Negotiating Success: Negotiating and Co-Constructing ...
ABSTRACT
An abstract of the thesis of Dax Alan Seutter for the Master of Arts in Teaching English to Speakers of Other Languages presented June 7, 2007.
Title: Negotiating Success: Negotiation and Interaction in ESL Pair Activities.
Verbal interactions of language learners provide opportunities for learners to master new skills, realize development needed, and test new language. Learners often emerge from verbal exchanges with slightly greater proficiency than when they entered. Conversation involving negotiation is particularly beneficial to second language learners.
Interaction and negotiation research was initially motivated by studies in child first language acquisition. Negotiation features were defined and studies sought to describe their influence on learning. Many variables that may influence negotiation, including those of task and interlocutor, have been investigated, yet a great deal is still unknown.
This research analyzes video/audio recordings of the pair work of one adult low-level ESL learner in a Lab School to describe how negotiation is influenced by the variables of task and partner. This study examines the interactions of students from the largely underrepresented demographic of adult low-level ESL learners in a classroom context rather than a more typical experimental setting. The data represents one focal student working on the same kind of task with different partners and with the same partner on different kinds of tasks. Negotiation features (Long, 1980, 1996) were identified and each interaction was analyzed with regard to it’s dyadic interaction pattern (Storch, 2002).
The focal student’s interactions revealed that there is a tendency for student pair work in this setting to be of a collaborative nature. In addition, task language support and interlocutor both play an interdependent role in structuring negotiation. Finally, student roles in an interaction may be related to their relative second language proficiencies.
In conclusion, these results show that adult low-level ESL learners work cooperatively and use negotiation to repair and guide the completion of tasks in pair work. Additionally, a lower amount of language support can help reveal slight differences in second language proficiency. Teachers should remember that the variables affecting negotiation are complex and interrelated, and that students will seek the language support they need from a variety of resources (teacher, fellow students, the task, etc.).
NEGOTIATING SUCCESS:
NEGOTIATION AND INTERACTION
IN ESL PAIR ACTIVITIES
by
DAX ALAN SEUTTER
A thesis submitted in partial fulfillment of the
requirements for the degree of
MASTER OF ARTS
in
TEACHING ENGLISH TO SPEAKERS OF OTHER LANGUAGES
Portland State University
2007
ACKNOWLEDGEMENTS
First and foremost, I thank God whose grace and spiritual love has allowed me to complete this endeavor. God gave strength to the little voice inside of me that always said I would finish even despite the setbacks and frustrations along the way.
The unconditional love of my mother Carrie A. Baird has been one of my greatest resources in completing this paper, this degree, and indeed my entire academic career. My mother is the most giving, caring, and loving woman I have ever known. Thank you mom for always giving me your full, unconditional support and for always being that person I know I can confide in completely. Thanks are also due to my loving sister and brothers; April Dawn, Cary Andrew, and Bodie Thomas. Grandma Case and Grandma Charlotte also deserve my gratitude. I am very honored to call you all my family and am deeply grateful for your presence in my life. To my loving father Jerry Richard Seutter who I will be with again in the life after, I owe you a great debt for the values and work ethic you instilled in me at a young age. I love you and am forever grateful.
To my advisor Kathryn Harris, if patience and encouragement were droplets of water, you would be an ocean. Page after page, revision after revision, draft after draft; you always afforded me the pep talk and the meeting extension that I needed to keep moving forward. Thank you so very much for all your help during this journey.
Finally, to my dear friends Caleb Francis and Megan Unruh, I love you both very much. You both provided immeasurable support along the way. Thank you all!!
TABLE OF CONTENTS
ACKNOWLEDGMENTS ii
LIST OF TABLES vi
LIST OF FIGURES vii
GLOSSARY viii
CHAPTER 1 INTRODUCTION 1
Background 1
Purpose and Goals 3
Relevance & Implications for Teaching 4
Scope & Design 4
Research Questions 5
CHAPTER 2 LITERATURE REVIEW 6
Child L1 Acquisition Studies 7
Negotiation of Meaning & Comprehensible Input 10 Negotiation and Acquisition 16
Comprehensible Output 20
Swain’s Output Hypothesis 22
Co-Construction & Collective Scaffolding 25
Task 28
Information Exchange 28
Content and Task Familiarity 32
Features of the Interlocutor 34
Interlocutor Familiarity 35
Gender 36
Differences in L1, Education, and L2 Proficiency 36
Interlocutor Relations 38
Summary 41
CHAPTER 3 METHODOLOGY 43
Research Questions 43
Design 43
Setting 43
The Classroom Recording Context 46
Participant Selection 47
Data Selection Criteria 49
Pilot Study 53
Data Analysis 54
Phase One: Negotiation Framework 54
Phase Two: Mutuality and Equality 55
CHAPTER 4 RESULTS 58
Clip #1: Personal Info Exchange 59
Clip #2: Sibling Similarities 60
Clip #3: Giving/Receiving Directions 62
Clip #4: Favorite Family Member 63
Results from Phase 1: Negotiation Features 65
Results from Phase 2: Collaboration Continua Placement 66
Support for Placement along the Collaboration Continua 66
Personal Info Exchange 67
Sibling Similarities 68
Giving and Receiving Directions 69
Favorite Family Member 70
Discussion of Results Found & Comparisons between Clips 70 Similar Interlocutor / Dissimilar Task 71
Differences in Negotiation 73
Comparative Placement on the Collaboration Continua 79
Dissimilar Interlocutor / Similar Task 81
Differences in Negotiation 84
Proficiency Level & Task Language Support 86
Comparative Placement on the Collaboration Continua 87
Similar Interlocutor / Similar Task 88
Differences in Negotiation 90
Comparative Placement on the Collaboration Continua 91
Summary 92
CHAPTER 5 DISCUSSION 93
Summary of Findings 93
Collaborative Majority 94
Task and Interlocutor Interdependence 95
Relative Proficiency Levels 97
Task and Interlocutor as Variables of Negotiation 98
Students as Teachers 101
Use of Available Resources 102
Benefits of Being Off-Task 103
Limitations of this Study 104
Questions for Further Research 105
Implications for Teaching 107
REFERENCES 109
LIST OF TABLES
Table 1 Definitions for PCC ESL Class Levels 45
Table 2 Negotiation Features 65
Table 3 Degree of Collaboration 66
Table 4 Highlights from Personal Info Exchange 67
Table 5 Highlights from Sibling Similarities 68
Table 6 Highlights from Giving/Receiving Directions 69
Table 7 Highlights from Favorite Family Member 70
Table 8 Personal Info Exchange Compared with Giving/Receiving Directions 74
Table 9 Giving/Receiving Directions Compared with Favorite Family Member 84
Table 10 Sibling Similarities Compared with Personal Info Exchange 90
LIST OF FIGURES
Figure 1 Dyadic Interaction Model 39
Figure 2 Clip Comparison Combinations 58
GLOSSARY
Comprehensible Input: Input that a learner is able to comprehend and thereby acquire.
