A Study of the Effect of Dictogloss as a Medium of Form-focused Instruction on Vocabulary versus Grammar Development of Iranian EFL Learners

Introduction

Deciding on how to teach grammar and vocabulary has always been the focus of an ongoing debate in the field of foreign language acquisition. In fact, considering other research areas, teaching grammar has gained considerable amount of empirical research in foreign language acquisition (Borg & Burns, 2008). Therefore, the importance of grammar as a prerequisite for achieving communicative competence is a proven fact. On the other hand, the pivotal role that vocabulary plays in foreign language teaching and learning has been constantly acknowledged in theoretical and empirical foreign language acquisition. In this regard, Hunt and Beglar (2005, p. 2) stated “The heart of language comprehension and use is lexicon”.

Therefore, the need to identify the best method of teaching grammatical structures and vocabulary seems to be the vital one, particularly in case of learners of English as a Foreign Language (EFL) in Iran, who find vocabulary and grammar learning tasks challenging. According to recent studies in the field, vocabulary instruction can be considered demanding since many teachers are barely sure about the best practice in vocabulary teaching and occasionally cannot decide where to begin to form an instructional focus on word learning (Berne & Blachowicz, 2008). Therefore, the principal focus of the present study is to examine the effectiveness of DG technique on vocabulary versus grammar learning of Iranian EFL learners at intermediate level. Dicotgloss (hereafter DG) is a focus on form technique that requires learners to utilize the four language skills and incorporates their knowledge of grammar and vocabulary to complete the task. Accordingly, Wajnryb (1990) argues that DG is an effective method to create conditions for learning grammar and vocabulary. DG is often mistaken for the traditional dictation task, but it is quite different from dictation regarding both procedure and task aims. In DG procedure, the teacher reads a short passage twice at normal speed to a class of learners. At first, the learners are asked to listen cautiously to get the story and the second time they can take some notes while listening. At the end of the dictation stage, the learners work in groups to rewrite their own version of the original text by getting use of their notes. Finally, the reconstructed texts are analyzed within groups by the help of teacher (Wajnryb, 1990).

The present study is an attempt to examine the effect of DG, as a focus on form task on vocabulary versus grammar learning of Iranian EFL learners. There have been some studies concerning the effect of DG on grammar learning, but to the best of the researchers’ knowledge, its effectiveness on vocabulary improvement has barely received any attention among researchers, particularly in the context of Iran. Moreover, the scarcity of research which compares the effectiveness of DG on grammar versus vocabulary was the main motivation behind conducting the current study. In addition, this study may provide English teachers with information about DG, which can motivate them to try out this technique or any of the other form-focused collaborative activities to teach language skills. The present study also provides additional data as it explores the effectiveness of the DG task compared to the traditional method of grammar and vocabulary teaching.

Review of the Related Literature

Dictogloss can be considered a relatively recent procedure in language teaching which is used in the field of focus on form instruction as one of the main techniques. Here is the summary of some main studies in the field.

Jacobs and Farrell (2003, as cited in Stockwell, 2010, p. 48) clearly summarized the advantages of DG as “When implemented conscientiously, DG embodies sound principles of language teaching which include: learner autonomy, cooperation among learners, curricular integration, focus on meaning, diversity, thinking skills, alternative assessment, and teachers as co-learners”. Furthermore, Shak (2006) defines DG as a type of form-focused task in which a meaning-based environment is provided for learners to focus their attention on the use of earmarked linguistic features while communicating with each other. Therefore, noticing and interaction, which are considered significant prerequisites for learning a second language, are in fact the theoretical assumptions underlying the DG technique. Moreover, some studies (Garrett, 1998; Potthast, 1999 & Ravenscroft, 1997) have found that collaborative learning activities have had positive effects on learners’ language acquisition. Murray (2001, as cited in Dewi, 2014) also argues that DG helps learners to make use of their grammar resources to rewrite a story which helps them to become aware of their  deficiencies and  their needs.

