| December 2009 home | PDF Full Journal |

Volume 11. Issue 4
Article 5

Title
The Impact of Gender on the Incidence and Quality of Form-focused Episodes in Task-based Conversational Feedback among EFL Learners

Authors
Parviz Birjandi
Islamic Azad University, Science and Research Campus, Iran

Omid Tabatabaei
Islamic Azad University, Najafabad Branch, Iran

Bio Data: 
Parviz Birjandi is a full professor holding an M.A in applied linguistics from the Colorado State University and a Ph.D in English education; minor: Research methods and statistics from the University of Colorado. He is currently the Dean of the College of Foreign Languages and Persian Literature in the Islamic Azad University, Science and Research Branch. He has published 16 articles in the area of TEFL and he is the author of English textbooks for high school and pre-university levels, used nationwide, 5 university textbooks and 4 practice textbooks.
 
Omid Tabatabaei is an assistant professor at the English department of Islamic Azad University, Najafabad Branch, Iran and currently the Head of the English Department in that university. His areas of interest are testing and assessment, research methodologies, psycholinguistics, language acquisition, and syllabus design. He has published and presented papers in international conferences and journals.

Abstract
Relatively few studies in second language acquisition and particularly foreign language learning have investigated the role of gender in task-based interactions and the provision of various types of feedback. The small number of studies such as Kasanga (1996), Oliver (2002), and Tannen (1990) have found potentially important differences between males’ and females’ interactions. The current study investigates whether the gender of EFL learners in conversational interactions influences the incidence and quality of one salient type of feedback, namely, form-focused episodes. Forty Iranian EFL students who were all English teaching majors completed the interactional tasks with both male and female interlocutors. Their language production in terms of the feedback exchanged was analyzed for the incidence and quality of form-focused episodes. The study investigated whether (1) the incidence and quality of form-focused episodes in feedback is different according to the group type (matched vs. mixed), (2) the feedback of males and females interactingin mixed-gender groups is different from each other in terms of form-focused episodes, and (3) the use of form-focused episodes in feedback is different according to the interlocutor’s gender. The results of a statistical analysis indicated that the gender composition of the groups influenced the occurrence of form-focused episodes, but this difference depended on the types of tasks the participants were engaged in.

Keywords: interaction, task, feedback, form-focused episodes

1. Introduction
A large body of research on language and gender (e.g., Oliver, 2002; Cameron, 2003a, 2003b) has demonstrated differences between the ways that males and females use languagewhen they interact. However, as Piller and Pavlenko (2001) have pointed out the role of gender in second language acquisition (SLA) “continues to be under-theorized and under-researched” (p. 1). Since many significant differences between male and female speech have been identified in conversational interactions between native speakers (e.g. Goodwin 1990; Tannen, 1990), one area of SLA, and specifically foreign language learning in which the impact of gender in terms of the use and quality of feedback being exchanged among learners might be particularly significant is in research conducted within the framework of the Interaction Hypothesis (Long, 1983, 1996). 
   The Interaction Hypothesis as put forward by Long (1996), suggests that engaging in conversational interaction facilitates SLA by providing learners with opportunities to receive comprehensible input and feedback (Gass, 1997, 2003; Long, 1996; Pica, 1994b, cited in Mitchell & Myles, 2004) as well as to make changes in their output (Swain, 1995, cited in Kaplan 2002). While it has been acknowledged that input or interactional modifications may differ across classes, genders, and cultures (Long, 1996), most research in second language acquisition and particularly foreign language learning has not considered the ways in which the gender of the participants might influence second and foreign language interactions. Although some researchers (e.g., Gass & Varonis, 1985a, 1986; Pica, et al., 1991) report the numbers of participants of each gender involved in their studies, few have considered the influence that the gender of learners, or the gender groupings of the participants, might have on their interactions. Since the Interaction Hypothesis assumes that conversational interaction is a site for second language learning, differences between males and females in these interactions, particularly the provision and quality of feedback, may influence language learning through interaction. One way to measure feedback is through the identification of instances when learners focus on form (Loewen, 2003, 2004) or attend to linguistic elements as they arise incidentally in lessons whose overriding focus is on meaning or communication (Long, 1991). Occasions when learners focus on form have been referred to as form-focused episodes in the literature (e.g., Basturkmen, Leowen, & Ellis, 2002; Loewen, 2003, 2004). The purpose of the current research is, therefore, to investigate the question of whether the use and incidence of form-focused episodes as one type of task-based conversational feedback is influenced by the gender of the participants. The following sections provide an overview of the Interaction Hypothesis focusing on form-focused episodes as one significant type of feedback and review some of the research related to language and gender which have been conducted within the area of second language acquisition.

2. Background
2.1. An Overview of the Interaction Hypothesis
Using a second language in a conversation with a native speaker or fluent non-native speaker has traditionally been viewed as a means to practice what has already been learned. The Interaction Hypothesis, which was initially given prominence by Wagner-Gough and Hatch (1975, cited in Kaplan 2002) and refined by Long (1983) and others (Gass & Veronis, 1985a; Mackey,1999; Pica & Doughty, 1985), claims that one route to second language learning is through conversational interaction, which provides learners with opportunities to receive target language input, to produce output, and, through interactional adjustments, to draw their attention to mismatches between their interlanguage and the target language.
   Through interaction, learners receive comprehensible input and feedback from their interlocutors, and are provided with opportunities to test target language hypotheses (Swain, 1995) as well as to ‘notice the gap’ between their interlanguage and the target language. This is where task-based language teaching can be considered as an effective way to strengthen classroom interaction. Ellis (2006) argues that classroom participants should forget where they are and why they are there and to act in the belief that they can learn the language indirectly through communicating in it rather than directly through studying it. It is probably easier to achieve when students are interacting among themselves, without the teacher being present, as the greater symmetry of social roles this affords leads naturally to the kinds of risk-taking behavior required of task-based pedagogy (Ellis, 2006). This is one significant reason why pair and group work are seen as central to task-based teaching.
   As Long (1996) explains, during interaction learners receive feedback on the form and meaning of their messages, modify their speech in an attempt to enhance the comprehensibility of their message, and push their interlocutors, whether native speakers (NS) or non-native speakers (NNS), to do the same. Gass and Mackey (2002) argue that receiving information on what is target like in the language at the exact time that the learner has produced a non-target like utterance allows the learner to connect form with meaning, an important step along the road to more target-like usage of a structure. This is true whether the information, or feedback, comes to the learner in the form of explicit correction or more implicitly, as a model of target like language, or negotiation for meaning. Mackey (2006) considers negotiation for meaning and the provision of recasts as two helpful interactional processes, because both can supply corrective feedback letting learners know that their utterances were problematic.

