/a0029692
Brunstein, J. C., & Glaser, C. (2011). Testing a path-analytic mediation model of how self-regulated writing strategies improve fourth graders’ composition skills: A
randomized controlled trial. Journal of Educational Psychology, 103(4), 922–938. https://doi.org/10.1037/a0024622
Chen, J., & Zhang, L. J. (2019). Assessing student-writers’ self-efficacy beliefs about text revision in EFL writing. Assessing Writing, 40, 27–41. https://doi.org/
10.1016/j.asw.2019.03.002
Chen, J., Zhang, L. J., & Chen, X. (2022). L2 learners’ self-regulated learning strategies and self-efficacy for writing achievement: A latent profile analysis. Language
Teaching Research, 1–22. https://doi.org/10.1177/13621688221134967
Cheng, X. L., & Zhang, L. J. (2024). Engaging secondary school students with peer feedback in second language (L2) writing classrooms: A mixed-methods study.
Studies in Educational Evaluation, 81, Article 101337. https://doi.org/10.1016/j.stueduc.2024.101337.
Dörnyei, Z., & Taguchi, T. (2010). Questionnaires in second language research: Construction, administration, and processing (2nd ed.). Routledge.
Glaser, C., & Brunstein, J. C. (2007). Improving fourth-grade students’ composition skills: Effects of strategy instruction and self-regulation procedures. Journal of
Educational Psychology, 99(2), 297–310. https://doi.org/10.1037/0022-0663.99.2.297
Graham, S., & Harris, K. R. (2000). The role of self-regulation and transcription skills in writing and writing development. Educational Psychologist, 35(1), 3–12.
https://doi.org/10.1207/S15326985EP3501_2
Guo, W., & Bai, B. (2019). Effects of self-regulated learning strategy use on motivation in EFL writing: A comparison between high and low achievers in Hong Kong
primary schools. Applied Linguistics Review, 0(0). https://doi.org/10.1515/applirev-2018-0085
Howard, M. C., & Hoffman, M. E. (2018). Variable-centered, person-centered, and person-specific approaches. Organizational Research Methods, 21(4), 846–876.
https://doi.org/10.1177/1094428117744021
Huang, Y., & Zhang, L. J. (2020). Does a process-genre approach help improve students’ argumentative writing in English as a foreign language? Findings from an
intervention study. Reading & Writing Quarterly, 36(4), 339–364. https://doi.org/10.1080/10573569.2019.1649223
Jacobs, H. L., Zingraf, S. A., Wormuth, D. R., Hartfiel, V. F., & Hughey, J. B. (1981). Testing ESL composition: A practical approach. Newbury House. https://eric.ed.gov/
?id=ed217708.
Jung, T., & Wickrama, K. A. S. (2008). An introduction to latent class growth analysis and growth mixture modeling. Social and Personality Psychology Compass, 2(1),
302–317. https://doi.org/10.1111/j.1751-9004.2007.00054.x
Kim, D.-H., Wang, C., & Ahn, H. S. (2015). English language learners’ self-efficacy profiles and relationship with self-regulated learning strategies. Learning and
Individual Differences, 38, 136–142. https://doi.org/10.1016/j.lindif.2015.01.016
Kline, R. B. (2016). Principles and practice of structural equation modeling (4th ed.). The Guilford Press.
McCarthy, P., Meier, S., & Rinderer, R. (1985). Self-efficacy and writing: A different view of self-evaluation. College Composition & Communication, 36(4), 465. https://
doi.org/10.2307/357865
Muthen, L. K., & Muthen, B. (2017). Mplus user’s guide: Statistical analysis with latent variables (8th ed.). Muthén & Muthén.
Pajares, F. (2003). Self-efficacy beliefs, motivation, and achievement in writing: A review of the literature. Reading & Writing Quarterly, 19(2), 139–158. https://doi.
org/10.1080/10573560308222
Pajares, F., & Johnson, M. J. (1994). Confidence and competence in writing: The role of self-efficacy, outcome expectancy, and apprehension. Research in the Teaching
of English, 28(3), 313–331.
Pajares, F., Miller, M. D., & Johnson, M. J. (1999). Gender differences in writing self-beliefs of elementary school students. Journal of Educational Psychology, 91(1),
50–61. https://doi.org/10.1037/0022-0663.91.1.50
Pajares, F., & Valiante, G. (1997). Influence of self-efficacy on elementary students’ writing. The Journal of Educational Research, 90(6), 353–360. https://doi.org/
10.1080/00220671.1997.10544593
Pajares, F., & Valiante, G. (1999). Grade level and gender differences in the writing self-beliefs of middle school students. Contemporary Educational Psychology, 24(4),
390–405. https://doi.org/10.1006/ceps.1998.0995
Peugh, J., & Fan, X. (2013). Modeling unobserved heterogeneity using latent profile analysis: A Monte Carlo simulation. Structural Equation Modeling: A
Multidisciplinary Journal, 20, 616–639.
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). Academic.
Spurk, D., Hirschi, A., Wang, M., Valero, D., & Kauffeld, S. (2020). Latent profile analysis: A review and “how to” guide of its application within vocational behavior
research. Journal of Vocational Behavior, 120, Article 103445. https://doi.org/10.1016/j.jvb.2020.103445
Sun, T., & Wang, C. (2020). College students’ writing self-efficacy and writing self-regulated learning strategies in learning English as a foreign language. System, 90,
Article 102221. https://doi.org/10.1016/j.system.2020.102221
Teng, L. S. (2022). Self-regulated learning and Second Language Writing: Fostering strategic language learners. Springer Nature.
Teng, F., & Huang, J. (2019). Predictive effects of writing strategies for self-regulated learning on secondary school learners’ EFL writing proficiency. TESOL Quarterly,
53(1), 232–247. https://doi.org/10.1002/tesq.462
Teng, L. S., Sun, P. P., & Xu, L. (2018). Conceptualizing writing self-efficacy in English as a foreign language contexts: Scale validation through structural equation
modeling. TESOL Quarterly, 52(4), 911–942. https://doi.org/10.1002/tesq.432