A deeper look at the research carried out in the area of language learning reveals that the effectiveness of DG as a language learning task has been empirically evaluated in promoting grammar as well as other language forms and skills.

In a study by  Nabei (1996), learners’ interaction during DG procedure was examined to see the way it can smooth the path of language learning. The findings suggest that DG facilitates discussion of both meaning and form, but there were only four subjects in the study. DG task in Nabei’s study took an overall look at all grammatical forms.

In a later study by Swain and Lapkin (2001), a DG task and a jigsaw task were compared as two communicative tasks, in terms of learners’ task performance. The results showed no statistically significant difference between the two tasks regarding the occurrence of lexis-based and form-based LREs (language related episodes). Furthermore, no difference was identified for the time it took for students to perform the tasks. In addition, the quality of reconstructed stories and learning outcome were the same.

Lim and Jacobs (2001) investigated the performance of second language learners in secondary school where learners helped each other in the process of learning during two by two verbal interactions while performing a DG task. The findings of the study provided evidence that peer collaboration during a DG task between students can get better regarding both cognition and affect. In fact, collaboration results in better learning atmosphere.

In another study, Kuiken and Vedder (2002) examined the effectiveness of learners’ interaction during a DG task on acquiring second language grammar. The analysis of the data showed that the interaction between learners during the reconstruction stage of DG did not lead to any improvement on posttest nor in a more frequent use of the grammatical structures in the written texts.

In a study carried out by García Mayo (2002), the effectiveness of two form-focused tasks (DG and text reconstruction) was analyzed. The findings indicated that in a DG procedure, advanced learners were more focused on the words’ meaning than about task features. In fact, DG could not really help learners to focus on target linguistic elements.

 In another study by Yeo (2002), DG was compared with an input enhancement technique. The results of the study indicated that the group in which DG was used by learners outperformed the other group where learners engaged in an input enhancement task to learn English participial adjectives. Therefore, Yeo suggested that DG as an output focused task is more fruitful than input focused tasks.

Leeser (2004) studied learners’ proficiency level and focus on form through the use of the DG and analyzed language related episodes (LREs). The findings suggested that as the proficiency level of the learners increased, the number of LREs which were produced also increased and these LREs became more grammatical in nature.

Fortune (2005) analyzed the use of meta-language by learners engaged in DG, comparing advanced L2 students with that of intermediate learners. Her study provided evidence that metalanguage can play a more facilitative role to the advanced learners in focusing attention and deciding about which form is more proper to use.

Hornby Uribe (2010) examined the effectiveness of three different approaches using the DG to teach grammar to high school Spanish II students. Based on the results, using DG in this study helped learners to notice and begin to learn the target forms.

A study carried out by Donesch-Jezo (2011) analyzed the effect of DG task on the acquisition of  single words and modal verbs which were used to express indeterminacy. The results revealed that the participants in the DG group did better than the other two groups.

Kim (2008) examined the effect of collaborative and individual tasks on the second language vocabulary acquisition. The findings of the study indicated that the lexical LREs (language related episodes) produced in the DG procedure result in a better acquisition of second language vocabulary. Therefore, it can be concluded that collaborative tasks are more effective on vocabulary acquisition than individual tasks.

HoKang (2009) in his study examined the effects of DG on both listening comprehension and grammar learning compared with the traditional focus-on-formS. The findings of the aforementioned study showed that DG instruction was more beneficial on both listening and grammar learning. However, it seemed that students’ proficiency levels indeed influenced the advantageous effects of DG.

Lapkin and Swain (2013) compared the DG with a jigsaw task. In this study, students worked in pairs to write a story by making use of a series of pictures. The researchers noted that the DG results in more exact reconstruction of the target forms than the jigsaw tasks.

Nguyễn (2014) assessed the effects of DG on the learners’ grammatical mastery and their motivation for learning grammar. The findings revealed that the use of DG in the grammar lessons improved not only their grammatical mastery but also their motivation and attitudes towards learning grammar.