2.2. Feedback
Pica (1994b) points out that feedback provides learners with information about their language production, giving them the opportunity to modify their output, compare their utterances with a target like model, or explicitly discuss language form. In other words, feedback is the information that learners receive from their interlocutors about their language production. Feedback can be either positive, demonstrating comprehension of the learner’s language, or it can be negative, pointing out to the learner what was non-target like about his or her utterance. Feedback can be provided either implicitly, for example by demonstrating a lack of understanding of a learner’s utterance, or explicitly, by telling the learner what was non-target like about his or her utterance.
   Feedback can be examined in terms of form-focused episodes, recasts, and negotiation for meaning (Pica, 1994a). However, the focus of this study is on form-focused episodes.

2.2.1. Form-focused Episodes (FFEs)
One way to measure feedback is through the identification of instances when learners focus on form, and such occasions are referred to as form-focused episodes (FFEs) (Basturkmen, Leowen, & Ellis, 2002; Loewen, 2003, 2004). These units encompass discourse on matters of language form and meaning that negotiation and recasts cannot capture. FFEs differ from negotiation for meaning in that negotiation only implicitly focuses on form, while FFEs can be either implicit or explicit in their treatment of linguistic form. FFEs are different from recasts because a recast involves a restatement of the learner’s original non target-like utterance while an FFE, for example, could include an explanation of why the utterance was non target-like. 
FFEs have been found to occur frequently in classroom contexts (Ellis et al., 2001; Loewen, 2003). Williams (1999) argues that the incidence of FFEs can vary according to the learners’ proficiency level, the sort of activity they are engaged in, their cultural background, the learning context, and learner characteristics such as age and prior instruction. Gender is another learner characteristic that could influence the incidence of FFEs. The language discussed in an FFE is similar to a topic nominated for conversation, in that one learner raises an issue about the target language and the other learner has the option to either join in the discussion or move on with the task at hand. Research on gender and conversational interaction has demonstrated that males and females differ in the topics they discuss and how they treat them (Tannen, 1990).
Possible differences between males and females in the treatment of FFEs is especially important given that FFEs may lead to learning of the L2 forms learners discuss (Ellis et al., 2001). According to the findings of various investigations, it seems likely that several factors interact in facilitating SLA from FFEs. The present study examines the above-mentioned gender issues regarding FFEs with the hope of shedding some light on ultimate learning possibilities for males and females engaging in task-based interactions.

2.3. Interactional Style
While the focus of much of the research on interaction has been on interaction in the context of second language acquisition, it is important to remember that interaction refers to exchanges between native speakers of a language as well. Researchers have found gender differences when studying these everyday interactions. Tannen’s (1990) study of same-sex friendships, for example, found that across age groups, males and females acted differently when conversing with a same-sex friend. Males and females differed in the amount of talk they engaged in, with females generally talking more overall and discussing fewer topics than males, who discussed many topics briefly.

Differences between males and females have been found in mixed settings as well as matched-gender interactions. In one study of undergraduate students, Aries (1976) reported that males both initiated and received more interaction than females. In the interpretation of the findings, Aries mentioned that males spoke more than females, and individuals spoke more to males than to females; this latter finding raises the possibility that it is not just the gender of the speaker that influences interactional style, but also the gender of the interlocutor. She also pointed out that the mixed group setting seems to benefit men more than women by allowing men more variation in their interpersonal style.
   Interactional differences have also been found in giving and receiving compliments. Holmes (1998) examined interactions involving compliments and found that women both gave (68%) and received (74%) the vast majority of compliments. While men were in the minority on both ends, they were much more likely to give a compliment to a female (23%) than to a male (9%) and to receive a compliment from a female (17%) than a male (9%). Thus, even within a linguistic behavior as gender-differentiated as compliment giving, it is not just the gender of the person giving the compliment that matters, or the gender of the person receiving a compliment, it is the interaction of these two factors, namely, the gender of both the individual and the interlocutor.
   It is apparent from these studies that, depending on the context and the individuals involved, there may be differences in interactional style between males and females. These differences may be especially important if they affect other areas of life, such as education and achievement.

2.4. Gender and Second Language Interaction
Gender has been a subject of study in many fields closely related to second language acquisition, including other areas of linguistics, as well as psychology and education. A small number of studies (most of them carried out in a second language rather than foreign language context) have addressed the question of the role of gender in second language interaction (Aries, 1976; Gass & Varonis, 1986; Kasanga, 1996; Oliver, 2002; Pica et al., 1991). The results of the few studies that have been conducted on the impact of gender in SLA have pointed to possible gender differences in second language interactions. In interactions between learners, one study found that the most negotiation occurred in male-female dyads, followed by male-male dyads and then female-female dyads (Gass & Varonis, 1986), but another study found no significant differences between male-male and female-female dyads (Oliver, 2002). When looking at individual learner language production, studies have suggested that both males and females negotiate more in mixed-gender pairings than in matched-gender pairings, males indicate non-understanding with a greater frequency than females, and in mixed-gender pairings, males dominate in both the amount of talk and the performance of the task (Kasanga, 1996). In interactions between learners and native speakers, no significant differences were found for the incidence of negotiation in different types of dyads (Pica et al., 1991), but when looking at individuals, female NS were found to negotiate more with male learners than with female learners and female learners were found to negotiate more with female NS than with male NS. There were no significant differences for males, either learners or native speakers.
   While some studies (Gass & Varonis, 1986; Pica et al., 1989; Pica et al., 1991; Kasanga, 1996; Oliver, 2002) focused on the role of gender in conversational interactions, they were mostly concerned with only one type of feedback, namely, negotiation for meaning. Thus, the fields of SLA and especially foreign language learning are encountered with a serious shortage of studies related to another salient type of feedback, namely form-focused episodes.

