The Impact Students’ Preferences Have on Written Corrective Feedback Effectiveness

O impacto que as preferências dos alunos têm na eficácia do feedback corretivo escrito

ABSTRACT:

Although written corrective feedback (WCF) has been shown to promote linguistic accuracy among L2 learners, the debate regarding its effectiveness remains open due to differing results. In this respect, individual preferences have been identified as crucial factors that must be comprehensively researched as they could mediate the effect of a WCF treatment. Although this conjecture has attracted researchers’ attention, it remains an underexplored variable as the existent studies have primarily focused on the strategies learners prefer and how these align with teachers’ responses to written texts; however, there is a lack of a more experimental approach to observing this phenomenon. Thus, this quantitative research, with a pre/posttest design, aimed to determine the effectiveness of WCF when the treatment is tailored to learners’ preferences. The sample consisted of 61 Pedagogy in English students from a Chilean university. Students’ preferences for WCF were collected through a questionnaire. Then, the participants were divided into a control group and two experimental groups: direct plus metalinguistic explanation and indirect plus metalinguistic clarification. Each experimental group was composed of students who had selected the feedback strategy they received and those who had not. Results indicate that both experimental groups significantly outperformed the control group, but no statistically significant differences were found between them. Additionally, no significant relationship was found between tailoring the feedback strategies to students’ preferences and the effectiveness of any type of feedback, which is substantial in an educational context since considering student preferences does not always have a positive impact on the learning process.

Keywords:
written corrective feedback; students’ preferences; acquisition; grammatical accuracy

RESUMO:

Apesar de o feedback corretivo escrito (FCE) ter sido demonstrado como promotor de precisão linguística entre aprendizes de L2, o debate sobre sua real eficácia ainda permanece aberto devido a resultados divergentes. Nesse contexto, as preferências individuais foram identificadas como fatores cruciais que exigem uma pesquisa abrangente, uma vez que poderiam mediar o efeito de um tratamento com FCE. Embora essa conjectura tenha chamado a atenção dos pesquisadores, ela continua sendo uma variável pouco explorada, uma vez que os estudos existentes, na maioria das vezes, enfatizaram as estratégias que os aprendizes preferem e como elas se alinham com as respostas dos professores a textos escritos; no entanto, falta uma abordagem mais experimental para observar esse fenômeno. Portanto, esta pesquisa quantitativa, com um desenho de pré/pós-teste, teve como objetivo determinar a eficácia do FCE quando o tratamento é ajustado às preferências dos aprendizes. A amostra correspondeu a 61 estudantes de Pedagogia em Inglês de uma universidade chilena. As preferências dos alunos em relação ao FCE foram coletadas por meio de um questionário. Em seguida, os participantes foram divididos em um grupo de controle e dois grupos experimentais: explicação direta mais explicação metalinguística e clarificação indireta mais explicação metalinguística. Cada grupo experimental era composto por alunos que haviam selecionado a estratégia de feedback que receberam e alunos que não o haviam feito. Os resultados mostram que ambos os grupos experimentais tiveram um desempenho significativamente melhor do que o grupo de controle, mas não foram encontradas diferenças estatísticas entre eles. Além disso, não foi encontrada uma relação significativa entre o ajuste das estratégias de feedback às preferências dos alunos e a eficácia de qualquer dos tipos de feedback, o que é substancial em um contexto educacional, já que considerar as preferências dos alunos nem sempre tem um impacto positivo no processo de aprendizado.

Palavras-chave:
feedback corretivo escrito; preferências dos alunos; aquisição; precisão gramatical

1 Introduction

Written corrective feedback (WCF) has been shown to be an effective tool to promote the acquisition of linguistic forms (Bitchener; Knoch, 2008 ; Bitchener; Ferris, 2012 ; Shintani; Ellis; Suzuki, 2014 ; Ortiz, 2016 ; Muñoz, 2017 ; Muñoz; Ferreira, 2017 ; Muñoz; Sáez, 2019 ; Sinha; Nassaji, 2022 ); however, there still exist some discrepancies (see Karim; Nassaji, 2019 ). In this respect, Ellis ( 2010 ) emphasizes the need to examine WCF effectiveness from multiple perspectives. One of them corresponds to the surveillance of individual differences among second language (L2) learners, which have drawn researchers’ attention, as they may impact the effectiveness of WCF treatments. Nonetheless, while much research has explored how feedback promotes language acquisition, few studies have addressed individual elements influencing its effectiveness (Van der Kleij, 2019; Lee, 2020 ).

An individual difference that has drawn researchers’ attention corresponds to learners’ beliefs and preferences towards WCF strategies (Amrhein; Nassaji, 2010 ; Liu; Wu, 2019 ). According to Han ( 2017 , p. 2), “learner beliefs can not only influence teachers’ decision-making when providing WCF, but can also help explain variations among learners’ engagement with WCF and learning outcomes”. Thus, it is feasible to suggest that learners’ beliefs or preferences towards a specific type of correction could affect the reception, processing, and efficacy of WCF treatments.

Notwithstanding, most of the studies on this matter have adopted a descriptive approach providing evidence about learners’ most preferred correction techniques (see Table 1 ), and –to the best of our knowledge – only two have a more experimental approximation about tailoring WCF to learners’ preferences (Rummel; Bitchener, 2015 ; Sinha; Nassaji, 2022 ). Therefore, it is crucial to expand research on this matter to provide empirical evidence to demonstrate the actual impact of students’ preferences on WCF. This would not only contribute to broaden knowledge in an under-explored area in the field of error correction but would also sustain pedagogical decisions in an instructional context because – as Storch and Wigglesworth ( 2010 ) suggest – some students might reject feedback if they consider it inappropriate, which would lessen WCF effectiveness. Hence, the current study is aimed at determining the impact of tailoring WCF treatments to learners’ preferences.

2 Literature Review

2.1 Written Corrective Feedback

Written corrective feedback (WCF), also known as grammar correction or error correction (Ferris, 2012 ), is understood as a reaction to inaccuracies in students’ writing production. This has proven to be an effective tool when it comes to promoting the correct use of certain linguistic structures in second and foreign language settings (for a detailed review see Bitchener; Ferris, 2012 ). However, the discussion is far from over since “differential success in learners’ gaining from WCF has been observed” (Han, 2017 , p. 133).