In another study, Idek and Fong (2015) examined the effectiveness of DG in promoting students’ learning of irregular verbs. The result suggested that the students who were taught by the DG outperformed the students in the other group who were instructed by the conventional grammar tasks.

Another quasi-experimental research investigated the effectiveness of a DG exercise which had been revised to direct learners’ attention to particular phrasal expressions in a journal abstract (Lindstromberg, Eyckmans, & Connabeer, 2016). The result suggested that the modified DG which incorporates some more techniques to direct attention, shows a greater advantage over the standard DG task as a means of teaching phrasal words.

Snoder and Reynolds (2018) in a recent study investigated the effectiveness of DG on collocation learning of Swedish learners of English. The result showed that, the standard DG condition leads to better learning outcomes than the structured DG. Moreover, students completing the standard DG spoke and wrote more target features in their groups compared to the learners who were engaged in structured DG.

According to these studies, by using DG more opportunities can be given to students to build new vocabulary and learn about the grammatical rules of the text. DG is considered a multifunctional task which helps to improve students’ listening, writing and speaking skills. It also enhances grammar and discourse system to complete the task (Dewi, 2014). According to the issues discussed, three research questions are designed:

1. Does using DG task have any statistically significant effect on Iranian intermediate EFL learners’ vocabulary learning?

NH: Using dictogloss task does not have any statistically significant effect on Iranian intermediate EFL learners’ vocabulary learning.

2. Does using DG have any statistically significant effect on Iranian intermediate EFL learners’ grammar development?

NH: Using dictogloss does not have any statistically significant effect on Iranian intermediate EFL learners’ grammar development.

3. Is using DG task more effective on vocabulary learning or grammar development in the case of Iranian intermediate EFL learners?

NH: There is no statistically significant difference between EFL learners’ vocabulary and grammar development by using dictogloss task.

Method

Participants

As the population of the study, some 40 female learners of English as a foreign language were selected from two classes in Gofteman institute of Tabriz, which is a private institute for teaching English as foreign language at different levels. This institute was chosen as the researchers were also teachers there. All the participants were at intermediate level which was proven by the Oxford placement Test (OPT) administered before the selection.

Instruments and Materials

At the beginning of the study, Oxford Placement Test was administered in order to make sure of the homogeneity of the learners regarding their English level. Then, a pretest was administered in both classes before teaching the new grammatical structures and vocabulary words which were needed to be taught were identified using Oxford Word Skill for intermediate level (Gairns & Redman, 2008), Grammar in Use for intermediate level (Murphy, 2004) and Oxford Practice Grammar (Coe, Harrison, & Paterson, 2006). The same books were also used as sources for designing the pretest and posttest questions. The pretest and posttest included standard questions from entrance exams in Iran and the Oxford books.

In order to achieve the goal of the study, six stories were chosen from Steps to Understanding (Hill, L. A, 1981), used in the experimental group in which the DG task was used as a medium of instruction. Steps to understanding is a collection of short stories for English learners, in which the stories are classified into introductory, intermediate and advanced levels. The selected stories to be used in the present study were mainly from the intermediate part since, according to Read (1996), the stories need to be comprehensible for learners.

Procedure

After the learners took the pretest, one session was held to familiarize them with the DG task. In the following six sessions, the learners were engaged in a DG task, where the chosen stories were read to them by the teacher and then they were to reconstruct the story collaboratively, where they were divided into five groups, each group including four learners.

Then, the key vocabulary and grammatical structures were highlighted by the teacher, for example, to teach the new words the teacher made use of examples, synonyms and antonyms or writing them on the board and providing explanations regarding their meanings. The pre-teaching of new words and structures gives learners the chance to concentrate on the story itself, rather than looking for new words and losing track of the story.