Comprehensible Output: This language that a learner produces can be comprehended by that person’s interlocutor. Swain (1985) argued that in striving to make their output comprehensible to their interlocutor a learner is able to notice gaps or deficiencies in their interlanguage. This idea is part of ‘The Output Hypothesis’.
Conversational Adjustments or Conversational Modifications: These are ways by which language can be modified and manipulated to promote comprehensibility. This term is synonymous with negotiation of meaning (Foster, 1998; Hardy & Moore, 2004; Long, 1985).
Input: In relation to Second Language Acquisition, input refers to all of the linguistic information in all forms that is presented to a learner or interlocutor of that language.
Interaction: With regard to Second Language Acquisition, interaction is the exchange (verbal or written) between any two or more language users.
Interlanguage: An interlanguage is a language represented by a continuum between the native language and a target language along which all learners traverse. At any point along the continuum, the learners’ language is systematic, i.e. rule-governed, and common to all learners, any difference being explicable by differences in their learning experience. (Larsen-Freeman & Long, 1991), see also; (Selinker, 1972)
Negotiation (of Meaning): The name of a group of conversational techniques speakers use to promote comprehension during conversation. A speaker may use negotiation of meaning either to understand better or to help their speaking partner better understand the message. Negotiation of meaning is also used to problem solve when both speakers are having difficulty in comprehension.
Native Speaker, NS: The Oxford Companion to the English Language defines a native speaker as “a person who has spoken a certain language since early childhood” (Christophersen & McArthur, 1992, p. 682).
Nonnative Speaker, NNS: A person for which some given language is not their first language. A non-native speaker of English is a person who learned English after gaining a native command of their mother tongue.
CHAPTER 1
INTRODUCTION
“Hhmm?” “What did you say?” “Uh huh.” “Mm hmm.” “Really?” “Um hmm.” “Quiet, yeah?” “What is quiet?” “Yeah.” “Understand?” “Oh, gas station?” “Tigard, Tigard .. T .. T”. Taken out of context the above phrases seem confusing. However, in the right context they serve the purpose of dispelling confusion and misunderstanding. These are examples of language that speakers use to clarify information and to check understanding. When people talk, they often use language similar to the examples above to construct and guide the ebb and flow of conversation. All types of speakers use this kind of helpful language in conversation. The use of this language has been documented in speech between young children and their caretakers, native speakers and non-native speakers, and between pairs of non-native speakers and language learners. The factors involved in the use of this language by second language learners in a classroom setting are far from completely understood. This area of research is fertile ground for supplying answers to remaining questions of how second language acquisition takes place.
Background
People use language to talk to and interact with other people. This verbal interaction itself can play an important role in the learning process. Learning through interaction is especially important in the second language classroom. When misunderstanding occurs, phrases such as those above allow learners to use their shared knowledge to bridge the gap in comprehension. Working to resolve these difficulties contributes to the learning process. When speakers negotiate their communication breakdowns, one or both participants often leave with more information as a result. For language learners this can occur both in and out of the classroom. The focus here will be how this takes place in the classroom.
When using a second language, learners are trying to understand as well as be understood. While engaging in pair work in a language classroom, students often need to resolve communication breakdowns. One way to do this is by signaling a question (“What did you say?”) or giving an indication of non-understanding (“Huh?”). This process of negotiation is part of most aspects of human communication. Negotiation happens when language is manipulated in an attempt to increase comprehension. Some functions of negotiation are seeking clarification regarding what the other person said, confirming the correctness of what was heard, and confirming a partners correct comprehension. In addition to serving communication, negotiation is believed to facilitate language learning. Through the process of increasing comprehension, it is believed that negotiation also increases acquisition (Long, 1980, 1985). Negotiation helps to fill gaps in a person’s understanding of the second language. In attempting to be understood, learners often notice their own difficulties in the second language and, with the help of their partner, fill in these gaps with new knowledge through the processes of interaction and negotiation (Swain, 1985).
The concept that working through these misunderstandings leads to acquiring new language skills is one that has its origins in research of first language (L1) acquisition in children (Snow, 1972, 1986). Research revealed that parents and caretakers adjust their speech, making it more comprehensible, in response to cues given by the children. Those interested in speech between native and non-native speakers from a Second Language Acquisition (SLA) background began to take notice of these findings and ask how they might apply them to their own field of research.
In order to find answers to how language learning takes place, the Interactionist perspective of SLA focuses on what occurs during conversation and what learners do to get the language they need. Research on interaction and negotiation in the second language classroom can help reveal how learners traverse the journey toward language acquisition and how they overcome some of the obstacles they face along the way. Although all interactions do not necessarily include negotiation, all negotiation takes place within the context of interaction. Most verbal interactions, if continued long enough, will necessitate some type of negotiation. This fact is especially true in the second language classroom where, through the course-prescribed language tasks, learners are constantly presented with language that pushes them beyond their current understanding.
Purpose and Goals
The purpose of this study is to examine the quality and quantity of negotiation and the interaction types found in pair work by adult low-level ESL learners. The goal is to compare the interactions of one learner across the dimensions of interlocutor and task type. Task type will be considered with regard to the amount of language support being provided within the task. The intent of this research is to determine if negotiation and type of interaction differs along these variables and if so to describe the variation.
Relevance and Implications for Teaching
Although researchers know that negotiation is a common event in the classroom, there is still a great deal that is not understood. A better working knowledge of the factors surrounding negotiation will benefit language teachers and learners alike. By understanding the factors that lead to more or less negotiation, teachers may attempt to create classroom environments and tasks that maximize the amount of negotiation. If, for example, it is determined that certain partner combinations result in an increase in successful negotiation, then a teacher could actively construct those kinds of pairings. Furthermore, if it is found that inherent language support in the task influences the amount or kind of negotiation, the teacher could plan classroom tasks with this information in mind. Finally, finding answers to these questions will give teachers a better understanding of why they see the kind of negotiation and interaction they observe between learners in the ESL classroom.
Scope and Design
To better understand negotiation in the second language classroom, my research examines the negotiation and interaction pattern found in the pair work of one adult low-level ESL learner. In the data under analysis, one learner is working with the same partner across different task types and working with different partners on the same task type. Data comes from a focal student chosen with regard to several criteria described later in chapter three.
The data analysis contains two phases. The first phase of analysis involves identifying the number of negotiation features found in the interactions. During the second phase of analysis, a framework developed by Storch (2001, 2002) is used to describe the patterns of interaction present in the pair work. Following my data analysis, a concluding discussion explains the study results and their implications to the current body of research and teaching in the second language classroom. This study attempts to give a more complete understanding to how aspects of task and partnering influence the quality and quantity of interaction and negotiation during the course of pair work.