A well-documented concern corresponds to the most suitable strategy for providing correction. The existing techniques differ in their degree of explicitness or directness, as well as in the possibility of combining them with a metalinguistic component or considering them as an independent correction technique (for a detailed analysis, see Ellis, 2009 ; Sheen, 2011 ). The ones that have received the largest attention are direct and indirect techniques, but “research on which form of feedback is most effective has produced mixed results” (Storch; Wigglesworth, 2010 , p. 304). Hence, the evidence about the effects of different types of WCF is still inconclusive (for a review, see Karim; Nassaji, 2019 ).

Another matter of interest in this area of knowledge is the type of errors that benefit the most from WCF. According to Suzuki, Nassaji and Sato ( 2019 ), the types of errors addressed with WCF can impact the effectiveness and durability of the treatment because not every deviation reacts similarly to correction. Similarly, the scope of feedback has also called the attention of researchers. Some studies advocate for comprehensive correction of every single error in writing, while others favour a more focused approach targeting one or a couple of errors for more positive effects (see Lee, 2020 for a detailed discussion). Notwithstanding, results are still up in the air (Frear; Chiu, 2015 ).

The source of feedback is an additional factor that may influence WCF effectiveness. No deciding conclusions have been developed due to the lack of evidence (Bitchener; Ferris, 2012 ). However, some studies that have addressed this variable show that teachers’ feedback stands as the most trustworthy response to error among learners (Amrhein; Nassaji, 2010 ; Ghani; Asgher, 2012 ; Hu, 2019 ). Peer feedback and self-correction have also emerged as valid sources of correction, but with some concerns about their quality and reliability (Miao; Badger; Zhen, 2006 ).

Regardless of the wide-ranging empirical evidence about WCF effectiveness, it has not been possible to reach indisputable results. The existent research cannot be easily generalised due to diverse factors. Firstly, the effectiveness shown by some WCF studies differs in diverse settings and/or among diverse participants, so many of the effects are confined to the explored scenarios. This is connected to a second limitation, namely, underexplored places, especially those where the acquisition of an additional language is restricted to the classroom. Additionally, existing data reveal learners do not benefit from feedback likewise.

Thus, contextual and individual factors have been pointed out as fundamental variables to be considered when researching WCF because they might add evidence to nourish current knowledge and to solve discrepancies (Han, 2017 ; Atkinson; Tardy, 2018 ; Lee, 2020 ). One of these individual factors corresponds to learners’ preferences or beliefs towards WCF, as they might impact students’ involvement in the treatment and, consequently, may affect the ultimate effectiveness of feedback strategies (Han, 2017 ; Zhang; Hyland, 2018 ; Saeli; Cheng, 2019 ).

2.2 The Impact of Learners’ Preferences on WCF Treatment

Students’ preferences or beliefs for WCF have been mentioned as mediating factors that may influence the overall effectiveness of error correction strategies. According to Rummel and Bitchener ( 2015 , p. 70), “if a student believes that the type of feedback he/she is receiving is effective, he/she may be more willing to engage with the feedback than a student who does not hold that belief”. By the same token, Han ( 2017 ) suggests learners’ beliefs can interfere with their engagement with WCF. Storch and Wigglesworth ( 2010 ) – on their behalf – propose students might reject error correction if they consider it useless. Furthermore, Bitchener ( 2017 ) argues that students who are less engaged in the learning process may be less receptive to WCF. Learners’ preferences may also impact the processing of feedback and, subsequently, the students’ accomplishment in language acquisition (Amrhein; Nassaji, 2010 ; Chen; Nassaji; Liu, 2016 ; Gan; An; Liu, 2021 ; Saeli; Cheng, 2019 ; Zhang, 2018 ).

Nonetheless, there still exist some divergent ideas regarding the impact of learners’ preferences in a WCF treatment. Some scholars have suggested they would not be as beneficial as it has been stated. In this regard, Dembo and Howard ( 2007 ) express that a teacher should scrutinise whether it is always favourable to match students’ preferences because they might not boost academic achievement. As Mozgalina ( 2015 ) suggests, students engage more when they have less opportunities to choose. However, the scarcity of experimental research on the effect of learners’ preferences on WCF hampers conclusive judgments (Lee, 2008 ; Rummel; Bitchener, 2015 ; Chen; Nassaji; Liu, 2016 ; Han, 2017 ). Thus, as Zhang et al . ( 2021 , p. 2) state, “exploring learners’ preference of WCF type helps gain a better understanding of the role of WCF in L2 learning”.

2.2.1 Some Research on WCF and Students’ Preferences

Current data evidence WCF as a facilitative tool to improve accuracy in writing. However, this effect seems to be not only linked to linguistic factors, but also to learners’ perceptions of WCF (Zhang et al ., 2021 ). Some of the research that has been conducted in this respect is summarised in the following table:

Table 1
. Research on WCF and Students’ Preferences

Study	Participants	Context	Findings
Lee ( 2008 )	ESL secondary students	Hong Kong	Students of any proficiency level want more written comments from teachers. Higher proficiency learners are more interested in error feedback. Both low and high proficiency students prefer more explicit error feedback.
Amrhein and Nassaji ( 2010)	From beginners to advanced students of ESL and teachers	Canada	Students prefer comprehensive feedback and consider grammatical and vocabulary inaccuracies to be the most important to receive correction.
Schulz ( 2001 )	Colombian and US foreign language learners	Colombia	Both groups of students value error correction, but Colombian apprentices show a stronger belief in the role of grammar in language learning.
Elwood and Bode ( 2014 )	First-year tertiary students	Japan	Learners prefer direct feedback on both content and mechanical errors. They show less enthusiasm about indirect correction as it may be counterproductive.
Karim and Nassaji ( 2015 )	Intermediate adult learners of ESL	Canada	Students consider indirect correction plus metalinguistic explanation the most effective strategy for error correction, and they acknowledge the lack of linguistic knowledge hinders this process. Learners believe both grammatical and non-grammatical errors must be corrected.
Diab ( 2015 )	Sophomore level English as a Second Language (ESL) course at an American university in Lebanon	Lebanon	Four out of five students from experimental group 1 prefer metalinguistic feedback because it makes them think about the error. Experimental group 2 shows differing predilections towards feedback.
Chen; Nassaji; Liu ( 2016 )	Intermediate, advanced-intermediate, and advanced English learners in a major provincial university of Mainland	China	Some of the results show students tend to hold a positive view on WCF. They acknowledge the necessity of grammar instruction in writing classes, yet not explicitly instructed. Students’ positive opinions about grammar decrease when gaining proficiency. Organisation is the most important aspect to be corrected, followed by grammatical and vocabulary errors. Localising plus indicating the type of error stands as the most preferred technique between intermediate and advanced-intermediate learners. Extended comments on writing’s overall quality are the most preferred.
Irwin ( 2018 )	Second-year students at a Japanese university	Japan	Apprentices believe error correction is important and desirable. The majority has a desire for at least two forms of feedback. Students believe vocabulary and expressions are the most important error to focus on, followed by grammatical mistakes. Indirect feedback stands as the most preferred technique.
Zhang and Cheng ( 2021 )	English major sophomores	China	Students’ perceptions were measured focusing on the effects of comprehensive WCF on accuracy, syntactic complexity, and fluency. Results show participants feel WCF helps them improve grammatical accuracy and awareness, and after the intervention they were able to write faster. However, learners believe feedback has no impact on syntactic complexity, as they cannot expand basic sentence structures.
Zhang et al . ( 2021 )	Tertiary students	Thailand	Students prefer a metalinguistic explanation for correcting grammatical, orthographic, and pragmatic deviations, while overt correction is desirable for dealing with lexical errors; i.e. preferences vary according to error types. The least preferred WCF type is indirect feedback, and grammar is considered the most important aspect in writing. Concerning the scope of WCF, no clear tendency is shown, with students divided in their preference for selective and comprehensive feedback. There are some effects of proficiency level on participants’ preferred WCF strategy, with indirect feedback being negatively evaluated by lower-level students.
Gan; An; Liu ( 2021 )	University students	China	Learners prefer teacher’s correction to self/peer feedback. Also, teacher feedback indirectly influenced course satisfaction and course exam results.
Muñoz; Ortiz; Sáez ( 2023 )	133 A2 to C1 students of an English pedagogy major	Chile	Learners believe grammar instruction is essential for learning an L2. Error correction is highly valued. Grammar errors are the most important to be corrected, but this idea declines when students gain proficiency. Direct correction with metalinguistic clarification is considered the most useful technique; especially, among less competent learners. Focused correction provided by the teacher is highly valued.