After pre-teaching, the teacher read the story twice for the learners, during the first time the learners just listened but in the second reading they took some notes of the events in the story. Then they worked in group to reconstruct the story. The teacher’s role was a facilitator in this task. When the reconstruction stage was over, the learners were asked to exchange their written stories with other groups and read them for analyzing and correcting. On the other hand, in the control group the same grammatical structures and vocabulary items were taught using the traditional teaching method of present, practice and production.

Finally, a posttest was administered in both classes. The result obtained from the posttest was compared with the pretest’s results using SPSS in order to investigate the effectiveness of the task on learners’ vocabulary and grammar development.

Results

Examining the normality of the data is considered an important requirement before running many of the statistical analysis, since normal data are the underlying assumption in parametric testing. Therefore, this section will first focus on testing the normality of the data.

All the three null-hypotheses of the study are explored through assuming the normality of the data besides its specific assumptions. Table 1 displays the skewedness and kurtosis statistics and their ratios over the standard errors:

Table 1. Test of Normality

Since all the ratios were within -1.96 and +1.96, the data can be considered to be normally distributed enough to allow running one-way analysis of covariance.

Then, ANCOVA was run in order to examine the influence of DG task on vocabulary achievement of the learners, while putting aside the effect of the covariate factor. There are a number of assumptions associated with ANCOVA, which are needed to be considered before running the test.

Firstly, the covariates (i.e. pretest scores of the experimental and control groups) were measured prior to the treatment in order to avoid the scores on the covariate to be also influenced by the treatment, which reduces the likelihood of obtaining a significant result.

Secondly, Cronbach alpha, as the most common measure of internal consistency, is calculated to check the reliability of the scale. Cronbach’s alpha shows how closely related a set of items are as a group.

Table 2. Reliability Statistics of Vocabulary

According to Table 2, the Cronbach’s alpha for the two items is 0.704, which suggests that the items have relatively high internal consistency. Since values above 0.7 are considered acceptable; however, values above 0.8 are preferable.

Thirdly, some preliminary correlation analyses were run to explore the strength of the relationship among the covariates:

Table 3. Correlations between Control and Experimental Group Pretest Scores on Vocabulary

Since there was more than one covariate, Pearson Correlation was used to check the strength of the relationship. As they were not too strongly correlated (r<.8), the third assumption of ANCOVA is assumed.

Fourthly, the assumption of a linear relationship between the dependent variable and the covariate was checked by the scatter plots:

Figure 1. Scatter Plot of the Relationship between Control and Experimental Group on Vocabulary

Fifthly, the final assumption concerns the relationship between the covariate and the dependent variable, which was checked by the homogeneity of the regression slopes.

Table 4. Tests of Between-Subjects Effects of Vocabulary

Table 4 shows that the sig value or probability value is 0.756, safely above the cut-off point of 0.05. The sig value for the interaction is greater than 0.05 and the interaction is not statistically significant, indicating that the assumption is not violated.

After checking all the assumptions, a one-way between-groups analysis of covariance was conducted to compare the effect of DG on learners’ vocabulary learning. The independent variable was the use of DG task, and the dependent variable consisted of scores on the test after the intervention was completed. The participants’ scores on the pre-intervention administration were used as the covariate in the analysis.

Through analyzing the descriptive statistics of vocabulary provided in Table 5, the actual difference in the mean scores between the experimental in comparison with the control group is quite large. As Table 5 indicates, the mean score for the experimental (M=13.36, SD=3.83) was significantly different from that of the control group (M=10.89, SD=3.21). As a result, according to the mean scores, it can be stated that the experimental group outperformed the control group. However, other factors need to be considered before deciding on the final conclusion.

