Manuscript ID : TTLT-2304-2316 (R1) Visit : 65 Page: 205 - 222

Article Type: Original Research

The Role of Participatory Structure and Output Modality in Developing Iranian EFL Learners’ Writing and their Perceptions: A Mixed-Method Study

Subject Areas : All areas of language and translation

Mahdieh Pourhadi ¹ , Nasrin Hadidi Tamjid ² , Saeideh Ahangari ³

1 - Department of English, Tabriz Branch, Islamic Azad University, Tabriz, Iran
2 - Department of English, Tabriz Branch, Islamic Azad University, Tabriz, Iran
3 - Department of English, Tabriz Branch, Islamic Azad University, Tabriz, Iran

Received: 2023-04-30 Accepted : 2023-06-15 Published : 2024-08-13

Keywords: Participatory Structure, Output Modality, Writing, Perceptions, EFL Learners&rsquo, , Mixed-Method,

Abstract :

Educational aids and new educational technologies have transformed learning systems during the last decade. The present study was an attempt to investigate the role of participatory structure and output modality in enhancing Iranian EFL learners’ writing, and examine their perceptions toward participating in writing activities. The population of this study included all female EFL learners (90 learners) at intermediate level at Avayeh Daneshvaran Language Academy in Tabriz. Among them, 63 learners were selected as the sample group after conducting a sample of the PET. The participants were selected through convenient sampling; they included four intact classes. The instruments included two samples of the English language proficiency test of PET, which were administered as homogeneity test, and the pre-tests and posttests of writing. The collected data were analyzed using one-way ANOVA and two-way ANCOVA. The researchers analyzed the participants’ responses to the open-ended questionnaire qualitatively. The findings revealed that output modality, participatory structure, and the interaction of these two had no significant effect on EFL learners’ writing skill. Regarding the qualitative section, the findings showed that the learners had positive attitude toward the use of podcasts and summary telling and writing. The findings of the study have significant theoretical and practical significance for teachers, teacher educators, curriculum designers and researchers.

References:

Full-Text:

The Role of Participatory Structure and Output Modality in Developing Iranian EFL Learners’ Writing and their Perceptions: A Mixed-Method Study

Mahdieh Pourhadi, Ph.D. Candidate, Department of English, Tabriz Branch, Islamic Azad University, Tabriz, Iran

mah_pourhadi@yahoo.com

Nasrin Hadidi Tamjid, Assistant Professor, Department of English, Tabriz Branch, Islamic Azad University, Tabriz, Iran

nhadidi@iaut.ac.ir

Saeideh Ahangari, Associate Professor, Department of English, Tabriz Branch, Islamic Azad University, Tabriz, Iran

saeideh.ahangari@gmail.com

2023-04-30 2023/06/15

Abstract

Educational aids and new educational technologies have transformed learning systems during the last decade. The present study was an attempt to investigate the role of participatory structure and output modality in enhancing Iranian EFL learners’ writing and examine their perceptions toward participating in writing activities. The population of this study included all female EFL learners (90 learners) at an intermediate level at Avayeh Daneshvaran Language Academy in Tabriz. Among them, 63 learners were selected as the sample group after conducting a sample of the PET. The participants were selected through convenient sampling; they included four intact classes. The instruments included two samples of the English language proficiency test of PET, which were administered as a homogeneity test, and the pre-tests and post-tests of writing. The collected data were analyzed using one-way ANOVA and two-way ANCOVA. The researchers analyzed the participants’ responses to the open-ended questionnaire qualitatively. The findings revealed that output modality, participatory structure, and the interaction of these two had no significant effect on EFL learners’ writing skills. Regarding the qualitative section, the findings showed that the learners had a positive attitude toward the use of podcasts and summary telling and writing. The findings of the study have significant theoretical and practical significance for teachers, teacher educators, curriculum designers and researchers.