Research Questions
The following questions form the basis of this research study:
1) How does one adult low-level ESL learner’s negotiation vary when working on two different task types?
2) How does one adult low-level ESL learner’s negotiation change when working with two different interlocutors?
CHAPTER 2
LITERATURE REVIEW
Conversation was once viewed solely as a place to practice language that was learned in the classroom. Conversation, however, is now known to be an important language learning tool in its own right. Child first language acquisition studies helped motivate second language acquisition research in exploring the learning value of conversation. Conversation promotes learning through negotiation. Negotiation occurs when students attempt to work through a breakdown in communication. Negotiation serves many functions beneficial to language acquisition. Some researchers even suggest a causative link between negotiation and language acquisition.
Many previous studies have examined negotiation between native and non-native speakers as well as pairs of non-native speakers. The majority of these studies have taken place in an experimental setting, and very few have made use of an authentic ESL classroom. Past studies have also looked at a wide variety of participants, but very few have concentrated on adult low-level learners. A large portion of ESL learners in the United States today fit this demographic making it an important area of further research.
In understanding the variables that influence negotiation, task and interlocutor are very important. As teachers, we are often largely responsible for the kinds of tasks students complete in the classroom. Student grouping for pair work is also often initiated by the teacher. Both variables of task and interlocutor are complex and dynamic. Previous research has taken many different directions in investigating their roles in negotiation. In this research I explore the dimensions of task and interlocutor in the pair interactions of adult ESL learners in a non-experimental classroom setting.
Child L1 Acquisition Studies
Inquiries into SLA interactions have their foundations in research investigating child first language acquisition, interaction, and learning (Snow, 1986). Critical to understanding first language acquisition is the interaction that children have with their caregivers. In referring to the way adults talk to children, Snow (1986) uses the term “child directed speech” (CDS). She uses this term because of issues that she has with the connotations of other terms such as baby talk, motherese, and caretaker talk, all of which have been used to describe a similar register of speech. Child directed speech seems to be the most accurate label. What makes this speech so special that it needs a different name?
Child directed speech was initially thought not to differ in any significant way from the speech used among adults, however this was quickly discovered to be a false assumption (Phillips, 1973; Sachs et al., 1976). In fact, child directed speech is very different; it is simpler, more grammatical, and highly redundant. It seems that speech directed toward children is modified to accommodate the child’s linguistic ability. In a similar way, language spoken to non-native speakers is often modified to accommodate their second language ability. These similarities give ideas about what kind of language input is needed for successful language acquisition in general.
The next step after describing child directed speech was an attempt to answer why it contains the features it does. Snow (1972) found that feedback from the child was responsible for some of the features of child directed speech. In Snow’s study, mothers were asked to make tapes that would be played back to their children later. Although the language the mothers used on the tapes was simpler and more redundant than that which would be addressed to an adult, it was found that speech used by the mother when talking directly to the child in their presence was even simpler and more redundant still. This suggests that some of the features occurred as adjustments made in response to cues from the child and that they were indeed crucial to language acquisition. Child first language acquisition research eventually led to questions about Native Speakers (NSs) interactions with Non-Native Speakers (NNSs) or “Foreigner Talk” (Ferguson, 1971; Long, 1980, 1983). Foreigner talk is a term used to describe a register of speech that native speakers use when talking to non-native speakers. This register is different from that used between native speakers in the same way that the register of child directed speech differs from that used between two adults. Studies examining foreigner talk prompted questions about how it might benefit language acquisition by better enabling the NNSs to understand the discourse. NSs make adjustments in their speech when speaking to NNSs in order to make speech more comprehensible for NNSs. Additionally, NSs respond to cues given by the NNSs. These cues by the NNSs signal when there has been some misunderstanding or lack of understanding in the conversation. It is thought that this effort on the part of the speakers leads to comprehensible input. Comprehensible input is language that a learner is able to comprehend and thereby acquire. It has been at the root of the majority of interaction and negotiation research.
Negotiation is something that learners do in an attempt to make language comprehensible. It is their desire to understand the language, which creates the need for the discourse strategies of negotiation. Child first language learners and adult second language learners share many of the same discourse strategies and obstacles as they acquire and utilize language. Similarities exist between the ways in which children try to understand and be understood and the ways in which second language learners do. Additionally, the ways in which language is manipulated in order to make it more comprehensible is comparable between both sets of learners (Hatch, 1978a, 1978b). Examining how children interact verbally and negotiate discourse in their first language has revealed many similarities to adult second language learners. It is important to remember that negotiation is not exactly the same for child L1 and adult L2 learners; however, the similarities in negotiation performed by these two groups of learners suggests it may be universally important in language learning.
The way that parents/adults adjust their speech to meet a child’s linguistic needs is essentially the same as the way in which negotiation is used to make input more comprehensible for a second language learner. The following section will describe the conversational modifications that occur during speech directed toward L2 learners. The important information to be derived from all of this is that the commonality of these adjustments or modifications across these different types of discourse suggests they represent a vital component of language acquisition in general.
Negotiation of Meaning & Comprehensible Input
For many years, the field of second language acquisition (SLA) has concerned itself with the factors that underlie successful language learning. In looking for answers about the process of language learning, studies in SLA have looked at findings in the field of L1 acquisition. SLA research has led to many theories both contradictory and concordant, which have in turn created an ever-expanding body of knowledge about the process of language learning. One area which has been given significant attention since the late 1970s and early 1980s is that of conversational interactions (Hatch, 1978a; Long, 1980). This area of research is a result of findings suggesting parallel processes at work between SLA and L1 acquisition.
In the late 1970s, there occurred a shift in thinking with regard to the importance of conversation in SLA. This shift was partially motivated by findings that were being made at the time in the realm of child L1 acquisition. Prior to this critical juncture, it was believed that spoken discourse simply reinforced what was learned previously in the classroom and only gave learners an opportunity to put their knowledge on display. In the classroom, this resulted in speaking that was highly organized and orchestrated by the teachers and/or textbooks. This idea could be thought of as designing verbal performances for learners. A major ideological shift occurred when Hatch and associates (Hatch, 1978a; Hatch & Wagner-Gough, 1976) showed that conversational interactions give learners opportunities to produce and be exposed to L2 features that are not already known to them. They hypothesized that interaction actually fosters the development of L2 syntax rather than just being a place of demonstration. This process was thought to be parallel to that at work in child L1 acquisition in which verbal interactions with adults facilitate acquisition through exposure to new syntax and lexical items.
Researchers realized that conversational interactions gave learners an opportunities to hear and produce language that did not already exist in their interlanguage (Hatch, 1978a; Hatch & Wagner-Gough, 1976). A learner’s interlanguage can be described as the use of the target language at any given point along his or her learning continuum (Selinker, 1972). In addition to exposing learners to new forms, interaction also provides learners with feedback on the comprehensibility of their interlanguage. Yet another benefit of interaction is that it helps the learner receive input that is more comprehensible. Of the many benefits of interaction, comprehensible input is perhaps one of the most valuable.