Table 5. Descriptive Statistics of Vocabulary

In order to assess the equality of variances for a variable calculated for the two groups, Levene’stest was used as an inferential statistic and the results are presented in Table 6. According to this table, the sig value is 0.43, which is much larger than the cut-off value 0.05. So, the assumption of equality of variance is assumed:

Table 6. Levene’s Test of Equality of Error Variances of Vocabulary

In addition, the following table of tests for between-subjects effect of vocabulary shows that there is a significant difference in the vocabulary learning through the experimental in comparison with the control group:

Table 7. Tests of Between-Subjects Effects of Vocabulary

Here the absence of any statistically significant differences between the mean scores of the control and experimental groups in learning English vocabulary were tested. According to the result, the sig value is 0.01, which is less than 0.05. Thus, there were statistically significant differences at (α ≤ 0.05) level between the experimental group and the control one in favor of the experimental one, and consequently the null hypothesis was rejected. Therefore, the experimental group outperformed the control group on the vocabulary learning test scores and DG proved to be effective as our independent variable.

In order to answer the second research question, the same assumptions associated with ANCOVA, which were discussed in the previous section, need to be considered again for the second research question.

Firstly, the covariates (i.e. pretest scores of the experimental and control groups) were measured prior to the treatment in order to avoid scores on the covariate also being influenced by the treatment, which as a result reduces the likelihood of obtaining a significant result. Secondly, Cronbach alpha was used to check internal consistency:

Table 8. Reliability Statistics of Grammar

According to Table 8, the two groups have a good internal consistency with a Cronbach alpha coefficient of 0.778. Values above 0.7 are considered acceptable; however, values above 0.8 are preferable.

Thirdly, some preliminary correlation analyses were run to explore the strength of the relationship among the covariates:

Table 9. Correlations between Control and Experimental Group Pretest Scores on Grammar

Since there was more than one covariate, the Pearson Correlation was used to check the strength of the relationship. As they were not too strongly correlated (r<.8), the third assumption of ANCOVA is assumed.

Fourthly, the assumption of a linear relationship between the dependent variable and the covariate was checked by the scatter plots:

Figure 2. Scatter Plot of Relationship between Control and Experimental Group on Grammar

Fifthly, the final assumption concerns the relationship between the covariate and the dependent variable which was checked by the homogeneity of the regression slopes:

Table 10. Tests of Between-Subjects Effects of Grammar

Table 10 shows that the sig value or probability value is .678, safely above the cut-off point of 0.05. The sig value for the interaction is greater than .05 and the interaction is not statistically significant, indicating that the assumption is not violated.

A one-way between-groups analysis of covariance was conducted to compare the effect of DG on intermediate language learners’ grammar. The independent variable was the type of interaction (experimental and control groups), and the dependent variable consisted of scores on the test after the intervention was completed. The participants’ scores on the pre-invention administration were used as the covariate in the analysis.

The actual difference in the mean scores between the experimental in comparison with the control group is quite small. As Table 11 indicates, the mean score for the experimental group (M=12.15, SD=3.4) was not significantly different from that of the control group (M=11.68, SD=2.64):

Table 11. Descriptive Statistics of Grammar

Table 12 shows a sig value of .94, which is much larger than the cut-off of .05. So, the assumption of equality of variances is assumed.

Table 12. Levene's Test of Equality of Error Variances of Grammar

Table 13 indicates that there is no statistically significant difference in the grammar development through the experimental group in comparison with the control group.

Table 13. Tests of Between-Subjects Effects of Grammar

According to Table 13, the sig value is .63, which is more than .05; therefore, the experimental group had no statistically significant improvement over the control group. Thus, the second null hypothesis is not rejected confirming that using DG task does not have any statistically significant effect on Iranian intermediate EFL learners’ grammar development.

To answer the third research question, Independent sample t-test was run to determine whether there is a significant difference between EFL learners’ vocabulary and grammar:

Table 14. Descriptive Statistics of Vocabulary and Grammar Level of EFL Learners

Table 14 indicates that the mean score for the vocabulary (M=12.13, SD= 3.7) is higher than that of grammar (M=11.92 SD=3.01). However, the actual difference in the mean score between the vocabulary scores in comparison with the grammar squares is not significantly large.