3. Statement of the Problem
As discussed in the background section earlier, research conducted within the framework of the Interaction Hypothesis has indicated that conversational interaction can promote SLA (e.g., Ellis et al., 1994; Han, 2002; Iwashita, 2003; Leeman, 2003). Interaction may influence learning by providing learners with multiple opportunities: to receive input, produce output, and, through feedback on the comprehensibility and grammaticality of their own production to notice the difference between their interlanguage and the target language (Gass 2003; Pica, 1994b).
   Individual differences such as motivation, aptitude, working memory, anxiety, analytic ability, and the developmental levelamong learners may affect the ways that they interact and, possibly, the learning that results from such interaction (Mackey & Philip, 1998; Robinson, 2001). Gender can also be considered as another important factor which might influence the effectiveness of interactions.
   The research indicates that gender differences are apparent in L1 interaction beginning at an early age (e.g., Goodwin, 1990; Kyratzi & Guo, 2001). Males and females may have different educational experiences in the same classroom, use different learning strategies to different degrees, and their motivation and willingness to communicate may differ both in degree and by context. Finally, they may behave differently in mixed-gender situations than in matched-gender contexts. Studies of gender differences in second language interactions have further suggested that there are differences between males and females in the use of negotiation signals, or indicators of non-understanding (Pica et al., 1991; Kasanga, 1996). They have also indicated that it is possible that interactions in matched-gender pairs differ from interactions in mixed-gender pairs, both in the amount and type of interaction (Pica et al., 1991).
   However, despite the fact that the possible influence of gender on interaction has been called “a crucial issue” (Long, 1996, p. 421), there are relatively few studies of gender and interaction, particularly in EFL context and in terms of form-focused episodes, and these studies generally involve small numbers of participants (between 4 and 10 dyads). One very significant limitation of such studies is that they include participants with a limited set of L1s: it is unknown to what extent these findings apply to speakers of languages other than those spoken by the participants in the above studies. Although in these studies some learner characteristics such as proficiency level have been considered, most of the analyses appear to have treated learners as homogeneous and have overlooked individual differences and background variables. Furthermore, while these studies offered interesting findings with regard to gender, some were not designed to investigate gender differences in interaction. None of the studies have given participants the opportunity to interact in both mixed-and matched-gender groups, but rather have compared different individuals in each condition, raising the possibility that the findings, for example, of male dominance in mixed-gender groups, may be due more to the individuals in the groups and not to a specific gender-related characteristic.       
   Findings of gender differences in interaction and specifically form-focused episodes therefore remain speculative, especially in an EFL context. More concrete findings are necessary in order to gain a better understanding of the ways that gender might influence language learning through interaction. If males and females use different interactional strategies, or if their interactions are influenced by whether they interact with an interlocutor of the same or different gender, it is possible that these differences might lead to differences in language learning.
   The present study focuses on an analysis of the impact of one type of feedback, namely, form-focused episodes provided by males and females in mixed-and matched-gender groups in an EFL context (Persian academic community involving Iranian EFL students studying English teaching as their major at the B.A level) in an effort to determine to what extent and under what conditions learner gender affects the incidence and quality of form-focused episodes. It is important to note that this is the first time such a study has been conducted in Iran. Findings from this study will hopefully lay the foundation for future investigations of gender and foreign language development. The findings of this study can also be beneficial to researchers, helping them understand how gender may influence experimental findings and, ultimately, the language learning that results from interaction, and to teachers, who may wish to gain a better understanding of the dynamics of interaction in their classrooms and what implications these have for task design.

4. Research Questions
As mentioned previously, the current study aims to shed light on the impact of gender on task-based conversational interactions of foreign language learners by addressing the following major questions: Does the use and incidence of task-based conversational feedback vary according to the gender of the participants? If so, how?
These questions are further divided into three sub-questions:
1. Does the incidence and quality of FFEs in feedback differ according to group type
    (matched vs. mixed)?
2. Is the feedback of males and females in mixed-gender groups different in terms of    
    FFEs?
3. Is the use of FFEs in feedback different according to their interlocutors’ gender?

5. Research Hypotheses:
1. The incidence and quality of FFEs in feedback do not differ according to the group
    type (matched vs. mixed).
2. The feedback of males and females in mixed-gender groups is not different in terms
     of FFEs.
3. The use of FFEs in feedback is not different according to the interlocutors’ gender.

6. Method
6.1. Participants
Primarily, a group of 80 EFL students who were all Persian native speakers participated in this study. They were English Teaching majors taking laboratory classes as a four-credit course at the BA level. In fact, the researcher was the lab teacher as well, thus the participants were engaged in the completion of the assigned tasks as part of their class activities.  
   All participants took a pretest prior to the main phase of the experiment. This pretest consisted of the speaking and listening modules of IELTS taken from Cambridge IELTS 5 (Cambridge University Press). After the completion of this test, 40 (20 males and 20 females) of the whole population of 80 whose scores were approximately 5 (i.e., 4.5 to 6) were selected. The speaking scoring was done by three independent raters. The researcher interviewed the subjects and recorded all the exchanges. Then the raters listened to the interviews and scored them individually. The results were submitted to correlation analysis; the coefficient was 0.96. This was considered significantly high to determine that the rating was objective.

6.2. Materials
6.2.1. Treatment Tasks
Tasks have been defined by different researchers (Long, 1985; Richards, Platt and Webber, 1986; Breen, 1987; Ellis, 2003; Nunan, 2006, cited in Nunan, 2006) in various ways. Nunan (2006, p. 5) defines a task as:
"a piece of classroom work that involves learners in comprehending, manipulating, producing or interacting in the target language while their attention is focused on mobilizing their grammatical knowledge in order to express meaning, and in which the intention is to convey meaning rather than to manipulate form. The task should also have a sense of completeness, being able to stand alone as a communicative act in its own right with a beginning, a middle and an end."

Taguchi (2007) argues that a main objective in researching language tasks has been to identify a set of task characteristics based on the assumption that learner performance varies according to task characteristics.
   Under the interactionist framework, various aspects of tasks that are believed to influence interaction have been investigated. One of the early distinctions was made by Pica, Kanagy, and Falodun (1993), who classified tasks in terms of the source and flow of information. Tasks can be either one-way, with one participant holding all of the information to be conveyed to the other, or two-way, with each participant having part of the information that needs to be shared. Tasks can be further classified according to whether the exchange of information between participants is required, meaning that in order to complete the task, learners have to share information with each other, or optional, meaning that learners could choose to share information, but could also complete the task without doing so. Another distinction that can be made is between tasks with an open outcome, meaning that there is no predetermined solution or right answer and tasks with a closed outcome, in which participants are trying to find a specific solution (Ellis, 2003).
   Based on the findings of various studies mentioned in Ellis (2003), interaction is promoted by tasks that require a two-way exchange of information and a closed outcome. Accordingly, the tasks selected for this study are Picture Differences, Picture Placement, and Picture Story which are described in Table 1, and the characteristics of the tasks described are presented in Table 2.