As mentioned previously, the existent research has followed a descriptive approach with questionnaires and interviews to examine students’ preferences. To the best of our knowledge, only Rummel and Bitchener ( 2015 ) and Sinha and Nassaji ( 2022 ) followed a more exploratory procedure to determine the impact of adapting WCF to learners’ preferences. The former conducted their research in an English language school in Laos. They identified upper-level students’ preferred WCF strategies, using a questionnaire and interviews. Then, students were placed into three treatment groups (direct, indirect, and metalinguistic feedback). Eight learners received the predilected feedback type and the others were treated with a different strategy. The findings show seven out of the eight students who received correction tailored to their preferences were able to eliminate all the targeted errors (past simple) on their second delayed posttest, while only four out of 34 of the learners who received feedback opposed to their preferences could get rid of all the targeted inaccuracies on their final posttest. The scholars conclude: “There is a strong reason to believe that these results could not have been reached if beliefs had not had some effect on learners uptake of the written CF they received” (Rummel; Bitchener, 2015 , p. 79). On their part, Sinha and Nassaji ( 2022 ) provided direct and indirect WCF to adult students with an intermediate level of proficiency at an English language centre in a university context in Canada. After the treatment, learners filled out a perception questionnaire. During the analyses, participants were regrouped by identifying those who favoured direct and indirect feedback in each of the experimental groups. Contrary to Rummel and Bitchener’s results, Shina and Nassaji found no connection between the students’ perceptions about WCF strategies and their grammatical accuracy gains.

Despite these two exploratory approaches to investigate the implications of learners’ preferences in WCF effectiveness, there is a series of limitations which needs to be overcome. Firstly, these results are circumscribed to the researched contexts (Laos and Canada), so they cannot be generalised. Secondly, the number of students who received feedback according to their preferences is restricted and, according to what both studies inform, the possibilities for actual selection were restricted to two or three WCF strategies; other elements such as the focus and source of feedback were disregarded. Thirdly, there is a necessity to research other linguistic forms and to include participants with diverse levels of proficiency and from different settings. As Zheng and Yu ( 2018 ) declare, lower levels of proficiency biases learners’ engagement with WCF. Fourthly, in the case of Rummel and Bitchener ( 2015 ) there is no information about the number of correct uses in the pretest and the posttests, which hinders the possibility to determine what “seven were able to eliminate all their targeted errors” means (Rummel; Bitchener, 2015 , p. 78-79). Thus, avoiding these constraints, the current research aims to determine the effectiveness of WCF strategies and the impact of tailoring the treatment to learners’ preferences.

3 Methodology

3.1 The Research

The present quasi-experimental research approach with a pretest-posttest design is intended to determine the effectiveness of two WCF strategies and the effect of tailoring the treatments to students’ preferences for correction techniques. Thus, the research was designed to answer the following research questions:

1. Which is the effect of two WCF strategies on the accurate use of a linguistic form?

2. Which is the impact of tailoring WCF to learners’ preferences?

3.2 Context

The research was conducted in a Chilean university in the Biobío region. The program was a Pedagogy major in teaching English as a foreign language (TEFL). The course corresponded to a first-year reading and writing class at pre-intermediate level.

3.3 Participants

The study included three intact classes of 61 first-year students, 39 females and 22 males. Their ages varied between 18 and 29 years and all of them had Spanish as their first language. All of them were enrolled in a reading and writing pre-intermediate English course. According to the Placement Test they took at the beginning of the year, they were elementary learners.

The study employed a combination of simple random sampling and random assignment methodologies. Students were assigned to three distinct groups: a control group (N=12), a direct metalinguistic group (N=23) and an indirect metalinguistic group (N=26). Within the direct metalinguistic group, 11 learners received feedback tailored to their preferences, while 13 students in the indirect metalinguistic group received similarly tailored feedback.

3.4 Instruments

3.4.1 Questionnaire

Data aimed at gathering students’ preferences for WCF were collected using an online questionnaire from Muñoz, Ortiz and Sáez ( 2023 ). A pilot application was used to identify any difficulties to understand or answer the instrument. To avoid any problem interpreting technical terms, some examples were included. Judgement validity was conducted through experts’ assessment of each of the items. A Cronbach’s alpha of .802 indicates the acceptable reliability of the questionnaire (for a detailed revision of the instrument, see Muñoz; Ortiz and Sáez, 2023 ).

3.4.2 Writing Tests and Tasks

Students answered a 250-word pre and posttest and completed three writing tasks. Both tests and writing tasks were based on the Preliminary English Test (PET) examination and required the use of the target structure. Participants had 30 minutes to compose their writing; any external resources were forbidden. The genre corresponded to descriptive texts as this is widely used to assess freshmen in the pedagogical context that was investigated. Besides, as Rummel and Bitchener state, “a communicative writing task would allow students to write quite naturally” (2015, p. 72).