Keywords: Participatory Structure, Output Modality, Writing, Perceptions, EFL Learners’, Mixed-Method

INTRODUCTION

English language as the largely used language worldwide and a requirement to the attainment of a great deal of knowledge has grown into a significant issue in the educational systems of countries across the globe. The English-speaking population is constantly increasing all over the globe, thus, as stated by Graddol (2006), the number of non-native fluent English speakers is more than the number of native English speakers. Recently, the process of language learning is not restricted to the course books and computer technology is implemented in this process to develop the value of education and learning. In Computer Assisted Language Learning (CALL), teachers are able to encounter the traditional methods of teaching and learning (Hasan & Hoon, 2013) via integrating the computer technology into the process of language teaching. Podcast is one of the tools of CALL, which was not initially intended for language learning, but it can be used for it. Podcasting can have enormous potential in developing learners’ writing and speaking skills (Ramli, 2018; Dianithi, 2017; Bamanger & Alhassan, 2015).

In fact, using modern materials like podcasts in language classes can provide language learners with authentic language input. That is to say, podcasts can act as comprehensible input and be a role model for the learners in helping them to produce authentic and accurate spoken and written texts. The output that learners produce can be written or oral, which can be produced individually or through participating with others. However , lack of ability to use foreign language productively, that is, in speaking and writing, may demotivate learners from language learning since in most of the cases the ultimate expectation from foreign language learners is the mastery of speaking and writing skills.

Much like other EFL contexts, English is taught as a compulsory subject at Iranian high schools. However, learning English language has permanently been an excessive challenge for Iranian learners owing to the restricted interaction with target-language speakers and lack of occasions to practice it in their everyday lives and interact with native speakers. That is why most of students prefer to improve their English in language institutes and academies. Here, the administration of modern tools like podcasts can have an encouraging role in providing EFL learners with comprehensive input.

The use of podcasts in EFL contexts by teachers and learners can resolve this deficiency (not having access to NSs) to some extents, and facilitate the process of language learning and gaining authentic input through listening to and practicing podcasts to produce comprehensible outputs through active participation or individual efforts. Furthermore, numerous EFL learners face troubles in speaking and writing throughout the class; they feel apprehension and become anxious when interacting with their teacher or even with their peers. In this regard, Jahin and Idrees (2012) mentioned that writing seems burdensome regardless of being native or non-native students since every author or writer is anticipated to exert stability amongst numerous features in his/her text for instance the content, audience, purpose, mechanics, vocabulary, organization, and the like.

Previous studies (e.g., Sayadi & Mashhadi Heidar, 2018; NamazianDost, Bohloulzadeh, & Rahmatollahi, 2017; Davoudi & Rezaei, 2016; Samad, Bustari, & Ahmad, 2017; Yoestara & Putri, 2018) on the role of podcasts especially in EFL contexts have concentrated on learners in a variety of academic and non-academic settings in Iran and the world over. However, there is not any comprehensive study that examines the role of participatory structure and output modality in developing learners’ writing skill in classes in which podcasts are used.

Generally, EFL teachers can use the findings of the present study in providing comprehensible input for the learners and guide them to produce acceptable outputs that can be in the form of written performances. Learners will also recognize the influence of acting individually or participating with the classmates in order to perform a linguistic or communicative task. Teachers can make use of technology and modern material in order to enhance learners’ learning and performance. As believed by Setiyadi (2020), to language learners, language learning is not just about obtaining a group of unconscious behaviors, however rather a process of realizing the fundamental rules, and administering them in their performances. In order for this realization to take place, the learners have to get through numerous phases and procedures. Equipping learners with acceptable and tangible input and instructing them to work with others in dealing with language tasks can lead to professional outputs. Moreover, knowing about the perceptions of EFL learners toward task types and task conditions can inform teachers and material developers to provide the best contexts and appropriate materials for the learners in EFL contexts.

All these points inspired the researcher to conduct the present study and investigate the problems related to the improvement of writing skill in Iran; the study can provide a better understanding of using podcasts in producing comprehensible outputs in language classes, individually or in pairs. Thus, in the current study, podcasts were employed to provide authentic input and see how learners’ productive skills of writing can be improved within different task conditions.

The following research questions were posed to meet the purpose of the study:

RQ1: Is there any significant effect of output modality on intermediate EFL learners’ writing performance?

RQ2: Is there any significant effect of participatory structure on intermediate EFL learners’ writing performance?