Table 1:            Treatment Task-put task and description

Table 2:            Task Characteristics     

The Picture Differences task is known as an information gap task because each partner holds some of the information, and requires a two-way exchange of information; all members of the group must participate in the task. Information gap tasks, and in particular picture difference tasks, have been empirically demonstrated to provide opportunities for negotiation for meaning and other types of feedback (Mackey & Oliver, 2002).
   The Picture Placement task was adapted from Ellis (2001). Like the Picture Differences task, this task also requires an exchange of information, leading to participation by all members of the groups. However, it is possible to view this task as a combination of two one-way tasks, as each learner must share specific information about different items and their location with his or her partners. This task provides each participant with the opportunity to give directions, allowing comparisons with the findings of dominance by Gass and Varonis (1986) in direction giving.
   The Picture Story task provides an opportunity for learners to work together in a situation in which a two-way exchange of information is optional. In this task, based on one used by Swain and Lapkin (1998), optional completion of the task requires a discussion and resolution by all members of the group; however one partner could simply bypass the others and institute his or her own decisions. The Picture Story task also adds the element of writing, not required by the other two tasks. The addition of a writing component to this task, as well as the differing characteristics of the three tasks, may shed light on the different types of interaction elicited by different types of tasks, and allow an examination of whether different task characteristics affect the role of gender in task-based interactions.

6.2.2. Design
This study employed a repeated measures design, in which participants interacted in both mixed- and matched-gender groups. There were ten groups including four participants in each. This approach was chosen to ensure that results would reflect actual differences among interactions, rather than individual differences among participants other than gender. Participants, in groups of four, completed three tasks on each of two days. In order to ensure that results from all groups were comparable, and that any differences were due to gender rather than task-ordering effects, all groups completed the tasks in the same order, and the instructions for each version of the task remained constant. Gender groupings were counterbalanced: on the first day, half of the students participated in mixed-gender groups (i.e., groups consisting of two males and two females). The other half of the students participated in matched-gender groups (i.e., groups of four males or four females). On the second day, the students who had previously completed the tasks in mixed-gender groups completed the tasks in matched-gender groups and the students who had already completed the tasks in matched-gender groups worked in mixed-gender groups. Half of the total participants in each condition completed Task Set A on Day 1 and Task Set B on Day 2; half did the opposite. Task set A and Task set B each consist of a Picture Placement task, Picture Differences task and Picture Story task but which are different versions of each task. Since the same participants were supposed to participate in both matched- and mixed-gender groups on each of two days their performance might have been affected by their familiarity and experience of completing identical tasks, thus two versions of the same tasks were selected to avoid task wiseness. All interactions were audio-recorded for later transcription and coding.

6.2.3. Transcription and Coding Procedures
As mentioned earlier, participants’ speech was recorded on individual tape recorders during the task-based interactions; however, it was possible to hear all members of the group on each tape. Therefore, before transcription began, the researcher listened to all the tapes and chose the tape that had the best quality for all participants in each group interaction. All task-based interactions were then transcribed from these chosen recordings by the researcher or one of five research assistants. Following transcription from the better overall recordings, the researcher or one of two research assistants then checked the accuracy of each transcript by listening to the other participants’ recording and making any necessary modifications to the transcript. The researcher and research assistants always checked transcripts of interactions they had not originally transcribed. An inter-rater reliability check of the 45% of the random samples of identified FFEs from the transcribed data was subsequently done by two inter-raters. The percentage of agreement, found by counting the frequency of FFEs identified by the researcher and the inter-raters and then calculating the correlation coefficient, was 0.97. Interactions were then coded for the type and incidence of FFEs.
   A form-focused episode includes all interaction in which learners draw attention to form, that is, those that focus on form in the context of meaningful communication as well as those that are set apart from such communication and simply revolve around the questions of form itself (Williams, 1999). This may include instances of students asking for glosses of individual words or phrases or asking for specific feedback on form. FFEs were examined in an attempt to determine if gender differences exist in the frequency or type of focus on form.
   In order to code FFEs one should determine whether the FFE concentrates on lexis or form. An FFE is coded as being focused on lexis if learners are discussing the meaning of a word or which word to use, as in Example 1 below, in which female learner 1 is trying to remember the word for ‘blender’ so that she can ask her interlocutor where to place it in her kitchen.
        Example 1.   FFE: Lexis
      -Female learner 1: “How can I say that? Ay. I don’t remember, but I know the
name. It’s go whoo and you can make juice.”
      -Female learner 2: “Ah, blender?”
      -Female learner 1: “Blender! Yes, blender. Where do you have the blender?”

An FFE is coded as being focused on form if learners are discussing the syntax, morphology, phonology, or spelling of a word. FFEs that focus on syntax are those in which learners talk about word order or the suppliance or nonsuppliance of an element, as in Example 2 below, in which the learners discuss where they should say ‘two more minutes’ or ‘two minutes more’.

       Example 2.    FFE: Syntax
      -Male learner 1: “Sleeping for two …”
      -Female learner 1: “For two…”
      -Male learner 1: “more, two minutes. Two more minutes or two minutes more?”
      -Male learner 2: “Two more minutes.”
      -Male learner 1. “Two more minutes. All right.”

An FFE is coded as being focused on morphology if learners discuss issues of derivational morphology or of agreement (i.e., gender, number, tense), as in Example 3 below, in which the learners try to distinguish between the use of the past participle and the present progressive of ‘invite’.

      Example 3.    FFE: Morphology
     -Female learner 1: “The man…is invited her. He’s-…”
-Female learner 2: “No, no, inviting, I-N-G. No invite.”
-Female learner 1: “Past, no?”
-Female learner 2: “No past.”
-Female learner 1: “Okay, inviting.”

FFEs about phonology involve issues of pronunciation, as in Example 4 below, in which the female learner questions her pronunciation of ‘birds’.

     Example 4.     FFE: Phonology
-Female learner: “How many, how many birds, /birdz/?”
-Male learner: “Birds”
-Female learner: “…birds are flying?”

FFE about spelling involved learners discussing the spelling of words, as in Example 5 below, in which male learner 2 helps male learner 1 spell ‘each’.

Example 5.    FFE: Spelling
-Male learner 1: “Each other. Ho do you spell?”
-Male learner 2: “e-a-c.”
-Male learner 1: “t-?”
-Male learner 2: “c-h, other.”

Two inter-raters coded 35% of the transcribed material and the results were submitted to correlation analysis. The correlation coefficient was found to be 0.95.

7. Data Analysis
7.1. Incidence of FFEs
Research question one focuses on whether the use and quality of FFE differed in the three types of groups, M (all male), F (all female), and MF (mixed). The purpose of this question is to find out whether task-based feedback focusing on FFEs differs according to the gender composition of the group. To address this question, the FFEs produced by each group were considered, and the generated FFEs of the three types of groups were compared using repeated measures ANOVAs. Where significant results were obtained, post hoc Tukey tests were performed. To this end, the total incidence of FFEs was analyzed as a proportion of FFEs to total turns; thus the total number of FFEs for each group or individual was divided by the total number of turns taken by the group or individual in the task. In the case of individual language production, the participant who first questioned the language being used was credited as initiating the FFEs. Table 3 presents the descriptive statistics for the incidence of FFEs per turn by each group on each task.