3.4.3 Feedback Types

The types of feedback were determined once learners’ preferences for WCF strategies were identified (see 4.1), i.e., the treatments were designed according to the correction techniques that were mostly favoured; these were: direct correction with metalinguistic clarification and indirect correction with metalinguistic clarification. These were understood as follows:

1. Direct WCF with metalinguistic explanation (DML): both an indication of the error and the correct use of the form were provided, along with metalinguistic information about the nature of the error. An example of this is provided below:

writes*

My mom is an important writer. She write amazing novels.

* Remember that verbs in Simple Present in third person singular end in:

▪ -s > he wants // ▪ -es (verbs ending in: -ss, -x, -sh, -ch) > he passes. // ▪ -ies (verbs ending in -y) > fly > flies

2. Indirect WCF with metalinguistic explanation (IML): errors were localised and a question about the rules for the correct use of the target form was supplied, as in the following example:

*

My mom is an important writer. She write amazing novels.

* How do verbs in Simple Present in 3rd person singular end? What do you add to these verbs?

General comments on the composition were provided to participants in the control group. Some examples of these observations are: “Well done!! Very creative, keep going”.

3.4.4 Target Structure

This research adopted a highly focused correction, i.e., the treatment aimed to promote the correct use of one linguistic form. This methodological decision was made as students selected this kind of treatment as the most preferred type of correction (see 4.1). The form corresponds to the –s morpheme of verbs in the Simple Present due to the following reasons. On the one hand, it “is among the most frequently used structures in English” (Guo; Yang, 2018 , p. 846). On the other hand, this morpheme is signalled as complicated to acquire due to its low saliency in communication and its redundancy (Goldschneider; DeKeyser, 2005 ). Additionally, this structure is considered problematic among Chilean students regardless of their proficiency levels (Ortiz, 2016 ). This was confirmed when checking some of the compositions students previously wrote in the course.

3.4.5 Research Design

The study was conducted in nine sessions as summarised in Table 2 . During the first one, learners completed the questionnaire about their preferences regarding WCF strategies. Two weeks after this, the pen-and-paper pretest was administered. The following week, learners performed writing task 1. They had 30 minutes to write using approximately 250 words. The subsequent session, participants received feedback on the erroneous use of the –s morpheme according to their experimental conditions; other errors were left uncorrected. Participants had five minutes to check the feedback and understand the corrections. Then, following Zhang and Chen ( 2021 , p. 6), “their texts were collected, and they were given another 30 min to revise their writing while looking at the original uncorrected texts. That is, the students made revision without access to WCF”. This procedure was repeated on two more occasions. Finally, during session number nine, students were given the posttest.

The control group received general comments as previously stated and underwent the same processes as the experimental groups.

Table 2
. Research Design

Session	Activity
1	Questionnaire
2	Pretest
3	Writing Task 1
4	Feedback on Writing Task 1
	Rewriting of Writing Task 1
5	Writing Task 2
6	Feedback on Writing Task 2
	Rewriting of Writing Task 2
7	Writing Task 3
8	Feedback on Writing Task 3
	Rewriting of Writing Task 3
9	Posttest

3.4.6 Analyses

Descriptive statistics was used to analyse the questionnaire used to identify students’ preferences for WCF.

For the accuracy rates over time, Obligatory Occasion Analysis was used to count the tokens of the target morpheme and its obligatory occasions. As Ellis and Barkhuizen ( 2005 , p. 73) state, this method delves into “how accurately learners use specific linguistic (usually grammatical) features”. Thus, if a student accurately used the –s morpheme three times in 10 obligatory contexts, the percentage of accurate use would be 30%. The formula is the following:

$$\begin{equation} \text{Accuracy (\%)} = \left( \frac{\text{Total Obligatory Occasions}}{\text{Number of Correct Uses}} \right) \times 100 \end{equation}$$

To maintain the validity and reliability of the findings, inter-rater reliability was measured using Cohen’s kappa. Two independent reviewers corrected 30% of the texts independently. The result indicated substantial consistency among the two raters as it rated above 0.75 (κ= 0.895).

Once these percentages were determined, all scores were entered into SPSS 26 and descriptive and inferential statistics for the pretest and posttest were calculated per each group. Firstly, the distribution of the sample was calculated using Kolmogorov-Smirnov. The results showed that the data for all groups were distributed normally (Z= .200). Thus, parametric tests were selected for the analyses. Tests of statistical significance were carried out using one-way and two-way repeated measures ANOVAs. To assess the impact of tailoring the treatments to learners’ preferences for correction techniques, Fisher’s Exact Test for Count Data was conducted, following Rummel and Bitchener ( 2015 ).

4 Results and Discussion

4.1 The Questionnaire

Concerning the type of error students believe must be corrected, grammatical, vocabulary, and spelling deviations stand as the most chosen; the less preferred were organisation and punctuation flaws. These results mirror Amrhein and Nassaji’s ( 2010 ), who assert learners consider these types of inaccuracies as the most important to receive correction.

For the present research, grammar errors were adopted to receive corrective treatment because they were the most chosen. Besides, none of the students considered them as not important or somewhat important. This resembles Muñoz; Ortiz and Sáez’ ( 2023 ) findings, that suggest the level of competence of Chilean students influences their preferences towards grammar correction. These researchers consider that at a lower level of proficiency, students depend on grammatical instruction at a higher degree; while advanced apprentices do not surpass acquisition to grammar, giving further importance to more communicative experiences.

Figure 1
. Most important error to be treated with WCF

Regarding the type of strategies learners prefer, both direct correction with metalinguistic clarification and indirect correction with metalinguistic clarification were chosen. It is interesting to notice that, despite the huge variation in directness, these two strategies were highly valued among learners. This may indicate that students believe grammatical explanation is fundamental to learn or acquire an additional language. This idea is consistent with Schulz’ ( 2001 ) findings, who discovered Colombian students show a strong belief in the role of grammar in language learning. In the same way, in a study by Zhang et al . ( 2021 ), Thai tertiary students expressed that grammar is the most important aspect in writing. Sinha and Nassaji’s ( 2022 ) students also highlight the importance of WCF to promote grammatical accuracy. Similar results were found by Muñoz, Ortiz and Sáez ( 2023 ) when investigating similar learners in the Chilean context.