RQ3: Is there any significant interactional effect of participatory structure and output modality on intermediate EFL learners’ writing performance?

RQ4: What are the intermediate EFL learners’ perceptions toward the participatory structures in written tasks?

LITERATURE REVIEW

Task-Based Teaching

Trying to find the most operative methodology for SLA (Second Language Acquisition) has caused a rising body of empirical research and principles grounded on them, besides these have raised the application of numerous pedagogical means supported by scholars. The methods and means range from implicit ones, namely, the methods lacking any conscious training on the language structure, for instance input enhancement (Lightbown & Spada, 1990; Sharwood-Smith, 1993) and input flooding (Krashen, 1985), to explicit methods like metalinguistic explanatory feedback (Pica, et al. 1987; Gass & Mackey, 2007) and consciousness-raising procedures (Sharwood-Smith, 1993).

For years, teachers, researchers, testers, and syllabus designers have regarded the process of language teaching and learning as an area of interest. This attention has caused the emergence of diverse traditional methodologies. Following traditional styles, as argued by Skehan (2003), the experts assumed that it is not sufficient to emphasize on the language forms, however, teaching requires to be complemented by a focus to improve the skills to express meaning. In a task-based syllabus, the focus is primarily on meaning, not on linguistic form, which could promote functional and communicative foreign language development. Spade (1997) claimed task-based instruction, is the most beneficial to interlanguage development. Task-based instruction incorporates a task-induced focus on meaning and focus on form or an explicit, not necessarily task-induced, focus on forms. Although both approaches (task-induced focus on meaning and form, and explicit focus on form) have been shown to be effective, these approaches are more obtrusive, interrupting the flow of communication (Norris & Ortega, 2000).

Output Hypothesis

Swain (1985) developed the Output Hypothesis based in the Sociocultural or Social Interactional Theory, developed by Vygotsky. Accordingly, Swain pondered that the learning happens and to attain real learning the learners need to have the opportunity to test, produce, and to deliberate on different points. The Output Hypothesis incorporates the written and oral production (output).

Swain (1985) claimed that the Output Hypothesis, includes three functions that occur interchangeably: the hypothesis testing, the noticing function, and the metalinguistic function. Swain notes that during second language production, a learner could realize what he/she knows completely or partly. The learner could similarly check whatever he considers right way of writing or speaking in the foreign language. During this process, he could participate in a metalinguistic achievement. That is to say, Swain believed that in language production (writing/speaking) the learner can notice the gaps in his/her interlanguage, then in trying to create the foreign language, spontaneously he/she will check what he recognizes about it and think about it, which results in real learning.

Swain (1985, 1995, 2005) claimed that comprehensible output (i.e., second language production) has a significant position in second language acquisition as it confirms mental grammatical processing and is the most operative stimulus for the expansion of the learner’s interlanguage. The studies of Swain exposed that regardless of comprehensive contact with a target language input, the productive ability of immersion students is unlike that of native speakers. She maintained that it is merely throughout the production of the second language (in writing or speech), that the learners may notice that they cannot say what they try to say in the target language.

The Role of Participatory Structure in Productive Skills

According to Ellis (2004), Participatory Structure refers to “the procedures that govern how the teacher’s and students’ contributions to the performance of the task are organized, e.g., in terms of teacher-class or small group interactions” (p. 347). Actually, group work is a preferred participatory structure in all systems of teaching not just in task-based language teaching (TBLT).

Several supporters of TBLT (e.g., Prabhu, 1987) have excluded group work in their methodology. Prabhu (1987) contended that to have operative task-based teaching, learners are required to be extremely exposed to worthy models of the foreign language; Prabhu mentioned that this point necessitate the teacher to assume the responsibility of the mission. In fact, input-based tasks entail a teacher-class participatory arrangement; thus, the teacher will provide the input in these tasks and the learners will respond nonverbally to expose their understanding. In addition, the teacher interacting with the whole class can conduct the tasks related to speaking. For example, in information-gap tasks, the information can be divided between the teacher and the learners. The resulted interactions can provide special occasions for the teacher to remain much more than simply a supervisor and facilitator of a particular communicative task. The teacher is the main source of input (i.e., Prabhu’s ‘good models’) and is able to respond to any communicative and linguistic problems that happen.