14
J. Zhang and L.J. Zhang System 122 (2024) 103253

Teng, L. S., & Zhang, L. J. (2016). A questionnaire-based validation of multidimensional models of self-regulated learning strategies. The Modern Language Journal, 100
(3), 674–701. https://doi.org/10.1111/modl.12339
Teng, L. S., & Zhang, L. J. (2018). Effects of motivational regulation strategies on writing performance: A mediation model of self-regulated learning of writing in
English as a second/foreign language. Metacognition and Learning, 13(2), 213–240. https://doi.org/10.1007/s11409-017-9171-4
Woodrow, L. (2011). College English writing affect: Self-efficacy and anxiety. System, 39(4), 510–522. https://doi.org/10.1016/j.system.2011.10.017
Xu, T. S., Zhang, L. J., & Gaffney, J. S. (2022). Examining the relative effectiveness of task complexity and cognitive demands on students’ writing in a second
language. Studies in Second Language Acquisition, 44(2), 483–506. https://doi.org/10.1017/S0272263121000310
Xu, T. S., Zhang, L. J., & Gaffney, J. S. (2023a). A multidimensional approach to assessing the effects of task complexity on L2 students’ argumentative writing.
Assessing Writing, 55, 1–16. https://doi.org/10.1016/j.asw.2022.100690, 100690.
Xu, Z., Zhang, L. J., & Parr, J. M. (2023b). Incorporating peer feedback in writing instruction: Examining its effects on English-as-a-foreign-language (EFL) learners’
writing performance. International Review of Applied Linguistics in Language Teaching, 61(4), 1337–1364. https://doi.org/10.1515/iral-2021-0078
Zabihi, R. (2018). The role of cognitive and affective factors in measures of L2 writing. Written Communication, 35(1), 32–57. https://doi.org/10.1177/
0741088317735836
Zhang, J., Zhang, L. J., & Zhu, Y. (2023). Development and validation of a genre-based second language (L2) writing self-efficacy scale. Frontiers in Psychology, 14,
Article 1181196. https://doi.org/10.3389/fpsyg.2023.1181196
Zhang, L. J. (2022). L2 writing: Toward a theory-practice praxis. In E. Hinkel (Ed.), Handbook of practical second language teaching and learning (pp. 331–343).
Routledge. https://doi.org/10.4324/9781003106609-27.
Zimmerman. (2000). Self-efficacy: An essential motive to learn. Contemporary Educational Psychology, 25(1), 82–91. https://doi.org/10.1006/ceps.1999.1016
Zimmerman, B. J., & Bandura, A. (1994). Impact of self-regulatory influences on writing course attainment. American Educational Research Journal, 31(4), 845–862.
https://doi.org/10.3102/00028312031004845
Zimmerman, B. J., & Kitsantas, A. (1999). Acquiring writing revision skill: Shifting from process to outcome self-regulatory goals. Journal of Educational Psychology, 91
(2), 241–250. https://doi.org/10.1037/0022-0663.91.2.241
Zimmerman, B. J., & Kitsantas, A. (2002). Acquiring writing revision and self-regulatory skill through observation and emulation. Journal of Educational Psychology, 94
(4), 660–668. https://doi.org/10.1037//0022-0663.94.4.660
Zimmerman, B. J., & Risemberg, R. (1997). Becoming a self-regulated writer: A social cognitive perspective. Contemporary Educational Psychology, 22, 73–101.
Zimmerman, B. J., & Schunk, D. H. (Eds.). (2011). Handbook of self-regulation of learning and performance. Routledge. https://doi.org/10.4324/9780203839010.