Figure 2
. Most preferred WCF strategies

Regarding the scope of WCF, most learners favour correction directed to one or two types of errors; i.e. they prefer focused WCF, which could represent a strategy to avoid cognitive overload when processing the provided feedback (see Figure 3 ). These results are at variance with Amrhein and Nassaji ( 2010 ), who found that learners prefer comprehensive feedback. It is possible that both the dissimilar investigated context and the diverse levels of participants’ competence in both studies might have influenced the findings. Amrhein and Nassaji conducted their research in an immersive setting where the target language was part of everyday activities, while the Chilean context circumscribes the L2 input almost exclusively to classrooms. On the other hand, Amrhein and Nassaji ( 2010 ) included a range of participants from beginners to advanced, but in order to answer the questionnaire, these individuals were required to be proficient enough in reading and writing. In our study, participants had a basic level of competence, and this characteristic supported the decision of using L1 in the questionnaire. Thus, it is possible to suggest that proficiency shapes learners’ preferences towards the scope that WCF should adopt. Thus, as Muñoz; Ortiz and Sáez ( 2023 ) suggest, participants with lower levels of proficiency favour focused feedback. Probably, as beginner learners tend to make more errors, focusing on one or two types of inaccuracies would facilitate attention and noticing, essential factors that promote learning (Schmidt, 2010).

Figure 3
. Preferred scope of WCF

Finally, regarding the most suitable source to receive feedback from, most of the Chilean participants prefer it comes from the teacher (see Figure 4 ). This depicts what other studies have found (Amrhein; Nassaji, 2010 ; Ghani; Asgher, 2012 ; Gan; An; Liu, 2021 ; Muñoz; Ortiz; Sáez, 2023 ). It may obey the belief that teachers are more skilful, so their corrections are more reliable, or as Hu ( 2019 ) states, they may be recognized as “sources of authority”. Notwithstanding, this conjecture needs to be further examined in the Chilean setting.

Figure 4
. Preferred source of WCF

4.2 The Tests

Table 3 presents the descriptive statistics for total scores for each of the groups over the two testing periods: pretest and posttest. Each group had considerably different accuracy rates in the pretest. The control group produced higher scores than both the direct metalinguistic group and the indirect metalinguistic group. The averages of accuracy rates in each group improved over time. At the posttest, the IML group that started with the lowest rate of accuracy outperformed the DML and the control group.

Table 3
. Descriptive Statistics for Mean Test Scores by Groups and Testing Period

Group	N	Pretest		Posttest
		Media	SD	Media	SD
Control	12	54.48	28.89	70.99	28.42
DML	23	40.03	31.87	86.73	16.25
IML	26	30.87	35.21	92.04	10.82
Total	61	38.97	33.45	85.89	18.83
DML: direct metalinguistic, IML: indirect metalinguistic

To determine statistically significant differences, a one-way ANOVA was run. It showed no variation among the groups in the pretest F (2,58)= 2.140, p= .127. This means the groups performed similarly at the beginning. However, differences were found in the posttest scores F (2,58)= 6,028, p= .004. To determine if these differences were statistically significant a two-way repeated measures ANOVA was used. The test scores were entered as the dependent variable, with time and the written CF types as independent variables. Each group performed differently on the total test scores, which indicates a significant effect for WCF, F (1,2)= 89.48, p= .000. A significant interaction between time and WCF treatment was also found – F (2,58)= 7.82, p= .001. One-way ANOVAs with the post-hoc comparisons (Bonferroni test) were then performed. As shown in Table 4 , the results revealed that both feedback groups showed statistically significant improvements over time (DML p-value= .037, IML p-value= .003). No differences were found between them despite the IML group showing the lowest rate at the beginning and the highest achievement in the posttest. The control group improved in its mean score in the second testing time, but not significantly (p-value= .875).

Table 4
. Multiple Comparisons Posttest

	Control	DM	ILM
Control	-----	0,037*	0,003*
DML	0,042*	-----	0,875
ILM	0,003*	0,875	-----
DML: direct metalinguistic, IML: indirect metalinguistic

This reveals the effectiveness of both WCF treatments. These results are in line with other investigations that show the positive effects of WCF strategies (Bitchener; Ferris, 2012 ; Shintani; Ellis; Suzuki, 2014 ; Ortiz, 2016 ; Muñoz, 2017 ; Muñoz; Sáez, 2019 , Sinha; Nassaji, 2022 ), and provide empirical evidence to the open-ended debate about the effectiveness of these two differing correction techniques (Sheen, 2011 ). Despite some scholars suggesting that direct correction is more effective, particularly with learners at early stages of acquisition (Ellis, 2009 ), others have advocated that indirect amendment is more prolific due to the cognitive processes this type of correction activates (Bitchener; Knoch, 2008 ). Nonetheless, the current research shows both kinds of techniques being equally significant as no statistical differences were determined (p <.05).

This outcome can be attributed to the combination of these polarized techniques with metalinguistic clarification. It could be suggested that the grammatical guidance assisted learners in promoting the accurate use of the treated form by modifying the degree of transparency or directness of the correction technique, making implicit correction more explicit. Thus, it is plausible to suggest that combining opposing correction techniques with metalinguistic explanation scaffolds acquisition as it provides grammatical support to emend errors.

Additionally, although some researchers have questioned the genuine effectiveness of metalinguistic strategies, especially among low-proficiency students (Hyland; Hyland, 2006 ; Beuningen; De Jong; Kuiken, 2012 ), these findings disclose that metalinguistic clarification is effective even with beginners, since it promotes the accurate use of the treated linguistic form, as other investigations have demonstrated (Muñoz; Ferreira, 2017 ).

We could explain these findings under the umbrella of learner beliefs as an individual factor. As Han ( 2017 , p. 2) states, “learner beliefs […] can also help explain variations among learners’ engagement with WCF and learning outcomes”. In the researched setting, learners rely on grammar instruction as a meaningful and essential element when learning a second language, particularly when it comes to low-proficiency students (Muñoz; Ortiz; Sáez, 2023 ). Therefore, when learners deeply rely upon grammatical instruction as a basic component of the learning process, they would tend to select WCF strategies that provide grammatical information about the targeted errors under the tenet that this has a positive effect on acquisition.