Comprehensible input was and still is considered an extremely important requirement for language acquisition (Krashen, 1982, 1985). In order for learners to acquire some bit of language data, they must first be able to comprehend it. At face value this sounds like an obvious proposition; however, it leads to many important questions. For instance, what things help create comprehensible input in a classroom environment? If negotiation and interaction help to make input more comprehensible, by what mechanism do they?
Long (1980) also argued for the necessity of comprehensible input in the second language acquisition process. He stated that input was made comprehensible to and by nonnative speakers through what he called conversational adjustments or conversational modifications. By conversational modifications, Long (1980) was referring to features he called clarification requests, confirmation checks, comprehension checks, and repetition. These features of language are real-time cues put out by the interlocutors about the comprehensibility of the discourse at hand. A clarification request is a statement or question by one participant meant to elicit further elucidation regarding a previous statement. A classic example of this would be, “What did you say?” A confirmation check is a statement meant to verify that a previous statement was indeed understood correctly. For example, “You said X, right?” A comprehension check is a statement by the speaker meant to determine that the listener is following and understanding what is being said. “You know what I mean?” or “Do you understand?” are two instances of statements that could be used as comprehension checks. Finally, repetition is simply that, and occurs when a person repeats some part of their own utterance (self repetition) or their interlocutors (other repetition). Repetition is often a subtler means of achieving the same ends as these other techniques because instead of explicitly stating the need, it relays the message in the intonation of the repetition. Other repetition with a rising intonation may be an instance of a clarification request or confirmation check. A self repetition may also serve as a comprehension check. Additionally, because low-level learners often lack the linguistic knowledge to utilize full statement negotiation moves like those listed above, they frequently turn to repetition in combination with body language and intonation to achieve the same ends. It is important to remember that the ends achieved by these verbal acts of negotiation are sometimes reached through non-verbal (body language) or even sub-vocal (sounds, not words) means. Another more recently coined category considered here is that of untriggered other repair. This takes place when one student, upon hearing a mistake by his or her interlocutor, gives unsolicited corrective feedback to that interlocutor (Harris et al., 2006). Similar to untriggered other repairs are recasts. Recasts are a specific type of untriggered other repair in which the person giving the feedback repeats the previous utterance in the corrected form, to the person who made the error. Now that there is some definition as to what negotiation is, it is time to determine how it relates to acquisition.
What is the process by which negotiation leads to acquisition and how can this be supported by data? Long (1985) suggested a three-part process for providing support for this claim. First, data must provide evidence that better comprehension improves acquisition. Secondly, it must be shown that negotiation leads to better comprehension. Finally, coming full circle, the claim could then be made that negotiation benefits acquisition by way of increased comprehension. In his studies, Long showed that even though both NS and NNSs engaged in conversational modifications (the same as negotiation), the latter did so with much greater frequency. His conclusion was that NNSs used conversational modifications as a means of requesting and thereby receiving input that is more comprehensible, as well as producing output that is more comprehensible. Long (1983) suggested that conversational modifications were valuable because they allowed a learner to manipulate the input in a way that made it more comprehensible. In other words, a recipient was able to signal that the message was not understood, resulting in reformulations by the interlocutor until understanding was reached.
Early research looking at negotiation and interaction primarily followed the pattern of comparison used in Long’s initial investigations of NSs talking to NNSs (1980). Long sought to prove that NSs made adjustments in their speech through the process of negotiation in order to accommodate the proficiency level of NNSs. Work describing the talk used by NSs towards NNSs was carried out extensively by Ferguson (1971) who referred to this special register of speech as ‘foreigner talk’. Beyond describing the ways in which this register is different, it is important to understand how it benefits acquisition. The studies described below attempt to answer that question.
Pica, Young, and Doughty (1987) compared the comprehension of NS instructions by NNS students under the two input conditions of pre-modified input and interactionally modified input. The task involved the correct placement of various items on a board by NNSs. In the pre-modified input condition, learners were given input that had been simplified through repetition and paraphrase. Vocabulary was restricted to common or familiar items. Boundary markers and sentence connectors were also added to reduce sentence length and complexity. All of these things were done in order to make the input more comprehensible to the learner without the use of negotiation. In the second condition, the input was presented in its original unmodified form; however, both interlocutors were allowed to negotiate during the interaction in order to arrive at mutual understanding.
The hypothesis was that learners in the interactionally modified condition would receive input that was more comprehensible during the course of the task resulting in a greater degree of accuracy in object placement. This prediction was indeed correct. Furthermore, repetition was found to be the most influential component in comprehension, more so than adjustments in grammatical complexity. Interaction in general increased comprehension, but particularly when it contained confirmation checks, comprehension checks, and clarification requests that in turn resulted in more repetitions. The findings by Pica, Young, and Doughty (1987) provide further support to the claims made by Hatch (1978a) and Long (1980, 1985) regarding the benefits of negotiation in making input more comprehensible. Pica, Young, and Doughty (1987) do not claim a direct link between interactional modifications and learning; rather, they provide empirical support for theoretical claims that interactional modifications, or negotiation, aid in input comprehension. The next step in connecting negotiation to learning would be to provide evidence for the theory that comprehensible input leads to or is necessary for acquisition. This is the same as the three-part model suggested by Long (1985) to provide evidence that negotiation benefits acquisition. Finally, research by Pica, Young, and Doughty (1987) expanded the arena of negotiation/interaction research by establishing input type as an important consideration.
Negotiation and Acquisition
Although studies like the one described above suggest a positive connection between negotiation and later comprehension, the link between negotiation and acquisition is still nebulous in many areas. Much of the previous research has only been able to support claims of correlation rather than causation. The research in this area has thus far yielded more questions than answers. Chaudron (1988) points to the lack of classroom research and describes the conclusions as being limited in scope, though highly suggestive for further research. More recently, however some researchers have set out to design studies that provide evidence for a more direct link between negotiation and acquisition.
A crucial first step in exploring a direct empirical support for a causative link between negotiation and interaction is to determine how acquisition will be defined and measured. Acquisition of lexical items may be the most straightforward and simplest way to test for the effects of negotiation on acquisition, as it has thus far been the most commonly used measure to check for acquisition.
Do type of input and interactional context of the task affect the amount of comprehension and acquisition of second language word meanings? This is the premise of a study by Ellis et al. (1994). These researchers tested the vocabulary comprehension and acquisition across three different groups of learners: a baseline group (B), a group receiving pre-modified input (PI), and a group receiving interactionally modified (IM) input. All groups performed the same task, but the input that they were given to complete it varied. The students performed a listening task in which they heard a set of directions and then were required to choose the correct picture of kitchen utensils and write the correct location of those items in a picture of a kitchen. The baseline (control) group received unmodified directions at a normal rate (180 wpm) with no interaction. The pre-modified group listened to directions containing negotiation between students and a native speaker, which were also given at a slower rate (90 wpm). The interactionally modified group received the same directions given at the same rate as the baseline group, but was allowed and encouraged to interact with the teacher over problem areas in the directions. Students in this group were also provided with language frames for seeking clarification, to aid them in negotiation (i.e., “Will you please speak more slowly?” “What is a _____?”).