The discoveries of recent research revealed positive outcomes of collaborative writing on evolving thinking abilities. It is revealed from preceding research equally in first language and EFL learners that shared production technique is a way to facilitate reflective thinking (Storch, 2002; Storch, 2005). Farrah (2011) stated that regarding critical thinking skills, it was established that the collaborative approach allowed several of the learners to create a richer bulk of content and challenge other learners to think more cautiously related to the subject at hand.

Writing Skill

It is assumed that EFL teachers consider writing as the most challenging language skill to explain. The justification is that writing is a productive skill that encompasses the organization of the author’s information into a written design (Hu, 2009). In an analytical study of the difficulties of writing skill that Saudi learners experience, Raja and Zahid (2013) specified that academic writing is regarded as the most problematic skill. EFL learners have trouble in English writing owing to the dissimilarity between English and their native languages (Hu, 2009).

According to Elashri (2013), for EFL learners writing is not an easy issue, particularly when the learners’ English proficiency is not developed perfectly. There are four viewpoints for writing skill teaching and learning: the process-focused approach, the product-focused approach, the genre-based approach to teaching writing, and process and genre based approach to teaching writing. The process approach centers more on using techniques such as rewriting, brainstorming, exploring ideas, and peer editing. The product approach is a traditional methodology to teaching writing in which the teachers classically provide the learners with a model and encourage to mimic it to produce a comparable product. A genre-based approach rests on the kind of the texts that the learners write. The most modern methodology is to associate process and the genre style.

Podcasts and their Applications in Foreign Language Classrooms

Fast evolution of concentration in MALL and the pervasiveness of high-speed Internet have caused the development and approval of podcasts (Hawke, 2010). Levy (2009) defined podcasting as, “an audio/video file that can be broadcast via the Internet with sound files that are ‘pushed’ to subscribers, often at regular intervals” (p. 775). Podcast is a groundbreaking advanced innovation that can be utilized viably for foreign language learning. As indicated by Robinson (2009, as cited in Namazian Dost, Bohloulzadeh, & Rahmatollahi, 2017) the term podcast is a mix of two words: Pod means iPod-the name of a well-known MP3 player and Broadcasting. As he considered, podcasts are video or sound records on the Internet that can be downloaded to a computer and listened to using any PC or any compact playback gadget that supports MP3 files.

As stated by Yoshida (2013), in most of EFL contexts that the learners have insufficient opportunities to use the foreign language outside the class, offering authentic and real occasions for them to practice productive skills can be a valued activity. Prior to the opening of the twenty-first century, the concept of CALL was supposed to have very partial uses in teaching productive skills to the EFL learners (James, 1996). However, from the beginning of the twenty first century to date, owing to a growth in the possession of cellphones and the range of the Internet, new means of dealing with this matter have been suggested. One of these new approaches is regarded to be podcasting. Certainly, numerous scholars believed that podcasting can have significant effects on the learners’ productive and receptive skills (e.g., Pun, 2006; Stanley, 2006; Nikolou & Darra, 2018; Bueno-Alastuey & Nemeth, 2020).

Learner Perception

Learner perception is the process that gives the foundation for learning, understanding, knowing and learning or motivating a specific action or response (Greenberg & Baron, 2008). According to Davis (1989), learners’ perceptions involve two cognitive beliefs: learners’ perception of the usefulness of an object (PU) that is the amount to which a technology develops one’s output or skill in a certain job and the perception of the ease of use (PEU) that is the capability to apply a technology with little or no trouble. As argued by Davis (1989) learners make decisions centered on their impression of how they distinguish each method of learning. Thus, learners’ perceptions related to online learning may cause learners to consider learning with definite outlook that may improve or weaken their strength to administer particular resources. On the other hand, learner autonomy necessitates that learners manage their learning process through adapting, re-adapting, and enhancing their learning behavior in various learning conditions. Autonomous learners are frequently motivated which results in operative learning consequences via the establishment of varied occasions for learning.