4.3 Tailoring the Treatment to Students’ Preferences

To answer this research question, “students’ beliefs about written CF and their performance after receiving written CF that either matched or didn’t match their beliefs” (Rummel; Bitchener, 2015 , p. 78) were investigated. Before the treatment session, students were asked about their preferences towards WCF, especially, about the type of errors, strategies, scope and source of feedback. In this research, 24 out of the 61 students received the type of feedback they preferred. To identify any correlation between learners’ preferences and the increase of the effectiveness of the WCF strategies, a mean of accuracy score for the treated morpheme was established. The results in the posttest had to be 90% or higher. This decision is backed on what has traditionally been suggested as the acquisition level (cf. Ellis; Barkhuizen, 2005 ). Thus, the use of the targeted morpheme was scored either 1 as acquired or 0 as not acquired. After this, Fisher’s exact test was run. This methodological resolution prevents Rummel and Bitchener’s ( 2015 ) methodological weaknesses. They state that seven out of the eight learners that were treated following their choices “were able to eliminate all their targeted errors” (Rummel; Bitchener, 2015 , p. 78-79); however, they do not declare what that “all” means, as a learner might have correctly used the studied form (simple past) only once. Besides, setting the level of accuracy as high as one hundred per cent is hard to be accomplished because even native speakers “fail to perform at 100 percent accuracy themselves” (Ellis; Barkhuizen, 2005 , p. 85).

The findings show that only 11 students who received WCF tailored to their preferences were able to reach a level of accuracy of 90% or higher. Table 5 presents this and shows that the IML group had more students who reached or outperformed 90% correct use, while only four students in the DML behaved similarly. Hence, as in Sinha and Nassaji ( 2022 ), no relationship between preferred WCF and improvements in accuracy was found.

Table 5
. Results of Tailoring WCF to Students’ Preferences

Group	WCF tailored to preferences	LoA <90%	LoA ≥90%	Total
DML	NO	4 (17,4%)	8 (34,8%)	12 (52,2%)
	YES	7 (30,4%)	4 (17,4%)	11 (47,8%)
IML	NO	5 (19,2%)	8 (30,8%)	13 (50%)
	Yes	6 (23,1%)	7 (26,9%)	13 (50%)

Contrary to Rummel and Bitchener’s ( 2015 ) results, Fisher’s test shows there are reasons to suggest tailoring the treatment to students’ preferences does not impact the effectiveness of the WCF treatment (p=.256). This may be due to diverse factors. One of them corresponds to the level of proficiency of the participants. According to Zheng and Yu ( 2018 ), lower levels of proficiency biases learners’ engagement with WCF. Rummel and Bitchener’s participants were advanced learners while the sample of the current research was composed of beginners. This is in line with Sinha and Nassaji ( 2022 ), who included intermediate learners and did not find any mediating factor of students’ preferences in the effectiveness of WCF. Thus, it could be declared that the level of competence in the language is correlated to favourable outcomes when tailoring WCF to learners’ preferences.

This individual factor may account for the nature of the selected strategies as well. Learners with lower competence in the L2 believe grammar instruction is a fundamental component of the learning process (Schulz, 2001 ; Zhang et al ., 2021 ; Muñoz; Ortiz; Sáez, 2023 ). For this reason, the less preferred strategies were those that do not provide grammatical explanation, namely: direct and indirect techniques, unlike Rummel and Bitchener ( 2015 ) who included more advanced learners that did incline for these types of strategies. Thus, learners’ preferences of the current study might have been based on the belief that techniques that offer grammatical clarification would be the most convenient, as Sinha and Nassaji ( 2022 ) affirmed. However, this may not necessarily be the most appropriate premise since more demanding WCF strategies such as indirect correction may be further beneficial to promote accuracy when using certain linguistic forms. Hence, it could be suggested that learners with a basic level of competence in the L2 are not yet prepared to make advantageous decisions regarding the most beneficial WCF strategies to favour their acquisition process. Undoubtedly, the existent evidence is still in its infancy, so extensive research is needed to grant more genuine conclusions.

Learners’ beliefs about who should correct errors might have also impacted these findings. Most of the participants chose the teacher as the most suitable source of feedback. It is possible to assume that learners believe it is teachers’ responsibility to correct flaws; so, they do not need to take an active role. Hence, this notion might have grounded the selection of the teacher as the most suitable source of feedback. This aligns with other studies that have identified apprentices prefer to receive correction from teachers as they represent a trustful and authoritative source of knowledge (Gan; An; Liu, 2021 ). Now, in order to identify the motives that support this preference, a more qualitative approximation is required.

Another concomitant factor that might have mediated these findings corresponds to the influence of contextual factors. Potential cultural differences between the studied settings may have had a role in learners’ choices. It is possible to think that instructional experiences have moulded learners’ predilections towards error correction. Chilean students are not used to taking part in methodological decisions such as the type of correction they want to receive. So, they could have based their decisions on correction techniques that are not necessarily the most effective to promote learning. This is in line with Muñoz ( 2017 ), who concluded that no significant improvements in the effectiveness of the correction techniques are observed when participants can choose the errors that receive WCF. In this respect, Mozgalina ( 2015 ) states that when learners have less opportunities to choose, they engage more in class activities. Undoubtedly, this is a key aspect in language teaching as considering learners’ preferences would not always promote academic attainment (Dembo; Howard 2007 ).

Lastly, it is significant to mention that the extent to which learners’ preferences were considered in this investigation might have also shaped the results. We tailored feedback not only to learners’ preferred correction techniques, but also to the type of error considered the most important to be corrected, the number of deviations that should receive treatment (scope), and the source from where the feedback should emerge. This approach differs from Rummel and Bitchener ( 2015 ) and Sinha and Nassaji ( 2022 ), who just considered learners’ preferred strategies. Perhaps, doing this, and imposing the other elements, would have had a greater impact on the effectiveness of the treatments; however, this is just a conjecture that needs to be further researched. Future investigation should explore these elements to supply empirical evidence.

5 Conclusions

This research investigated: 1) the effect of two WCF strategies on the improvement of the accurate use of a linguistic form, and 2) the impact of tailoring WCF treatment to learners’ preferences. With this purpose, the methodological flaws of a previous alike study were overcome. The findings reveal the techniques that were examined, viz . direct correction with metalinguistic clarification and indirect correction with metalinguistic clarification, are equally effective to foster the correct use of the targeted error (-s morpheme of verbs in Simple Present). This suggests that the metalinguistic component is a pivotal factor that enhances strategies despite their disparity in explicitness.

This investigation also provides valuable evidence about the effects of learners’ preferences on the effectiveness of WCF strategies. The data show that feedback techniques are not boosted when they are selected by apprentices; on the contrary, these strategies are more effective when they do not fit students’ beliefs and are imposed by teachers. This suggests learners’ preferences is an individual factor that would not favour corrective feedback treatment, as many scholars have hypothesized (Storch; Wigglesworth, 2010 ; Han, 2017 ; Gan; An; Liu, 2021 ). However, these findings should be interpreted with caution as the experimental approach used to study these variables is still nascent and requires further development. Apart from the current research, only Rummel and Bitchener (2014) and Sinha and Nassaji ( 2022 ) adopted a similar methodological design; therefore, it is mandatory to expand research considering individual and contextual factors, as they might have an incidence in the discrepancies that still exist in the field of WCF.