The authors hypothesized that: 1) Input through interaction will be quantitatively and qualitatively different from pre-modified or unmodified input. 2) Exposure to direct or observed interaction will lead to higher L2 comprehension. 3) Learners receiving interactionally modified input will learn and retain more lexical items. 4) Learners who actively participate in interaction will comprehend and retain more than those that do not. Two post-tests and one follow-up test were used to test for vocabulary acquisition.
The results of this study supported some, but not all of the authors’ hypotheses. The amount of input was indeed qualitatively different in the three groups. The input in the pre-modified input and interactionally modified groups were longer and more redundant than the baseline group, with the interactionally modified group being by far the longest and most redundant. Additionally, the students participating in the interactionally modified group showed much greater comprehension than those in either the pre-modified or baseline group. With regard to acquisition, the interactionally modified group outscored the pre-modified group in vocabulary acquisition, but only in the short term. The researchers stated that it was not clear whether the interactionally modified group enjoyed the advantages that it did because of the negotiation itself or the greater processing time it allowed during the task. Results showed that students not only seem to benefit from directly participating in negotiated interaction, but may indeed also benefit from indirect participation. This suggests that learners who prefer a quieter approach to language learning may benefit from listening to the interactional activity of their peers. The authors comment that because of the complex nature of acquisition, even given the posttests, it is difficult to say whether the items seemingly acquired were indeed fully acquired by the learners. In closing, the authors do claim to have demonstrated support for a causative relationship between negotiated interaction and acquisition, but caution that because of the complexity of the acquisition process the results cannot be generalized to aspects of language beyond lexical acquisition.
In another study, Smith (2004) attempted to establish a link between learner-learner interaction and acquisition, within the context of a computer chat environment. This study used the acquisition of concrete nouns as a means to test the Interaction Hypothesis (Long, 1996). Students performed a jigsaw task and a decision making task and were later tested on their knowledge of concrete nouns presented in these tasks. To determine if lexical acquisition took place the students performed a receptive and a productive test of their knowledge of the nouns both immediately following their completion of the task and after one week’s time. The short time delay (7 days) was cited as a practical limitation. This delay would have been longer, but it was limited by the students’ school schedules. The results showed that nouns negotiated during the course of the tasks were retained at a much higher rate than those that weren’t. Higher retention of the negotiated items was found in both the immediate and the delayed posttests. Additionally because no significant differences were found between the students’ scores on the immediate and delayed posttests, the results were claimed to have held up over time. Smith’s findings are especially important because they provide evidence for a direct rather than a causal link between negotiation and acquisition.
Smith (2005) later used the same data as in Smith (2004) to explore the role that the complexity of the negotiation plays in learner uptake and the relationship between successful learner uptake and subsequent acquisition. Learner uptake is a learner’s response to a statement made by his/her interlocutor about some problem in his/her previous utterance. Uptake is generally thought to be successful when the speaker who makes the mistake immediately makes a correction based on the feedback they receive. Smith coded his data as either ‘No Uptake’, ‘Recognize’, or ‘Apply’; according to the amount of uptake. Smith’s data is the collection of negotiated interactions surrounding unknown lexical items from his previous study (Smith, 2004). Each negotiation of an item was labeled as being complex or simple with regard to both its length and the conceptual depth. Smith cautions several times that the results from this study cannot be generalized beyond the unique environment of the computer-chat language task from where the data came. Smith found that complexity of the negotiation did not have an effect on learner uptake. Additionally successful uptake did not seem to facilitate acquisition. Smith suggested that the notion of uptake in a computer chat environment must be adjusted to include delayed uptake and that the permanence of the chat log for later review may make immediate uptake less of an important variable. Finally, although these results shed some light on the function of negotiation, learner uptake, and acquisition, they cannot be generalized outside of the computer chat environment.
Comprehensible Output
Interaction research has begun to illuminate the importance of comprehensible input in second language learning. Studies have shown how input can provide a learner with feedback about the state of their interlanguage; and how learners can give cues, through negotiation, which help them receive input they are better able to understand. Research examining input shows that even in the cases of near optimal input something is still needed to make acquisition complete (Swain, 1985). A learner’s output or language production is also an important aspect of interaction and second language acquisition. When engaged in a conversation the speaker often receives cues from the listener about whether or not the message is being understood. Because the goal of conversation is comprehension of a message, the learner reacts to these cues by adjusting or modifying his or her speech in an attempt to convey the message. This, essentially, stretches proficiency in the L2 to new levels and informs learners about existing deficiencies in their interlanguage of which they may or may not have been previously aware. Additionally the learner will often modify his or her utterance using input that is supplied by the interlocutor in a previous utterance, thereby trying new language that he or she may not have used previously or been unable to generate on his or her own. In conversation, learners are also able to attempt language that they know, but may not be certain about, and then receive immediate on-the-spot feedback, as to whether or not it has been understood and sometimes about whether or not it is grammatically correct. In the next section, I will discuss research that has investigated the functions of output in SLA.
What role does output play in SLA? What is the connection between negotiation and output? Does negotiation indeed push learners to modify their language output? If so, then how? This section offers a brief outline concerning the connection between negotiation, output, and acquisition. Some have claimed that comprehensible input alone is not sufficient for acquisition, and that output is necessary as well. By producing language (output), learners are able to notice deficiencies or gaps in their interlanguage and are then assisted by negotiation in bridging these gaps. It is this combination of noticing and then negotiating that is thought to lead to acquisition. It is important to remember that the roles of input and output are by nature intertwined. This makes sense when we see that one interlocutor’s output is another’s input and vice versa.
Swain’s Output Hypothesis
The flipside of comprehensible input is comprehensible output. Output may become comprehensible when it is modified as a response to feedback indicating some problem in message comprehension. A speaker can become aware of a problem either by external or internal feedback. External feedback may come in the form of a clarification request or some other form of negotiation. Internal feedback may simply be the learners’ realization that they are not able to express what they wish to in the second language.