Table 3:            Descriptive Statistics: Incidence of FFEs in Each Group

Figure 1 is a graphical representation of the mean proportion of turns containing FFEs on each task in each of the three group types.
Figure 1:           Form-focused episodes in each group type

According to Figure 1 and also the obtained means, the incidence of FFEs across group type in each task seems to be uniform, meaning that the groups of mixed and matched generated almost equal proportion of FFEs. Task seemed to have an effect on the incidence of FFEs: participants engaged in form-focused episodes most often on the Picture Story task, followed by the Picture Placement task and the Picture Differences task. The results of repeated measures ANOVAs (see Table 4) largely reflect this trend:

Table 4: Analysis of Variance: FFEs in Each Group Type

   Despite the differences on the Picture Story task, group type was not found to be significant (F=2.54, df=2, p=0.81), meaning that the incidence of FFEs did not vary according to whether the group was comprised of four males, four females, or two males and two females. Task was a significant factor (F=425.74, df=2, p=0.00), meaning that there were significantly different proportions of FFEs on different tasks. Post hoc Tukeys performed on the significant finding for task reveal that there were more FFEs on the picture story task than on the Picture Placement (p=0.00) or Picture Differences (p=0.00) tasks, and there were more FFEs on the Picture Placement task than on the Picture Differences task (p=0.00). The interaction between task and group type was not significant (F=1.31, df=4, p=0.26). Therefore, the first research hypothesis is confirmed: the incidence and quality of FFEs in feedback do not differ according to the group type (matched vs. mixed).
   Research question two asks whether there are differences between the interactions of males and females in MF group in terms of FFEs. The purpose of this question is to investigate whether the interactional experiences of males and females in mixed-gender groups are similar. To address this question, the interactions produced by each participant in male-female groups were considered, and the interactions of the males and females were compared using repeated measure ANOVAs and post hoc Tukey tests. Table 5 presents the descriptive statistics for the incidence of FFEs per turn by males and females in mixed-gender groups.
Table 5:

Descriptive Statistics: Incidence of FFEs by Participants in Mixed-Gender Groups

Figure 2 is a graphical representation of the mean proportion of turns containing FFEs on each task.

From the above descriptive data, there are two trends that emerge: males initiated more FFEs, with the exception of the Picture Differences task, than females. And participants of both genders initiated more FFEs on the Picture Placement task than on the Picture Differences task, and more on the Picture Story task than on either of the other two tasks. The results of repeated measures ANOVAs (see Table 6) confirm these trends.

Table 6: Analysis of Variance: FFEs by Participants in Mixed-Gender Groups

Despite the differences among males and females on the Picture Story and Picture Differences tasks, gender was not found to be significant (F=0.16, df=1, p=0.60), meaning that there was no significant difference among the incidence of FFEs initiated by males and females. Task was a significant factor (F=244.34, df=2, p=0.00); Post hoc Tukeys performed on the significant finding reveal that there were more FFEs on the Picture Story task than on the Picture Placement or Picture Differences tasks (p=0.00), and there were more FFEs on the Picture Placement task than on the Picture Differences task (p=0.00). The interaction between task and group was not significant (F=0.05, df=2, p=0.90). Thus, the second research hypothesis is also confirmed: the feedback of males and females in mixed-gender groups is not different in terms of FFEs.
   Research question three focuses on whether learners’ FFE feedback is different in mixed- and matched-gender groups. The current analysis directly compares learners’ feedback in both mixed-gender and matched-gender groups in an effort to explore whether learners’ use and quality of FFEs change in these two types of groupings. The FFEs produced by each participant in mixed-gender groups were therefore compared to the interactions produced by each participant in matched-gender groups using paired-samples t-tests. A different type of analysis was required for research question three, because while the first two questions investigated differences between groups, research question three examines within group differences.
   Table 7 presents the descriptive statistics and results of paired-samples t-tests for the incidence of FFEs by males and females in mixed-and matched-gender groups on each task.

Table 7:FFEs by Participants in Mixed-and Matched-Gender Groups

Figure 3 is a graphical representation of the mean proportion of turns containing FFEs on each task.

Figure 3:FFEs by participants in mixed-and matched-gender groups

Males and females generated more FFEs in matched- than mixed-gender groups, although this trend is highly under the influence of task type, meaning that males and females tend to generate more FFEs in the Picture Story task followed by the Picture Placement and then the Picture Differences tasks. Hence, the third null hypothesis is rejected: The use of FFEs in feedback is different according to their interlocutors’ gender.

8. Discussion
The previous section presented the analysis and results for each research question. This section will discuss those results with the goal of bringing together the findings from the various analyses in order to create an overall picture of the role of gender in foreign language learner interactions in the given context. The discussion will be centered on two themes: gender effects and task effects.