6 Limitations

Despite the contribution of this research, some limitations are identified. In the first place, it is difficult to generalize the results as the participants were part of the same major in just one university. Thus, other studies should cover different settings to identify the influence of diverse contextual experiences. Besides, learners with different proficiency levels should be included since it may be a mediating factor that would shape learners’ preferences about WCF. It is also necessary to enlarge the number of participants since they may react differently to error correction. Targeted errors should be diversified as they do not respond in a similar manner to correction. This would help reach more solid results. Finally, a more qualitative approximation is needed to identify latent learners’ beliefs about error correction. In this way, it could be possible to deeply explain what statistical data offer.

Authors’ Contribution

The authors confirm that they all take public responsibility for the work submitted for review. The authors confirm contribution to the article as follows. Belén Muñoz: conceived of the presented idea, designed and directed the project, developed the theory, verified the analytical methods, analysed and interpreted results, wrote the manuscript in consultation with Marisol Mora and Marcos Riveros. Marisol Mora: carried out the intervention, collected the data, performed descriptive analysis of the results, designed figures, contributed to the writing of the manuscript. Marcos Riveros: carried out the intervention, collected the data, performed descriptive analysis of the results, contributed to the writing of the manuscript.

References

• AMRHEIN, H.; NASSAJI, H. Written Corrective Feedback: What Do Students and Teachers Think is Right and Why? Canadian Journal of Applied Linguistics, v. 13, n. 2, p. 95–127, 2010.
• ATKINSON, D.; TARDY C. SLW at the Crossroads: Finding a Way in the Field. Journal of Second Language Writing, v. 42, p. 86–93, 2018.
• BEUNINGEN, C. van; DE JONG, N.; KUIKEN, F. Evidence on the Effectiveness of Comprehensive Error Correction in Second Language Writing. Language Learning, v. 62, n. 1, p. 1–41, 2012.
• BITCHENER, J.; FERRIS, D. Written Corrective Feedback in Second Language Acquisition and Writing. New York, NY: Routledge, 2012. 232p.
• BITCHENER, J.; KNOCH, U. The Value of Written Corrective Feedback for Migrant and International Students. Language Teaching Research, v. 12, n. 3, p. 409–431, 2008.
• BITCHENER, J. Why Some L2 Learners Fail to Benefit From Written Corrective Feedback. In: NASSAJI, K.; KARTCHAVA, E. (ed.). Corrective Feedback in Second Language Teaching and Learning: Research, Theory, Applications, Implications. New York, NY: Taylor & Francis, 2017. p. 129–140.
• CHEN, S.; NASSAJI, H.; LIU, Q. EFL Learners’ Perceptions and Preferences of Written Corrective Feedback: A Case Study of University Students From Mainland China. Asian-Pacific Journal of Second and Foreign Language Education, v. 1, n. 5, p. 1–17, 2016.
• DEMBO, M.; HOWARD, L. Advice About the Use of Learning Styles: A Major Myth in Education. Journal of College Reading and Learning, v. 37, n. 2, p. 101–109, 2007.
• DIAB, N. Effectiveness of Written Corrective Feedback: Does Type of Error and Type of Correction Matter? Assessing Writing, v. 24, p. 16–34, 2015.
• ELLIS, R. A Typology of Written Corrective Feedback Types. ELT Journal, v. 63, n. 2, p. 97–107, 2009.
• ELLIS, R. Epilogue: A Framework for Investigating Oral and Written Corrective Feedback. Studies in Second Language Acquisition, v. 32, n. 2, p. 335–349, 2010.
• ELLIS, R.; BARKHUIZEN, G. Analyzing Learner Language. Oxford: Oxford University Press, 2005. 416 p.
• ELWOOD, J.; BODE, J. Student Preferences vis-à-vis Teacher Feedback in University EFL Writing Classes in Japan. System, v. 42, n. 1, p. 333–343, 2014.
• FERRIS, D. Written Corrective Feedback in Second Language Acquisition and Writing Studies. Language Teaching, v. 45, n. 4, p. 446–459, 2012.
• FREAR, D.; CHIU, Y. The Effect of Focused and Unfocused Indirect Written Corrective Feedback on EFL Learners’ Accuracy in New Pieces of Writing. System, v. 53, p. 24–34, 2015.
• GAN, Z.; AN, Z.; LIU, F. Teacher Feedback Practices, Student Feedback Motivation, and Feedback Behavior: How Are They Associated With Learning Outcomes? Frontiers in Psychology, v. 12, p. 1–14, 2021.
• GHANI, M.; ASGHER, T. Effects of Teacher and Peer Feedback on Students’ Writing at Secondary Level. Journal of Educational Research, v. 15, n. 2, p. 84–98, 2012.
• GOLDSCHNEIDER, J.; DEKEYSER, R. Explaining the “Natural Order of L2 Morpheme Acquisition” in English: A Meta-Analysis of Multiple Determinants. Language Learning, v. 55, n. 1, p. 27–77, 2005.
• GUO, X.; YANG, Y. Effects of Corrective Feedback on EFL Learners’ Acquisition of Third-Person Singular Form and the Mediating Role of Cognitive Style. Journal of Psycholinguistic Research, v. 47, n. 4, p. 841–858, 2018.
• HAN, Y. Mediating and Being Mediated: Learner Beliefs and Learner Engagement With Written Corrective Feedback. System, v. 69, p. 133–142, 2017.
• HU, G. Culture and Peer Feedback. In: HYLAND, K.; HYLAND, F. (ed.). Feedback in Second Language Writing: Contexts and Issues. Cambridge: Cambridge University Press, 2019. p. 45–63.
• HYLAND, K.; HYLAND, F. Feedback on Second Language Students’ Writing. Language Teaching, v. 39, n. 2, p. 83–101, 2006.
• IRWIN, B. Written Corrective Feedback: Student Preferences and Teacher Feedback Practices. IAFOR Journal of Language Learning, v. 3, n. 2, p. 35–58, 2018.