Questions regarding the role of output were motivated by data on learners of French in immersion programs (Swain, 1985). Although these students received an abundance of comprehensible input and began learning French at a young age, they were still clearly identifiable as NNSs in their speaking and writing after many years of study. Also despite much comprehensible input, they actually spoke very little French. Additionally, the French they did speak was not of any significant length, therefore lack of production was thought to be a factor in the impediment of their acquisition. These findings provided counterevidence to the notion that only comprehensible input is sufficient for SLA, therefore there must be more that is required for successful language acquisition. Swain’s study indicates that output places demands on learners, in the areas of L2 forms and syntax, in a way that input simply cannot. Swain and Lapkin (1995) make the assertion that when second language learners run into difficulty during language production or output, they become aware of, or notice, deficiencies in their interlanguage. Again, this noticing can occur by either external or internal feedback. Once learners have noticed a gap or inconsistency in their output, they modify their output so their message will be understood or seek out the new language they need. In summary, it is through output that learners are able to become aware of problems in their use of the second language and then stretch beyond their current level of proficiency through negotiation.
This noticing during output not only occurs in pair work, but also when students are working alone. In a study that was a partial replication of previous research by Cumming (1990), Swain and Lapkin (1995) asked a group of learners in a French immersion program to perform a writing task in French. During this writing task, students were instructed to think aloud or verbalize their thought processes as they constructed the written dialogue, specifically when they came across difficulties. It was hoped that this running dialogue would provide a window into the linguistic problem-solving that took place as students performed the task. These think-aloud problem-solving events were termed language-related episodes (LREs). These can happen when a learner comments on a language problem and then solves it, correctly or incorrectly, or simply solves it in the absence of commentary. The conclusion reached was that when learners notice a gap in their knowledge and engage in a language-related episode, they take part in introspective linguistic analysis that may be beneficial to second language acquisition. The results also show that noticing can take place without external feedback since this task involved only one learner. In this example, language-related episodes are akin to negotiation that takes place in solitude. In other words, LREs with one learner are instances of negotiation that a learner performs by himself or herself. Although Swain and Lapkin call for more research on output to trace the benefits of the cognitive processes to learning, they do claim involvement of the following processes during LREs: extending first language knowledge (particularly meanings) to second language contexts, extending second language knowledge to new target-language contexts, and formulating/testing hypotheses about linguistic forms/functions. Noticing is an important part of the Output hypothesis. As stated by Swain (2000), “One role for output in second language learning is that it may promote ‘noticing’ (p. 99)”. Swain illustrates several levels of noticing: 1) noticing salient/frequent features in the target language, 2) noticing a difference between the target language and one’s own interlanguage, and 3) noticing a lack of ability to say what is desired. Noticing like this causes learners to analyze their interlanguage to solve linguistic difficulties. The act of producing language often helps learners realize what they cannot say in the second language. Learners come to this realization when they notice a discrepancy between what they intend to express in the second language and what they ultimately are, or are not, able to produce. In summary, output aids acquisition by allowing learners to notice their insufficiencies in the second language and then by pushing them to correct those problem areas.
Having shown that output could aid in acquisition even in the absence of a fellow interlocutor, Swain and Lapkin (1998) went on to investigate the advantages of the output produced between two learners. In this later study, the authors conclude that dialogue between two learners can act as a cognitive tool used to co-construct the target language itself as well as knowledge about it. The linguistic work done by two learners is much the same as the work done by a solo learner like those in Swain and Lapkin’s earlier study; however, another learner lends an additional pool of linguistic resources from which to draw upon when creating a collaborative and mutually beneficial dialogue about the second language.
Co-Construction & Collective Scaffolding
When two learners work together on a language task or when they are simply in the midst of communication, a phenomenon called co-construction can occur. Co-construction is based upon the principle that two minds are better than one. When learners confront a language problem with some but not all of the necessary information or knowledge, they are often able to work together to solve the problem. This happens because one learner is able to add to the knowledge that the other possesses in order to solve the problem; both often end up with more knowledge or a better grasp of the problem than either one had initially.
Donato (1994) describes the occurrence of co-construction in a phenomenon he calls collective scaffolding. Scaffolding is a process in which a teacher or a more proficient student provides support and assistance to lower proficiency students, allowing them to work at a level slightly higher than their current ability. Donato (1994) talks about collective scaffolding within the context of the Activity Theory (Wertsch, 1979) of language acquisition. The Interactionist theory is cognitive in nature, which means that the focus on language learning is based on what a learner does to get the language necessary for SLA. Sociocultural theories of language learning such as the Activity Theory and Vygotsky’s theory have a different view of what is important for learning and where learning takes place. Activity Theory places emphasis on the motives of those involved and the internalization that occurs within the social context of the learning. In Vygotsky’s views of language development, learning cannot be examined outside of the realm of social context. Learning first takes place in the shared space between two individuals, and then, through the process of internalization, it is incorporated into the individual learner’s mindset. According to Vygotsky (1962), consciousness is co-knowledge and any individual consciousness that exists is of a secondary nature, being originally derived from the shared space. For cognitive, as well as, linguistic development Vygotsky (1978) described the zone of proximal development (ZPD). In the ZPD, learners can bridge the gap between their current level of knowledge and the next level of development with adult guidance or peer collaboration. This view places great emphasis on social interaction as a mechanism for individual, cognitive development. Social interaction can be viewed as such a mechanism in the case of scaffolding. Scaffolding occurs when a novice is drawn into, and engages within, the strategic problem-solving space of a more proficient speaker.
Scaffolding has typically been investigated in the case of speakers of unequal proficiency levels such as NS/NNS, teacher/student, parent/child, etc. Donato (1994) attempted to ascertain whether the same processes are at work within NNSs of similar proficiency levels. Donato’s focus was on observing the construction of co-knowledge. Also of importance was how that construction process results in linguistic change. The ultimate goal was to work toward answering the question of how negotiation influences L2 development. Donato found that learners are able to provide guided support in collaborative interactions in the same way that teachers do in an expert scaffolding scenario. In collective scaffolding, linguistic development by individual learners is a result of their engagement in the collaborative, dialogic process of interacting with another learner, rather than mere input crunching.
Other examples of co-construction can be seen in data from pairs of learners studied by Swain (1998). These learners built on top of one another’s language related episodes (LREs) to solve the language tasks. The final transcripts of these learners show that each learner added information that the other did not possess in order to solve the problem. Later it was found that many of the learners displayed this jointly produced language in individual tests. These examples illustrate that co-construction and collective scaffolding often involve negotiation and occur in collaborative interactions. This link to negotiation makes co-construction and collective scaffolding important considerations for second language acquisition.
Tasks make up the classroom context in which negotiation, interaction, co-construction, and collective scaffolding take place. Learner-learner interaction often takes place in the classroom during work on language tasks. Because interaction with negotiation is believed to benefit acquisition, it is important to ask if teachers might be able to select certain types of tasks to maximize the amount of negotiation and acquisition that takes place in the classroom.