8.1. Gender Effects
Regarding the first research question, i.e., “does the incidence and quality of FFEs in feedback differ according to group type (matched vs. mixed)?”, gender did not seem to play a role in the incidence of FFEs. This finding is in direct contrast to earlier research on language and gender by Tannen (1990) who found that in matched-gender interactions between friends, females focused on form more than males did. However, Tannen’s research was on people with pre-existing friendships who were involved in the conversations, while the current research involved individuals who, while acquaintances and classmates, were not necessarily friends, and who were involved in classroom tasks. Kasanga (1996) has hypothesized that in the L2 “classroom setting, where the purpose of interaction is more than expressing one’s opinion or solving a problem per se, male and female participants, with a common aim of learning new target language items or structures, may differ less significantly in their patterns of language behaviors than outside the classroom” (p. 163). This may in part explain the contrast between Tannen’s research and the current finding.
   Concerning the second research question, i.e., “is the feedback of males and females in mixed-gender groups different in terms of FFEs?”, males and females in the current study did not differ significantly in the incidence of FFEs they initiated in mixed-gender groups. This is in contrast towork by Fishman (1978) in language and gender. He found that in first language interactions between men and women, topics initiated by women were often abandoned, while topics initiated by men rarely were. Fishman concluded that women do the work necessary for interaction to occur smoothly. But men control what will be produced as reality by the interaction. The fact that an FFE is by definition collaborative makes it comparable in some respects to a topic nominated for conversation. The difference between the findings of the current study and the previous one can be attributed to the fact that in Fishman’s study the participants were interacting in their L1 while in this study learners were performing the tasks in L2 and in an educational setting.
   Kyratzis's (2000) study of same-sex friendships found that across groups, males and females acted differently when conversing with a same-sex friend. Females discussed fewer topics than males, and males initiated more topics which were more abstract focusing on less personal issues than females. This is also in contrast with the findings of the current study, the reason can be due to the fact that Kyratzis's research was on people with pre-existing friendships who were involved in the conversations, while the current research involved individuals who, while acquaintances and classmates, were not necessarily friends, and who were involved in classroom tasks. Another reason can be attributed to the Iranian context. In Iran from the very early time of schooling, male and female students are separated so they do not have the experience of interacting with partners of the opposite sex.
   Gender influences on interactional style do not begin in adulthood; they emerge in early childhood and are reinforced throughout life. For example, children across cultures prefer same-sex play, often demonstrating this preference as early as two or three years of age (Sheldon, 1990). As children continue to interact with friends of the same gender, different kinds of social skills, especially competences in language use become entrenched (Ochs, 1993). These findings thus fit in with previous research on language and gender. The presence of the females in the groups have caused attempts by learners to question language use to be picked up and developed into FFEs.
   Regarding the third research question, i.e., “is the use of FFEs in feedback different according to their interlocutors’ gender?”, males and females generated more FFEs in matched- than mixed- gender groups. Here the concept of face threat may shed some light on such a finding. FFEs may be a fairly face threatening interactional move; when one learner focuses on form produced by another, he is not only telling the other learner that she is wrong, he is saying that he knows what is correct. Both males and females in this study may have been more willing to take this risk with an interlocutor of a different gender. This is also true regarding the Iranian context. As it has been highlighted before, concerning the religious issues, the educational system, even from the primary school, is segregated. Thus, it is not unexpected to witness participants prefer to provide their interlocutors of the same sex with more feedback than interlocutors of a different gender. As far as the personal experience of the researcher, teaching English as a foreign language for ten years at Iran universities, this has always been a significant and at the same time problematic issue encouraging students of opposite sex to be involved in interactions and class activities especially in  conversational classes
   Because FFEs have been empirically demonstrated to facilitate foreign language acquisition (Han, 2002; Iwashita, 2003), the fact that learners received significantly more FFEs from interlocutors of the same gender than from the interlocutors of a different gender may mean that learners in different group types have different opportunities to push their interlanguage towards the target. What is evident from this study is that the opportunities for language learning are different in the two types of groups. Learning from FFEs is thought to arise, at least in part, from the learner’s opportunity to compare his non-targetlike utterance to a targetlike model of that utterance. When learners in this study interacted in matched-gender groups, they may have had more opportunities to make this comparison and notice the gap between their interlanguage and the target than when they interacted in mixed-gender groups.
   The results discussed above demonstrate that gender is one of many factors that influence interaction. However, as shown in the findings of the current study, the effect of gender is not the same for all tasks. In all the areas in which gender differences were found, they were often found on one task but not another. This raises the issue of context including, among other factors, the task the learners are engaged in. This is not surprising, as research has found that “there are few, if any, context-independent gender differences in language” (Romaine, 2003, p.111) The next section will discuss the influence of task on the interactions among learners in the current study, for task did not influence only the effect of gender on interactions, but also consistently influenced the interactions of participants across group type.

8.2. Task Effects
Most FFEs were initiated on the Picture Story task, followed by the Picture Placement task and finally the Picture Differences task. This may be related to the learners' perceptions of the objectives of the activity, as suggested by Williams (1999). Participants were told that the goal of the Picture Differences task was to find ten differences between the pictures; because they did not have to write anything, accomplishing this goal may not have led to much focus on language itself. Rather, they only used language as a tool to complete the task. The goal of the Picture Placement task was for learners to make their pictures look the same; because they were asking each other what items were called and may have felt compelled to use the specific words provided, the incidence of FFEs was higher. Finally, on the Picture Story task, learners had to write out their language to create a copy. It seems that having to write the story that they would then turn in focused their attention on matters of language use, both lexis and form.
   Another reason for the initiation of more FFEs on the Picture Story task might be related to the fact that this task required learners to write out the story portrayed in the pictures, which may have led learners to view language as an object, not just a tool for communication, and thus focus their attention not only on meaning but also on form (Williams, 1999).
   From the examination of task effects on feedback in the current study, it is apparent that different tasks generate various amounts of FFEs. This is in line with previous research such as the study by Gass and Mackey (in press), finding strong effects of tasks on interactional features. In Gass and Mackey’s study, the researchers were investigating whether the interactional setting (classroom or laboratory) influenced the incidence of negotiation, recasts, and FFEs. They found that while the setting did not have an effect, task did. Participants engaged in the most negotiation for meaning on the picture differences and map tasks, provided their interlocutors with the most recasts on the picture differences task, and initiated the most FFEs on the picture differences and map tasks. While the tasks used in the two studies differ, the findings by both suggest that task is a strong factor in predicting the incidence of interactional features.

9. Implications
The results of this study suggest that the experience of engaging in task-based interactions with other learners may be a different experience for males and females, and that learners’ interactions may differ depending on whether their interlocutor is of the same or different gender. This is more noticeable in a foreign language context like Iran where from the very early years of schooling male and female learners, because of the religious beliefs, are separated.
   The gender composition of the group influenced the incidence of FFEs only on the Picture Story task, on which M groups engaged in significantly fewer FFEs than either F or MF groups. Whether participants interacted in mixed- or matched-gender groups did not significantly influence the incidence of FFEs initiated by males or females.
   Task influenced the incidence of FFEs, with groups engaging in significantly more FFEs on the Picture Story task than on either of the two other tasks. When males and females worked together, significantly more FFEs on the Picture Story task than on the Picture Placement task, and more on the Picture Placement task than the Picture Differences task were noticed.
   Teachers want to give their students the best possible opportunities to learn and use the target language. Given the findings that gender may influence learners’ experiences in task-based conversational interactions, teachers may wish to encourage their students to work with interlocutors of a particular gender, in the case of FFEs, matched-gender groups. Teachers should consider planning for times when learners can work together in matched-gender groups in order to allow them the best possible context for the learning opportunities that arise from FFEs.
   Teachers should also be aware that the influence of gender on task-based interactions might vary with the task the learners are engaged in, with different tasks mitigating the effect of gender differently. Tasks and task types should be carefully chosen to elicit the kinds of interactions desired. Given the finding that some tasks were more likely to uncover gender differences than others, it is important to use a variety of tasks and task types when investigating the role of gender in second or foreign language interaction. Many of these implications are relevant for language teachers as well as researchers.