Jianhua Zhang is a Professor of English Applied Linguistics at the School of Foreign Languages, Sichuan University of Arts and Science, Dazhou, China. He has earned a
PhD in Education (Applied Linguistics/TESOL) from The University of Auckland, New Zealand. His research interests lie in educational psychology, second language
writing and second language acquisition. He is interested in teaching English as a foreign language (EFL), particularly writing and has published research in SSCI-
indexed journals such as Asia Pacific Journal of Education (Routledge), Frontiers in Psychology (Frontiers), Journal of Quantitative Linguistics (Routledge), Reading &
Writing Quarterly (Routledge), Learning and Instruction (Elsevier), System (Elsevier), Linguistics and Education (Elsevier), and Computer Assisted Language Learning
(Routledge)

Lawrence Jun Zhang, Ph.D., is Professor of Linguistics-in-Education and Associate Dean for the Faculty of Education and Social Work, The University of Auckland, New
Zealand. His major interests and publications are on the psychology of language learning and teaching, especially learner metacognition, L2 reading-writing devel­
opment and teacher assessment literacy. His publications have appeared in journals such as Applied Linguistics (Oxford), Applied Linguistics Review (de Gruyter), British
Journal of Educational Psychology (Wiley), Discourse Processes (Routledge), Journal of Multilingual and Multicultural Development (Routledge), Journal of Second Language
Writing (Elsevier), Journal of Psycholinguistic Research (Springer), Measurement (Routledge), Modern Language Journal (Wiley), Perceptual and Motor Skills (Sage), TESOL
Quarterly (Wiley), Language Teaching Research (Sage), Learning and Instruction (Elsevier), RELC Journal (Sage), System (Elsevier), Frontiers in Psychology (Frontiers), among
others. He serves on editorial boards for Applied Linguistics Review (de Gruyter), Australian Review of Applied Linguistics (Benjamins), Chinese Journal of Applied Linguistics
(de Gruyter), Journal of Second Language Writing (Elsevier), Metacognition and Learning (Springer), Journal of Second Language Studies (Benjamins), Language Teaching for
Young Learners (Benjamins) and RELC Journal (Sage). He is Co-Editor-in-Chief for System (Elsevier). He was honored by the TESOL International Association (USA) in
2016 with the award of “50 at 50”, which acknowledged “50 Outstanding Leaders around the world in the field of TESOL”. In the Stanford University Rankings 2022 and
2023, he was listed in the top 2% of Scientists in the World in the disciplinary areas of Linguistics/Applied Linguistics/Educational Linguistics.

15