Table 15. Independent Samples Test of Vocabulary and Grammar

According to Table 15, the p-value (sig two-tailed) is 0.015, which is less than the cut-off point of 0.05. Therefore, there was a statistically significant difference at the p < .05 level in the test scores of the two groups and the third null hypothesis is rejected. As mentioned before, the mean score for the vocabulary (M=12.13, SD=3.7) is higher than that of grammar (M=11.92 SD=3.01). Thus, it can be concluded that DG task proves to be more effective on vocabulary achievement of the learners than the grammar.

Discussion

1. Does using DG task have any statistically significant effect on Iranian intermediate EFL learners’ vocabulary learning?

The findings of the first research question indicated that there was statistically significant difference at (α ≤ 0.05) level between the experimental group and the control one in favor of the experimental group. Therefore, using DG proves to have statistically significant effect on Iranian learners’ vocabulary learning. These findings could be rationalized considering the nature of DG task which provided organized steps of instruction with focusing on a learner-centered instruction.

The findings of this research question are in line with a study conducted by  Kim (2008) who examined the effectiveness of collaborative and individual tasks on the acquisition of vocabulary through comparing learners’ performance on a DG task. The results of his study indicated significant effect of collaborative work such as a DG task on vocabulary acquisition. In addition, as DG can be considered a context-based method of learning vocabulary, the findings of this research are also in agreement with Arıkan and Taraf (2010) who found out that contextualized methods can improve learners’ vocabulary knowledge in an EFL classroom.

Furthermore, the results gained from the current research regarding vocabulary learning strongly agree with the research conducted by Sadeghi and Safari (2012) who examined the effectiveness of collaborative task versus a direct method task on acquiring  English vocabulary. The findings suggest a significant difference between the experimental and control group regarding their vocabulary knowledge in favor of the group taught by a collaborative task. To sum up, the present research proved the effectiveness of DG task as a collaborative task on vocabulary learning of EFL intermediate learners.

A study conducted by MacKenzie (2011) also supports the findings of this study. In her study, MacKenzie evaluated the effectiveness of using DG task as a tool to increase students’ breadth and depth of knowledge of business vocabulary. The findings revealed that collaborative tasks provide learners with varied and multiple encounters with given words, and as a result, different lexical features are highlighted. Therefore, DG can be considered as an aid to develop learners’ vocabulary knowledge.

Moreover, Kusumalatif (2016) carried out a study investigating the effect of DG technique on improving learners’ vocabulary knowledge. In this study, some 40 learners were selected as participants and a quasi-experimental design was employed to achieve the goal of the research. The findings proved the DG task to be an effective method of teaching on learners’ vocabulary improvement, which is again in line with the results gained from the present study.

2. Does using DG have any statistically significant effect on learners’ grammar development?

According to the findings, there was no statistically significant difference between the mean scores of the experimental and control group in grammar posttest, which reveals that the DG task was not indeed effective on learners’ grammar development. These findings can be in line with a research carried out by Kuiken and Vedder (2002). They investigated the effectiveness of between-learner interaction while performing a DG task on grammar acquisition. The findings of this study revealed that the learners’ interaction during the DG task did not lead to any improvement on the posttest nor in a more frequent use of the intended grammatical forms in the reconstructed story. Moreover, the findings also revealed that interaction often stimulated noticing the grammatical forms.

However, the findings for this research question disagree with the one conducted by Idek and Fong (2015) who argued that the use of DG task proves to be effective on learners’ grammatical achievement. The reason for this can lie in the type of grammatical structure investigated in the research, which was past irregular tenses. In that research, learners were required to learn the past form of irregular verbs, and the use of verbs in past sentences was not emphasized. However, in the current research, five tenses were focused on (simple present, simple past, present perfect, past perfect, and future with will) and the practical use of each tense in English sentences were examined.

The results of this study also sharply disagree with the one conducted by Hornby Uribe (2010) who explored the effect of DG on grammar teaching in a meaningful context. The findings of his study showed a significant within-group difference between groups participating in the study. Therefore, the DG was effective on learners’ grammar learning. The reason for this disagreement can be the fact that DG has a different effect on learners’ performance with different levels of proficiency. In Hornby Uribe’s study participant were in the beginner level, but, the present study investigated DG’s effectiveness on intermediate level learners.