10. Limitations
When considering the implications of the study, discussed above, there are several key limitations that researchers and teachers should be aware of. The study was conducted with intermediate-level learners of English as a foreign language and Persian as L1. It is possible that differences between the current study population and other populations, with regard to first language, educational background, learning environment, and other variables, may affect the influence of gender on feedback.
   Another notable limitation is related to the way participants were selected. As it is difficult to access the real IELTS test, the pretest was selected from an IELTS course books, i.e., Cambridge IELTS 5. Thus the obtained results might not have been a true reflection of the proficiency level of the participants. Furthermore, the score of participants based on this course book test was on average 5, but more specifically 4.5 to 6. This fairly broad range of scores was thus likely to have influenced the nature of the feedback in interaction of the participants in the various groups.
   Related to the idea of gender varying according to context is the fact that, in both the current study and in previous research (e.g., Kasanga, 1996; Newton, 1991, as cited in Ellis, 2003), interactions have been demonstrated to vary widely according to tasks and task characteristics. Because the current study demonstrated that different tasks influence the effect of gender in different ways, it is important to keep in mind that gender effects apparent in tasks used in this study may not influence interactions to the same degree on the other tasks or task types.
   This study also described interactions between L2 learners, rather than between learners and native speakers. The results of this study should not be extended to native speaker-learner interactions without additional research.
   The influence of gender on L2 interactions might also be different in an authentic classroom. While the data for the current study were collected from intact classes, the curriculum was essentially put on hold for the purposes of data collection. The research tasks were not embedded in the curriculum nor were they administered by the learners’ regular teachers, although the teachers were present. The dynamics of learners in a teacher-fronted classroom completing curriculum-based tasks might be different from the current context, and therefore interactions might be different, and influenced differently by learner gender.

11. Suggestions for Future Research
While learners clearly were involved in more FFEs when they worked in matched-gender groups, perceptions of their interlocutors’ linguistic knowledge may come into play in determining whether actual learning results from the FFE. The fact that a learner concedes to her interlocutor’s solution to a problem does not necessarily imply that the learner accepts the solution as worthy of incorporation into his developing interlanguage. It may be that FFEs with learners of one gender may be more memorable than with learners of another gender. Qualitative research is needed to explore some of these issues.
   The focus of this study is merely on the incidence and quality of FFEs. For future research, investigations can be fulfilled concentrating on whether gender differences exist in the frequency or type of focus on form and also attempts can be made to find out whether FFEs are resolved or not, and in the case of resolution, whether the produced utterance is target like or not. Furthermore, this study is only concerned with one type of feedback, FFEs, the impact of gender can also be investigated in the light of other sorts of feedback such as negotiation for meaning and recasts.
   Additional research on the role of learner characteristics and their relationship to learner gender and L2 interaction is crucial. Among the most important characteristics are ethnicity, L1, and interaction (Reid, et al., 1995); future research should replicate the current study with learners who differ from the current study population in these characteristics. It would be interesting to investigate how these characteristics interact with each other by conducting a study with learners with differing L1s and educational backgrounds, as well as ages and proficiency levels.
   Of course it is not only learners’ gender that may influence foreign language interactions; it is possible that the gender of the teacher may play a role. Given the role of the teacher as a primary source of language input for students, research on the role of teacher’s gender in foreign language interactions is necessary.

12. Conclusion
The results of the statistical analysis indicated that the gender composition of the groups influenced the occurrence of form-focused episodes, but this difference depends highly on the types of tasks the participants are engaged in. When males and females worked together, significantly more FFEs on the Picture Story task than on the Picture Placement task, and more on the Picture Placement task than the Picture Differences task were noticed. The gender of interlocutors seems to have a significant impact. Males and females tend to generate more FFEs in matched-gender groups, although this trend is highly under the influence of task type.
   To the best knowledge of the researcher, this study is the first systematic investigation of gender and interaction that compared male and female interactional patterns in different group types, in male-female interactions, and as learners interact in mixed- and matched-gender groups. As a result, despite acknowledged limitations, it provides the first opportunity to make claims about the role of gender in foreign language interactions. This was also the first study of gender and interaction to examine form-focused episodes in an Iranian EFL context. This study has attempted to lay the groundwork for greater research into the role of gender in task-based interactions among foreign language learners, including the role of the task the learners are engaged in, variation among individual learners, and developmental investigations of gender and foreign language learning through interaction. Most importantly, the current study shows that the gender of both the learner and the interlocutor may influence the incidence and quality of FFEs in task-based interactions among foreign language learners.

Acknowledgement
The authors would like to thank Jeremy Cross for his inestimable help, constructive views, and comments throughout the process of modifying and revising this paper.

References
Aries, E. J. (1976). Interaction patterns and themes of male, female, and mixed groups. Small Group Behavior, 7, 7-18.

Basturkmen, H., Leowen, S., & Ellis, R. (2002). Metalanguage in focus on form in the communicative classroom. Language Awareness, 11, 1-13.

Brown, J. D. (1988). Understanding research in second language learning. A teachers’ guide to statistics and research design. Cambridge: Cambridge University Press.

Cameron, D. (2003a). Gender and language ideologies. In J. Holmes & M. Meyerhoff (Eds.), Handbook of language and gender (pp. 447-467). Oxford: Blackwell.

Cameron, D. (2003b). Gender issues in language change. Annual Review of Applied Linguistics, 23, 187-201.

Ellis, R. (2001). Non-reciprocal tasks, comprehension, and second language acquisition. In M. Bygate, P. Skehan, & M. Swain (Eds.), Researching pedagogic tasks: Second language learning, teaching, and testing (pp. 49-74). Harlow, England: Pearson.

Ellis, R. (2003). Task-based language learning and teaching. Oxford: Oxford University Press.

Ellis, R. (2006). The methodology of task-based teaching. Asian EFL Journal, 8(3), 19-45.

Ellis, R., Basturkmen, H., & Leowen, S. (2001). Learner uptake in communicative ESL lessons. Language Learning, 51, 281-318.

Ellis, R, Tanaka, Y., & Yamazaki, A. (1994). Classroom interaction, comprehension and the acquisition of L2 word meanings. Language Learning, 44, 499-491.

Fishman, P. M. (1978). Interaction: The work women do. Social Problems, 25, 397-406.

Gass, S. M. (1997). Input, interaction, and the second language learner. Mahwah, NJ: Lawrence Erlbaum Associates.

Gass, S. M. (2003). Input and interaction. In C. Doughty & M. H. Long (Eds.), The handbook of second language acquisition (pp. 224-255). Oxford: Blackwell.    

Gass, S. M. & Mackey, A. (in press). Task-based interactions in classroom and laboratory settings. Language Learning.

Gass, S. M. & Mackey, A. (2002). Frequency effects and second language acquisition. Studies in Second Language Acquisition, 24, 249-260.

Gass, S. M. & Varonis, E. (1985a). Task variation and nonnative/nonnative negotiation of meaning. In S. M. Gass & C. Madden (Eds.), Input in second language acquisition (pp. 149-161). Rowley, MA: Newbury House.