METHOD

Participants

The population of this study included all female EFL learners at intermediate level who were studying English at Avayeh Daneshvaran Language Academy in Tabriz. They were almost 90 EFL learners at intermediate level. Among the population of this study, 63 learners were selected as the sample group of the study after conducting a sample of the PET English proficiency test. The participants were selected through convenient sampling; they included four intact classes. Yet, to counteract the effect of selection bias, the groups were randomly assigned to four experimental groups, including two writing groups, that is, 16 learners in the individual summary writing group and 16 in the paired summary writing group and two speaking groups, that is, 16 learners in the individual summary telling group and 16 in the paired summary telling group. They ranged in age from 16 to 35 years old with different educational background.

Materials and Instruments

The instruments of the present study included two samples of the English language proficiency test of the Preliminary English Test (PET), which were administered as homogeneity test, and the pre-tests and posttests of writing. Moreover, a researcher-made questionnaire about the students’ perceptions was administered.

PET was a complete preliminary English proficiency test published by Cambridge University Press. Initially all parts of PET, including the reading listening, writing and speaking sections, were administered to ensure the homogeneity of the participants, and the writing section of this test were also used as the pre-tests of the study. As the posttests of the study, the writing section of another sample of the PET test was administered. However, in line with the delimitations of the study, only the writing section of anther sample of the PET was administered as the posttest of the present study. The responses were scored according to the scoring rubric given in the test. The PET scoring rubric for writing includes four subscales: Content, Communicative Achievement, Organization, and Language.

The final instrument of this study was an open-ended questionnaire with 5 questions developed by the researcher that examined learners’ perceptions toward participatory structure and output modality of the tasks in their language classes. The content validity of this questionnaire was ensured consulting a panel of experts.

The materials of the current study included the podcasts and the course book. The podcasts used in this study were selected from the English as a Second Language (ESL) Podcasts (http://www.podcastinginenglish.com). The coursebooks were Four Corners series and the teachers in all groups taught the same syllabus.

Procedure

Having received permission from the administrators in the institute, the researchers started the data collection procedure. In the present study, initially, through a one-way ANOVA test the homogeneity of the learners in terms of their language proficiency was examined. Based on the participants’ PET scores, those students who scored 1SD below or above the mean were considered as the main participants. Thus, out of 90 students, 63 students were selected as the participants of the study. The participants were in four intact groups of intermediate level, who were randomly assigned to four experimental groups. The participants were grouped in four classes based on task modality and task condition; in Group 1, the participants were asked to write the summary of the podcasts individually; in Group 2, the participants were asked to write the summary of the podcasts in pairs; in Group 3, the participants were asked to tell the summary of the podcasts individually; and in Group 4, the participants were asked to tell the summary of the podcasts in pairs.

Prior to the treatment, the writing scores of the PET test were used as the participants’ writing pre-tests scores. Then, the treatment was conducted which lasted for twelve 15-minute sessions. Each session, one podcast was played in all classes. The topics of the podcasts were similar to the ones they had in their course books.

After the 10-session treatment, the participants were asked to take another sample of PET test (only writing section) as their writing posttest. It should be mentioned that the writing pre-tests and posttests were scored by two raters and the inter-rater reliability were checked. For scoring the writing of the participants, the scoring rubric of PET test for writing assessment was provided by the researcher in order to be used by the raters. At the end of the treatment, the participants in each group were requested to answer an open-ended questionnaire developed by the researcher, which asks about their perceptions toward output modality and participatory structure of the tasks.

Data Analysis

To answer the research questions posed in the present study, the researchers collected the related data and conducted the following statistical analyses:

A one-way ANOVA to compare the writing scores of the participants on PET

A two-way ANCOVA to answer RQs 1, 2, and 3

To answer RQ 4, the researchers analyzed the participants’ responses to the open-ended questionnaire qualitatively.

RESULTS

The following section provides the results of the data analysis.

Testing the Normality of the Distributions

Initially, the normality of the writing data gathered in the pre-tests and post-tests was probed by computing the ratios of skewness and kurtosis indices over their standard errors (Table 1). For continuous data, the test of normality is important to decide what statistical methods should be used for the data analysis. When the data has normal distribution, parametric tests; otherwise, nonparametric methods are used to compare the groups.