• KARIM, K.; NASSAJI, H. ESL Students’ Perceptions of Written Corrective Feedback: What Type of Feedback Do They Prefer and Why? The European Journal of Applied Linguistics and TEFL, v. 4, n. 1, p. 5–25, 2015.
• KARIM, K.; NASSAJI, H. The Effects of Written Corrective Feedback. Instructed Second Language Acquisition, v. 3, n. 1, p. 28–52, 2019.
• LEE, I. Understanding Teachers’ Written Feedback Practices in Hong Kong Secondary Classrooms. Journal of Second Language Writing, v. 17, n. 2, p. 69–85, 2008.
• LEE, I. Utility of Focused/Comprehensive Written Corrective Feedback Research for Authentic L2 Writing Classrooms. Journal of Second Language Writing, v. 49, p. 1–7, 2020.
• LIU, Q.; WU, S. Same Goal, Different Beliefs: Students’ Preferences and Teachers’ Perceptions of Feedback on Second Language Writing. Journal of Writing Research, v. 11, n. 2, p. 299–330, 2019.
• MIAO, Y.; BADGER, R.; ZHEN, Y. A Comparative Study of Peer and Teacher Feedback in a Chinese EFL Writing Class. Journal of Second Language Writing, v. 15, p. 179-200, 2006.
• MOZGALINA, A. More or Less Choice? The Influence of Choice on Task Motivation and Task Engagement. System, v. 49, p. 120–132, 2015.
• MUÑOZ, B. Contribución del feedback correctivo escrito indi-recto en el aprendizaje del morfema –s de verbos en inglés en tercera persona singular, en estudiantes de enseñanza básica. Literatura y Lingüística, v. 35, n. 1, p. 275-296, 2017.
• MUÑOZ, B.; FERREIRA, A. El feedback correctivo escrito indirecto en el aprendizaje de la forma comparativa de adjetivos en inglés. Logos, v. 27, n. 1, p. 73-89, 2017.
• MUÑOZ, B.; ORTIZ, M.; SÁEZ, K. Preferencias y opiniones de estudiantes de un programa de Pedagogía en Inglés con distinto nivel de competencia lingüística acerca del tratamiento de los errores en la escritura en LE: estudio de caso en una universidad chilena. Literatura y Lingüística, v. 47, p. 279-306, 2023.
• MUÑOZ, B.; SÁEZ, K. El feedback correctivo escrito indirecto en el tratamiento de la concordancia sujeto-verbo en tercera persona singular entre estudiantes de Inglés como LE. Alpha, v. 49, p. 275-290, 2019.
• ORTIZ, M. Uso de la retroalimentación correctiva focalizada indirecta con claves metalingüísticas en la adquisición del sufijo -s en la tercera persona del singular en inglés, en estudiantes de un programa de formación pedagógica en EFL de una universidad chilena. Revista Folios, v. 44, p. 127–136, 2016.
• RUMMEL, S.; BITCHENER, J. The Effectiveness of Written Corrective Feedback and the Impact Lao Learners’ Beliefs Have on Uptake. Australian Review of Applied Linguistics, v. 38, n. 1, p. 64–82, 2015.
• SAELI, H.; CHENG, A. Student Writers’ Affective Engagement With Grammar-Centred Written Corrective Feedback: The Impact of (Mis)Aligned Practices and Perceptions. Canadian Journal of Applied Linguistics, v. 22, n. 2, p. 109–132, 2019.
• SCHMIDT, R. Attention, Awareness, and Individual Differences in Language Learning. In: CLS INTERNATIONAL CONFERENCE (CLaSIC), 4., 2010, Singapore. Proceedings […]. Singapore: National University of Singapore/ Centre for Language Studies, 2010. p. 721–737.
• SCHULZ, R. Cultural Differences in Student and Teacher Perceptions Concerning the Role of Grammar Instruction and Corrective Feedback: USA-Colombia. Modern Language Journal, v. 85, n. 2, p. 244–258, 2001.
• SHEEN, Y. Corrective Feedback, Individual Differences and Second Language Learning. New York, NY: Springer Verlag, 2011. 216 p.
• SHINTANI, N.; ELLIS, R.; SUZUKI, W. Effects of Written Feedback and Revision on Learners’ Accuracy in Using Two English Grammatical Structures. Language Learning, v. 64, n. 1, p. 103–131, 2014.
• SINHA, T.; NASSAJI, N. ESL Learners’ Perception and its Relationship With the Efficacy of Written Corrective Feedback. International Journal of Applied Linguistics, v. 32, n. 1, p. 41–56, 2022.
• STORCH, N.; WIGGLESWORTH, G. Learners Processing, Uptake, and Retention of Corrective Feedback on Writing: Case Studies. Studies in Second Language Acquisition, v. 32, n. 2, p. 303–334, 2010.
• SUZUKI, W.; NASSAJI, H.; SATO, K. The Effects of Feedback Explicitness and Type of Target Structure on Accuracy in Revision and New Pieces of Writing. System, v. 81, p. 135–145, 2019.
• VAN DER KLEIJ, F. Comparison of Teacher and Student Perceptions of Formative Assessment Feedback Practices and Association With Individual Student Characteristics. Teaching and Teacher Education, v. 85, n. 1, p. 175–189, 2019.
• ZHANG, L.; CHENG, X. Examining the Effects of Comprehensive Written Corrective Feedback on L2 EAP Students’ Linguistic Performance: A Mixed-Methods Study. Journal of English for Academic Purposes, v. 54, p. 1–15, 2021.
• ZHANG, T.; CHEN, X.; HU, J.; KETWAN, P. EFL Students’ Preferences for Written Corrective Feedback: Do Error Types, Language Proficiency, and Foreign Language Enjoyment Matter? Frontiers in Psychology, v. 12, p. 1–12, 2021.
• ZHANG, T. The Effect of Focused Versus Unfocused Written Corrective Feedback on the Development of University-Level Learners’ Explicit and Implicit Knowledge in an EFL Context. 2018. Thesis (Doctorate on Philosophy) – The University of Sydney, Sidney, 2018.
• ZHANG, Z.; HYLAND, K. Student Engagement With Teacher and Automated Feedback on L2 Writing. Assessing Writing, v. 36, p. 90–102, 2018.
• ZHENG, Y.; YU, S. Student Engagement With Teacher Written Corrective Feedback in EFL Writing: A Case Study of Chinese Lower-Proficiency Students. Assessing Writing, v. 37, p. 13–24, 2018.

• Funding
  This work was funded by the National Agency for Research and Development (ANID), Fondecyt Iniciación N°11230546.

Publication Dates

• Publication in this collection
  22 Nov 2024
• Date of issue
  2024

History

• Received
  14 May 2024
• Accepted
  26 May 2024