Gass, S. M. & Varonis, E. (1985b). Variation in native speaker speech modification to non-native speakers. Studies in Second Language Acquisition, 7, 37-58.

Gass, S. M. & Varonis, E. (1986). Sex differences in NNS/NNS interactions. In R. R. Day (Ed.), Talking to learn: Conversation in second language acquisition (pp. 327- 351). Rowley, MA: Newbury House.

Goodwin, M. H. (1990). Tactical uses of stories: Participation frameworks within girls’ and boys’ disputes. Discourse Processes, 13, 33-71.

Han, Z. (2002). A study of the impact of recasts on tense consistency in L2 output. TESOL Quarterly, 36, 543-572.

Holmes, J. (1998). Complimenting- A positive politeness strategy. In J. Coates (Ed.), Language and gender: A reader (pp. 100-120). Oxford: Blackwell.

Iwashita, N. (2003). Negative feedback and positive evidence in task-based interaction: Differential effects on L2 development. Studies in Second language Acquisition, 25, 1-36.

Kaplan, R. B. (2002). The Oxford handbook of applied linguistics. Oxford: Oxford University Press.

Kasanga, L. A. (1996). Effect of gender on the rate of interaction: Some implications for second language acquisition and classroom practice. I.T.L. Review of Applied Linguistics, 111-112, 155-192.

Kyratzis, A. & Guo, J. (2001). Preschool girls’ and boys’ verbal conflict strategies in the United States and China. Research on Language and Social Interaction, 34, 45-74.

Kyratzis, A. (2000). Tactical uses of narratives in same-sex groups. Discourse Processes, 29, 269-299.

Leeman, J. (2003). Recasts and second language development: Beyond negative evidence.Studies in Second Language Acquisition, 25, 37-63.

Loewen, S. (2003). Variation in the frequency and characteristics of incidental focus on form. Language Teaching Research, 7, 315-345.

Loewen, S. (2004). Uptake in incidental focus on form in meaning-focused ESL lessons. Language Learning, 54, 153-188.

Long, M. H. (1983). Native speaker/non-native speaker conversation and the negotiation of Comprehensible input. Applied Linguistics, 4, 126-141.

Long, M. H. (1991). Focus on form: A design feature in language teaching methodology. In K. de Bot, R. Ginsburg, & C. Kramsch (Eds.), Foreign language research in cross-cultural perspective (pp. 39-52). Amsterdam: John Benjamins.

Long, M. H. (1996). The role of the linguistic environment in second language acquisition. In W. C. Ritchie & T. K. Bhatia (Eds.), Handbook of research on language acquisition (Vol. 2, Second language acquisition, pp. 413-468). New York:
Academic Press.

Mackey, A. (1999). Input, interaction, and second language development: An empirical study of question formation in ESL. Studies in Second Language Acquisition, 21, 557-587.

Mackey, A. (2006). Feedback, noticing and instructed second language learning. Applied Linguistics, 27(3), 405-430.

Mackey, A., & Oliver, R. (2002). Interactional feedback and children's L2 development. System, 30, 459-477.

Mackey, A., & Philip, J. (1998). Conversational interaction and second language development: Recasts, responses, and red herrings? Modern Language Journal, 82, 338-356.

Mitchell, R. & Myles, F. (2004). Second language learning theories. Oxford: Oxford University Press.

Nunan, D. (2006). Task-based language teaching in the Asia context: Defining 'task'. Asian EFL Journal, 8(3), 12-18.       

Ochs, E. (1993). Indexing gender. In B. D. Miller (Ed.), Sex and gender hierarchies (pp. 146-169). New York: Cambridge University Press.

Oliver, R. (2002). The patterns of negotiation for meaning in child interactions. Modern Language Journal, 86, 97-111.

Pica, T. (1994a). Questions from the language classroom: Research perspectives. TESOL Quarterly, 28, 49-79.

Pica, T. (1994b). Research on negotiation: What does it reveal about second language  learning conditions, processes, and outcomes? Language Learning, 44, 493-527

Pica, T. & Doughty, C. (1985). The role of group work in classroom second language acquisition. Studies in Second Language Acquisition, 7, 223-248.

Pica, T., Holliday, L., Lewis, N. E., Berducci, D., & Newman, J. (1991). Language learning through interaction: What role does gender play? Studies in Second  Language Acquisition, 13, 343-376.

Pica, T., Holliday, L., Lewis, N. E., & Morgenthaler, L. (1989). Comprehensible output as an outcome of linguistic demands on the learner. Studies in Second Language Acquisition, 11, 63-90.

Pica, T., Kanagy, R., & Falodun, J. (1993). Choosing and using communication tasks for second language instruction and research. In G. Crooks & S. M. Gass (Eds.), Tasks and language learning: Integrating theory and practice (pp. 9-34). Clevedon, UK: Multilingual Matters.

Piller, I. & Pavlenko, A. (2001). Introduction: Multilingualism, second language learning, and Gender. In A. Pavlenko, A. Blackledge, I. Piller, & M. Teutsch-Deyer (Eds.), Multilingualism, second language learning, and gender (pp. 1-13). New York: Mouton de Gruyter.

Richards, J., Platt, J., & Weber, H. (1985). Longman dictionary of AppliedLinguistics. London: Longman.         

Reid, P. T., Hariots, C., Kelly, E., & Holland, N. E. (1995). Socialization of girls: Issues of ethnicity in gender development. In H. Landrine (Ed.), Bringing cultural diversity to feminist psychology: Theory, research, and practice (pp. 93-111). Washington, DC: American Psychological Association.

Robinson, P. (2001). Individual differences, cognitive abilities, aptitude complexes and learning conditions in second language acquisition. Second Language Research, 17, 368-392.

Romaine, S. (2003). Variation in language and gender. In J. Holmes & M. Meyerhoff (Eds.), Handbook of language and gender (pp. 98-118). Oxford: Blackwell.

Sheldon, A. (1990). Pickle fights: Gendered talk in preschool disputes. Discourse Processes, 13, 5-31.

Swain, M. (1995). Three functions of output in second language learning. In G. Cook & B. Seidlhofer (Eds.), Principles and practice in applied linguistics: Studies in honor of H.G. Widdowson (pp. 125-144). Oxford: Oxford University Press.

Swain, M. & Lapkin, S. (1998). Interaction and second language learning: Two adolescent French immersion students working together. Modern Language Journal, 82, 320-337.

Taguchi, N. (2007). Task difficulty in oral speech act production. Applied Linguistics, 28, 113-135.

Tannen, D. (1990). Gender differences in topical coherence: Creating involvement in best friends’ talk. Discourse Processes, 13, 37-90.

Williams, J. (1999). Learner-generated attention to form. Language Learning, 49, 583-625.