Table 1

Descriptive Statistics; Testing Normality of Data

P-Structure	Modality		Skewness			Kurtosis
P-Structure	Modality		Statistic	Std. Error	Ratio	Statistic	Std. Error	Ratio
Individual	Writing	Homogeneity	.291	.580	0.50	-.730	1.121	-0.65
		Pre-Speaking	.504	.580	0.87	-.460	1.121	-0.41
		Post-Speaking	.514	.580	0.89	.131	1.121	0.12
		Pre-Writing	.264	.580	0.46	-1.001	1.121	-0.89
		Post-Writing	.301	.580	0.52	-.131	1.121	-0.12
Pair	Writing	Homogeneity	-.457	.564	-0.81	-1.108	1.091	-1.02
		Pre-Speaking	-.247	.564	-0.44	-1.691	1.091	-1.55
		Post-Speaking	-.436	.564	-0.77	-1.338	1.091	-1.23
		Pre-Writing	-.606	.580	-1.04	-.903	1.121	-0.81
		Post-Writing	-.614	.564	-1.09	-.503	1.091	-0.46
Note. P-Structure = Participatory Structure, Pre = Pretest, and Post = Posttest.

Since the absolute values of the ratios were lower than 1.96, it was concluded that the normality assumption was retained. It should be noted that the ratios of skewness and kurtosis over their standard errors are analogous to Z-scores which can be compared against critical values of +/- 1.96 at .05 levels (Field, 2018).

Exploring the First Three Research Questions (RQs 1, 2, & 3)

The first three research questions aimed at investigating the effect of participatory structure, output modality and their interaction on the writing performance of intermediate EFL learners. In particular, they were as follows:

RQ1: Is there any significant effect of output modality on intermediate EFL learners’ writing performance?

RQ2: Is there any significant effect of participatory structure on intermediate EFL learners’ writing performance?

RQ3: Is there any significant interactional effect of participatory structure and output modality on intermediate EFL learners’ writing performance?

However, initially, the researchers had to ensure the homogeneity of the participants in terms of their language proficiency. Therefore, based the participants’ PET scores, those students who scored 1SD below or above the mean were considered as the main participants. Thus, out of 90 students in four intact classes, 63 students were selected as the participants of the study, and the rest attended the classes but were excluded from the analyses (Table 2).

Table 2

Descriptive Statistics for PET Scores

Statistics
pet
N	Valid	90
	Missing	0
Mean		65.7278
Std. Deviation		11.29063
Skewness		.032
Std. Error of		.254
Skewness
Range		52.50
Minimum		41.00
Maximum		93.50

Additionally, the participants’ writing and speaking pretests and posttests were scored by a second rater and the inter-rater reliability was calculated. Table 3 displays the results of the Pearson correlations computed to estimate the inter-rater reliability indices of the two raters who rated the participants’ performance on the pretest and posttest of writing.

Table 3 Inter-Rater Reliability of Pretest and Posttest of Writing
		Pre-Rater 2	Post-Rater 2
Pre-Rater 1	Pearson Correlation	.987**
	Sig. (2-tailed)	.000
	N	63
Post-Rater 1	Pearson Correlation		.981**
	Sig. (2-tailed)		.000
	N		63
**. Correlation is significant at the 0.01 level (2-tailed).

Based on the results in Table 3, it can be concluded that there were significant agreements between the two raters on the pretest of writing r (61) = .987, representing a large effect size, p = .000, and the posttest of writing, r (61) = .981, representing a large effect size, p = .000.

To probe the first three research questions, a two-way ANCOVA was employed. Besides the assumption of normality, two-way ANCOVA has three more assumptions, that is, homogeneity of variances, linearity of relationship between covariate (pretest of writing) and dependent variable (posttest of writing) and homogeneity of regression slopes, which had to be met.

First, the analysis of two-way ANCOVA requires that the groups enjoy homogeneous variances on the posttest of writing performance. The result of the Levene’s Test of Homogeneity of Variance is displayed in Table 4.