Task
Tasks are some of the primary elements of activities that make up language learning and teaching. Often associated with the communicative language learning environment (Long & Crookes, 1992), tasks are comprised of students’ significant use of the target language to achieve some shared goal as designed by the teacher within the curriculum. In thinking about the issue of task and how it influences negotiation, it is important consider the many variables within a language task. Some of the aspects of task that have been dealt with in previous literature are: information exchange (one-way or two-way flow, and optional or required information exchange) (Foster, 1998; Gass & Varonis, 1985), the students’ familiarity with the given task, the students’ familiarity in working with each other or the task content (Hardy & Moore, 2004; Plough & Gass, 1993), the amount of structural or language support found in the task (Hardy & Moore, 2004; Ouellette, 2004), and other factors surrounding the task type in general.
Information Exchange
In a study by Gass and Varonis (1985) speakers’ negotiation behavior was analyzed as it related to the flow of information within communicative tasks. Other factors of interest in their study were the influence of situational and/or social roles. They performed a study that looked at two tasks, which differed in the directional flow of information. One task contained a one-way, unilateral flow of information and other a two-way, bilateral flow of information. The data was analyzed using a previously developed framework for negotiation that contained four components: 1) a trigger, 2) an indicator, 3) a response, and 4) a reaction to the response (Varonis & Gass, 1985). A trigger is some utterance or portion of an utterance made by the speaker, which causes the hearer to signal non-understanding. An indicator is the hearer’s signal that understanding has not been complete. Indicators were the primary focus of analysis with a differentiation between direct vs. indirect indicators. Direct indicators let it be known without a doubt that there has been a lack of understanding, usually through an explicit question. This is often in the form of a Wh-question utilizing words such as what or when. Indirect indicators on the other hand, are more subtle in their indication of incomprehensibility and usually appear as repetitions.
In the one-way task employed by Varonis and Gass (1985), one partner describes a picture that he or she can see to a partner who cannot see it, but who must draw it on another paper. This task does not have a one-way flow of information as would be found in a lecture (i.e., absolutely no two-way exchange of information); instead, one student holds all the information and there is some exchange between the participants. The results showed that more indicators were used by the drawer than by the describer, and more indirect than direct indicators were found overall. It makes sense that the person seeking the information (the drawer) is more likely to negotiate meaning than the person who holds the information (the describer). An interesting finding, however, is that while the describer produced roughly equal direct and indirect indicators, the drawer produced far more indirect than direct indicators of non-understanding. The study also contained a two-way task in which both participants possessed and exchanged information that was necessary for task completion. The information that each participant had was different and complementary to his or her partner’s information. Each participant listened to taped information regarding a robbery and the alibis of 2 out of 4 suspects. During the second part, the partners came together and shared the information that each of them had heard. The shared goal was to determine which alibi was not valid, and thereby which suspect was responsible for the robbery. For the two-way task, there was also a higher frequency of indirect rather than direct indicators. A comparison of the one-way and the two-way results showed a slightly higher average for both types of indicators in the one-way design, although it was not statistically significant. Finally, men gave more indicators than women overall, with a greater percentage of those being direct rather than indirect indicators. There was more of a gender balance of these in the two-way versus the one-way task. These results help to establish the relevance of task and interlocutor as they influence negotiation type and amount.
Gass and Varonis (1985) claim that task familiarity is an important issue and suggest that increased familiarity leads to decreased negotiation. The belief is that as the task is more familiar to the students, there is less information that is unknown or not understood and therefore less of a need for negotiation. Also realized was the importance of a student’s role within the task, as it determined the amount and type of negotiation. In both of the drawing tasks, the drawer had more indicators than the describer. This makes sense because the drawer had a more pressing need to make sure that the information was understood correctly. Another interesting fact was that although the describer had an equal number of direct and indirect indicators, the drawer showed more indicators that were indirect. Results from Gass and Varonis (1985) that showed more negotiation occurring in a one-way versus a two-way task contradict findings by Long (1983). Varonis and Gass explain, however, that there are substantial differences in what is considered a “one-way” task in their study versus that of Long (1983). Varonis and Gass describe their task as less one-way because there is some exchange between the participants, though the information flow is primarily in one direction. Long’s task is strictly one-way with absolutely no exchange between participants and information flowing only in one direction. Varonis and Gass also point out that tasks analyzed in different studies often use different modalities in the tasks that are examined. Examples are a Gass and Varonis (1985) task which goes from linguistic input to drawing output, and a Pica et. al. (1987) study task which goes from linguistic input to object-manipulation output. Comparison of results between studies is difficult when tasks contain differences in the type/amount of information exchange and the mediums of input and output for the task.
In another study, Foster (1998) found an increase in negotiated interaction in dyads performing a two-way task. This particular study, however, noted that many students actually seemed to produce little or no negotiated interaction. Results were so highly variable between the different dyads that the author does not claim any generalization; rather, she suggests that the dyads producing the most negotiations were somewhat atypical instead of representative. Overall Foster claimed that a dyad setting in which students were required to exchange information led to the most negotiations and modified output. An important conclusion, however, was the fact that many students were not inclined to negotiate at all. Foster suggests several possible explanations for this and claims that negotiation of meaning may not be a natural strategy that students employ. Foster suggests that negotiation might be encouraged by using tightly designed information-gap tasks in conjunction with instruction on ‘check, clarify, and modify’ techniques. Also at issue, here is the fact that the data came from a classroom rather than an experimental design. It may be that the classroom environment is more relaxed than an experimental context; therefore, students are more likely to pretend to understand when they do not, hoping that clarification will surface later in the dialogue rather than checking and clarifying at the time of the problem. Students may forego negotiation as a face-saving strategy and may find that if they just wait and listen that they will eventually understand.
Content and Task Familiarity
In any given task, students may have varying degrees of familiarity with the content involved. Whether this affects conversational negotiations was examined by Hardy and Moore (2004) in a study involving twenty-eight pairs of third semester German foreign language students. Also of concern in their study was the role that differing degrees of language support play on negotiation within the task.
Hardy and Moore (2004) investigated student performances on comprehension tasks utilizing material from video segments of German television dramas. Content familiarity was determined by the students’ previous exposure to the plot, characters, and cultural setting of the video segments used. It was hypothesized that a higher degree of familiarity (containing more shared knowledge) would require less negotiation during the course of a collaborative task. Hardy and Moore were unable to support their hypothesis with statistically significant evidence and cite several possible confounding variables within their methodology that might explain why. They also mention that their findings are at odds with those of Varonis and Gass (1985) who determined that a higher degree of shared assumptions did indeed result in less negotiations of meaning because of a decrease in the need to establish common ground.
Comparisons between these two studies, however, are not entirely clear as they focus on slightly different aspects of familiarity and its relation to negotiation. While both of these studies examine the factor of shared knowledge on the part of the participants and its influence on subsequent negotiations, in one the shared knowledge is content material for a task and in the other, it is the speakers’ first language and proficiency level. Additionally, the data used by Hardy and Moore came from classroom language tasks while that of Varonis and Gass came from informal conversations.