In another study conducted by Dewi (2017), the effectiveness of DG method on learners’ grammar improvement in second grade high school was examined. The results gained from Dewi’s study indicated that learners’ grammar knowledge was highly improved by using DG as a medium of instruction. These findings do not support the results obtained from the present study. The reason for this disagreement may again lie in learners’ different proficiency levels.

3. Is using DG task more effective on vocabulary learning or grammar development in the case of Iranian intermediate EFL learners?

Considering the third research question, the findings of the study indicate that there is a statistically significant difference between learners’ vocabulary learning and their grammar development when using DG as a medium of instruction. According to the analysis of t-test in the previous section, the experimental group did better on vocabulary posttest than in grammar posttest. As a result, it can be concluded that the DG had a higher effect on learners’ vocabulary learning than their grammar development.

The findings of this research actually agree with the study carried out by García Mayo (2002) who investigated the effectiveness of two form-focused tasks (DG and text-reconstruction). The findings of the study revealed that when learners are completing the DG task they seemed to pay more attention to word meanings and forms than targeted features by tasks. A possible explanation could be that when the learners make use of their resources such as their notes to reconstruct the text, they may tend to make use of their discourse/composition strategies. For example, students mostly made use of simple sentences and ignored the use of connector which was asked from them by the task.

The same explanation can be true for the current study. As the learners were required to make use of new words and grammatical structures at the same time, they tended to be more concerned with the new words, than grammatical structures. As a result, the grammatical structures which were highlighted by the teacher at the beginning of the process were neglected by most of the learners, since they just wanted to reconstruct the text even by using simple sentences. Therefore, the experimental group did better in vocabulary posttest than in grammar posttest.

Conclusion

The present research was an attempt to investigate the effectiveness of DG technique on Iranian EFL learners’ vocabulary versus grammar development. Through analyzing the data obtained from learners’ scores on pretest and posttest, the findings revealed that the experimental group outperformed the control group in learning vocabulary. Therefore, DG proved its effectiveness on developing the learners’ vocabulary skill. In addition, the DG task encouraged interaction and communication among learners in each group in the third stage (reconstruction) and among all groups in the last stage (analysis and error correction), which led to a learner-centered classroom. However, there was no statistically significant difference between the experimental and control group regarding grammar development. Therefore, it can be concluded that using DG task was more effective on vocabulary learning than grammar development.

English teachers could make use of DG in their classes as an alternative technique to teach new words, since DG provides a cooperative and acceptable competitive learning environment where learners work collaboratively in groups to reconstruct a text by using their recourses. This encourages communication in each group, therefore learners have the opportunity to use the new words in the context. As a result, the DG task is obviously beneficial for both teachers and learners. In addition, material developers and syllabus designers could make use of this task and include it in their English materials as a focus on form technique for learning new vocabulary.

Limitation of the Study

Although the findings of the present study have offered some implications for EFL research, there are some limitations in this study that are worth mentioning. One limitation concerns the size of the sample which was rather small (N= 40). Clearly this number of participants cannot be enough to make generalizations about all Iranian learners of English as a Foreign Language. In addition, the participants were only female learners of English. However, conducting the same research with male participants may lead to different results. Another limitation which may have affected the result of the study can be the teachers who also acted as researchers. As the same teachers taught for the two groups, their attitudes toward a condition in one class may have influenced learners' performance in the experimental group compared with the experimental groups in previous studies.

Suggestion for Further Research

The participants of the present study were only female English learners, so the same study can be conducted on the male participants in the context of Iran in order to investigate if it leads to any different results. Moreover, the current study was carried out during a period of six weeks and in seven sessions. Therefore, further research can be conducted in a longer period to achieve more accurate results. It is also recommended to conduct a study with classes taught by different teachers in order to examine whether the orientation of the teachers has any effect on the learners' performances.