Table 4

Levene's Test of Homogeneity of Variances; Regarding Writing Posttest

df1

df2

Sig.

2.991

.038

The results shown in Table 4, indicated that the assumption of homogeneity of variances was not retained on the posttest of writing, F (3, 58) = 2.99, p < .05. To compensate for the violation of this assumption, the results of two-way ANCOVA will be reported at .01 levels instead of .05 (Tabachnick & Fidell, 2014).

Second, a two-way ANCOVA assumes that the relationship between the posttest of writing performance (dependent variable) and its pretest (covariate) is linear. The results of the linearity test in Table 4.6 indicated that the assumption of linearity was met, F (1, 29) = 206.62, p < .05, η2 = .891 representing a large effect size¹.

Table 5 Test of Linearity of Relationship between Writing Posttest and Pretest
							Sum of Squares				df	Mean Square	F	Sig.
Post-Writing * Pre-Writing	Between Groups	(Combined)	1270.990		32	39.718		7.435		.000
		Linearity	1103.763		1	1103.763		206.62		.000
		Deviation from Linearity	167.227		31	5.394		1.010		.491
		Within Groups		154.915			29		5.342
		Total		1425.904			61
		Eta Squared		.891

Eta Squared was computed as Sum of Squares Between Groups / Sum of Squares Total; and should be interpreted using these criteria, .01 = Weak, .06 = Moderate, and .14 = Large (Field, 2018, p. 737).

Finally, a two-way ANCOVA assumes homogeneity of regression slopes (Table 6), that is, the relationships between the pretest and posttest should be roughly equal across the groups.

Table 6 Tests of Homogeneity of Regression Slopes; Regarding Writing Posttest
Source	Type III Sum of Squares	Df	Mean Square	F	Sig.	Partial Eta Squared
P-Structure	10.125	1	10.125	2.164	.147	.038
Modality	5.664	1	5.664	1.211	.276	.022
P-Structure * Pre-Writing	15.732	1	15.732	3.363	.072	.058
Modality * Pre-Writing	9.468	1	9.468	2.024	.161	.035
P-Structure * Modality * Pre-Writing	14.035	1	14.035	3.000	.089	.052
Error	257.321	55	4.679
Total	35936.188	62

The non-significant interaction between participatory structure, output modality and covariate (pretest), as shown in Table 6, indicated that the assumption of homogeneity of regression slopes was retained on the writing performance after controlling for the effect of the pretest, F (1, 55) = 3, p > .05, partial η2 = .052 representing a weak effect size. Since the assumptions of conducting two-way ANCOVA were not violated, it was legitimate to use this analysis to investigate the effects of modality and participatory structure on the writing performance of intermediate EFL learners.

RQ1: Is there any significant effect of output modality on the intermediate EFL learners’ writing performance?

To answer RQ1, the researchers conducted a two-way ANCOVA the results of which were used to answer RQ2 and RQ3, as well. The results of the tests of between subjects’ effect are represented in Table 7.

Table Tests of Between-Subjects Effects; Regarding Writing Posttest
Source	Type III Sum of Squares	df	Mean Square	F	Sig.	Partial Eta Squared
Pre-Writing	976.178	1	976.178	196.982	.000	.776
P-Structure	16.229	1	16.229	3.275	.076	.054
Modality	11.376	1	11.376	2.296	.135	.039
P-Structure * Modality	12.609	1	12.609	2.544	.116	.043
Error	282.473	57	4.956
Total	35936.188	62

These results and the descriptive statistics displayed in Tables 7 to 10 were used to probe the first three research questions. It should be noted that all results were reported at .01 levels because the assumption of homogeneity of variances was violated.

Table 8 displays the descriptive statistics for the effect of modality on the posttest of writing after controlling for the effect of the pretest.

Table 8 Descriptive Statistics for the Writing Posttest Considering Modality
Output Modality	Mean	Std. Error	95% Confidence Interval
			Lower Bound	Upper Bound
Summary Writing	23.147a	.408	22.329	23.964
Summary Telling	24.011a	.395	23.220	24.802
a. Covariates appearing in the model are evaluated at the following values: Pre-Writing = 14.88.

As illustrated in Table 8, summary writing (M = 23.14, SE = .408) and summary telling (M = 24.01, SE = .398) had roughly equal means on the writing posttest after controlling for the effect of the pretest.

The results of two-way ANCOVA in Table 8, representing a weak effect size, indicated that there was not any significant difference regarding modality on the posttest of writing after controlling the effect of the pretest, F (1, 57) = 2.29, p > .01, partial η2 = .039. Thus, the first null-hypothesis stating that there was not any significant effect of output modality on intermediate EFL learners’ writing performance was confirmed.

RQ2: Is there any significant effect of participatory structure on the intermediate EFL learners’ writing performance?

Table 9 displays the descriptive statistics for the effect of participatory structure (individual and pair work) on the posttest of writing after controlling for the effect of the pretest.

Table 9 Descriptive Statistics for the Writing Posttest Considering Participatory Structure
P-Structure	Mean	Std. Error	95% Confidence Interval
			Lower Bound	Upper Bound
Individual	23.060a	.403	22.253	23.866
Pair	24.098a	.403	23.291	24.905
a. Covariates appearing in the model are evaluated at the following values: Pre-Writing = 14.88.

As indicated in Table 9, the individual (M = 23.06, SE = .403) and pair work (M = 24.09, SE = .403) groups had roughly equal means on the posttest of writing performance after controlling for the effect of pretest.

The results of the two-way ANCOVA in Table 9, representing a weak effect size, indicated that there was not any significant difference between individual and pair work groups’ means on posttest of writing performance after controlling for the effect of the pretest, F (1, 57) = 3.27, p > .01, partial η2 = .054. Thus, the second null-hypothesis, stating that there was not any significant effect of participatory structure on the intermediate EFL learners’ writing performance was also confirmed.

RQ3: Is there any significant interactional effect of participatory structure and output modality on the intermediate EFL learners’ writing performance?

Table 10 displays the descriptive statistics for the interaction between participatory structure and output modality on the posttest of writing after controlling for the effect of pretest.

Table 10

Descriptive Statistics for the Effect of the Interaction between Modality and Participatory Structure on the Writing Posttest

P-Structure	Modality	Mean	Std. Error	95% Confidence Interval
P-Structure	Modality	Mean	Std. Error	Lower Bound	Upper Bound
Individual	Summary Writing	22.169a	.578	21.012	23.327
Individual	Summary Telling	23.950a	.557	22.834	25.067
Pair	Summary Writing	24.124a	.591	22.941	25.307
Pair	Summary Telling	24.072a	.558	22.954	25.189
a. Covariates appearing in the model are evaluated at the following values: Pre-Writing = 14.88.

As displayed in Table 10, the pair work groups in both summary telling and summary writing had higher means as compared to the individual groups. However, the results of two-way ANCOVA, representing a weak effect size, indicated that there was not any significant interaction effect between participatory structure and output modality on the posttest of writing after controlling for the effect of the pretest, F (1, 57) = 2.54, p > .01, partial η2 = .043.

Thus, the third null-hypothesis, stating that there was not any significant interactional effect of participatory structure and output modality on intermediate EFL learners’ writing performance was confirmed as well.

Qualitative Analyses

RQ4: What are the intermediate EFL learners’ perceptions toward the output modality and participatory structures in written summary tasks?

The RQ4 is related to the qualitative section of the study. In this regard, a semi-structured interview was conducted by the researcher to investigate the participants’ attitudes toward the use of podcasts in the class and the role of summary writing in developing their writing skill. The interview questionnaire was sent to all of the participants and they were asked to answer the questions; however, only 12 of them took part in this phase. The participants’ responses were read and categorized as illustrated in the following tables. The responses are provided under each question posed in the interview. Tables 11-15 represent the analyses related to writing summaries individually.

Table 11

What is the Role of Writing Summary Individually in Improving Writing Skill?

Share To

Article Url

The Role of Participatory Structure and Output Modality in Developing Iranian EFL Learners’ Writing and their Perceptions: A Mixed-Method Study