Another study by Gass and Varonis (1985) found support for the idea that familiarity with the task type and the interlocutors decreases the occurrence of negotiation, but does not entirely compensate for the need to negotiate for meaning.
Plough and Gass (1993) attempted to determine the effect on NNS interaction of both familiarity in task and interlocutor. In their study, it was found that the students involved in the unfamiliar task showed more involvement in and commitment to the task as evidenced by the number of interruptions. However, the numbers of confirmation checks and clarification requests were higher in the task-familiar group. Plough and Gass stated that by looking at the discourse it appeared that the students in the task-familiar group had become disinterested in the task, possibly because of their familiarity with it. The students seemed eager to complete the task as evidenced by a high number of interruptions and sentence completions. Due to the somewhat uncertain results regarding task familiarity, the authors suggest that a more fruitful area of investigation may be the variables that exist within the interlocutors.
Features of the Interlocutor
An interesting and important aspect of interaction studies is a high degree of variability in the results received from different interlocutor pairings. In their study which looked at collaborative dialogues produced by French immersion students, Swain and Lapkin (1998) found a significant difference in the numbers of LREs or Language Related Episodes between different student pairings. It appeared that some students were more attentive and spent more time on task than did others. Foster (1993) found similar variation in her work. While this is not a surprising fact, it is indeed important to note when qualifying research results and setting up task-related studies. This suggests that some student pairs may work better together than others.
There are many interlocutor variables involved in pair work, which influence the amount and type of negotiation that takes place. Interlocutor familiarity, gender of the participants, similar or different L1 backgrounds, amount of L1 education, and relative L2 proficiency may each play a role.
Interlocutor Familiarity
As mentioned above, Plough and Gass (1993) compared the difference in the type of negotiations that were found between interlocutors that were either familiar or unfamiliar with working together. It was found that more overlapping speech occurred with pairs that were familiar with each other. A possible explanation is that in familiar pairs, simultaneous speech is more acceptable and common because it is less likely to be interpreted as an attempt to dominate the conversation. Although interruptions were found to be roughly the same between familiar and unfamiliar interlocutors, sentence completions appeared considerably more often in familiar pairs. The occurrence of echoes in the discourse seemed to be much more dependent on the type of task than the familiarity between the interlocutors. The phenomenon of back channeling seemed to be a variable of individual conversational style rather than familiarity with the partner. When it comes to confirmation checks and clarification requests, there seemed to be a much greater occurrence in pairs that were familiar with each other. This suggests that familiar interlocutors feel more secure in expressing lack of comprehension and confirming that his or her partner understands. In other words, familiarity with their partner turns the exchange into a non-threatening forum for expressing non-understanding.
Gender
In a study by Gass and Varonis (1985) which examined the variation of negotiation as related to information flow, a further analysis was performed regarding the role of gender. Their study found that men signaled incomprehensible input more often than women and, on average, in a more direct manner than women did. Other studies, however, suggest that the role of gender in pair interactions is of minimal importance. In looking at interactions of mixed gender pairs of adult low-level learners, Topolski (2004) found that gender was not a factor in the pair dynamic in terms of the amount of talking. Topolski compared the amount of talk respectively by males and females in the two conditions of topic-oriented and task-oriented classroom activities. No difference was found in the quantity of talk by males or females in either condition. The divergent results in the two studies above demonstrate a far from conclusive understanding of the role of gender in peer interactions, indicating it is an important area for continued research.
Differences in First Language, Education, and Second Language Proficiency
In a study by Canada (2004) involving 12 dyads of adult low-level learners, the interactions of students with varying levels of similarities across three areas were compared. The students had the same or different L1 backgrounds, L1 educational backgrounds, and relative L2 proficiencies. The interactions were rated as positive, neutral, or negative, using a scale developed by Canada. Canada’s scale was based on four aspects; body language, wait time, scaffolding, and tone of voice/polite speech acts.
Dyads sharing L1s represented slightly higher scores, but positive interaction was also observed in partners who did not speak the same L1. No statistical relationship was found between L1 education level and negotiation score; however, the author cautions that the results may have been affected by the conflation of first language literacy skills with first language education levels. The primary findings related to relative L2 proficiency of the students and similarity in L1. Many of the dyads analyzed appeared to have disparity in the students’ mastery of English, although this was not always reflected in their oral proficiency scores. It was found that of the three dyads that had distance in oral proficiency scores, the two that received positive scores contained students that shared an L1. Along with examples presented by the author, this suggests that sharing an L1 can be a mediating factor between students who have slightly different L2 proficiencies. The author presented examples in which a student with higher L2 proficiency effectively used a shared L1 to scaffold her partner’s learning. It is also noteworthy that the majority of the dyads with positive scores were those who self-selected their partners or were initially not sitting together as opposed to those who were a product of assigned seating. Despite the predominance of positive interactions found in Canada’s data, we can see that sometimes things do not go so smoothly. Canada’s work contributes greatly because it questions what factors affect the quality of interaction and negotiation. In addition to there being many influencing factors, they seem to interact with each other in regards to their relationship with negotiation. Some of these factors are related directly to the interlocutors involved. One aspect of the interlocutor that influences negotiation is the role that learner plays in the pair dynamic. The learner’s personality (i.e., dominant, passive, etc.) and ability (i.e., expert, novice) can be seen to influence the type of interaction that takes place. The learner’s role in the pair relationship and its influence on the interaction are the topic of discussion in the following section.
Interlocutor Relations
It is important to take into account a learner’s personality and language ability when considering how they will participate in pair work with another learner. In a longitudinal classroom-based study, Storch (2002) set out to answer the following three research questions: 1) What patterns of dyadic interaction can be found in an ESL university-level class? 2) Does task or passage of time affect the pattern of dyadic interaction? and 3) Do the differences in the nature of dyadic interaction result in different outcomes in terms of second language development? Storch’s (2002) study took place in a theme-based ESL course at an Australian university. The participants ranged in age from 19 to 42 and all had college-level ESL proficiency, but placement tests had revealed their need for further work in writing and grammatical accuracy. The data from 10 pairs was used in the final analysis. Three language tasks were used to collect data: a short composition, an editing task, and a text reconstruction task. The students completed three similar versions of each task; one to familiarize them with the task, a second to gather pair data, and a third (done individually) to measure the transfer of negotiated information to personal performance. Additional data came from an editing task (used as a pre and post-test), a survey (to determine student attitudes to group/pair work), and the researcher’s observation notes describing student behavior during pair work. Data analysis consisted of two stages: the first stage categorized the student pairs according to their patterns of dyadic interaction, and the second used a process-product approach to develop possible links between these patterns and the students’ language development. Storch observed that the students’ pair work could be described as fitting into one of four patterns. This model, represented in Figure 1 (Storch, 2002, p. 128), describes four unique types of pair relationships with reference to their degrees of equality and mutuality.
Figure 1. Dyadic Interaction Model
high mutuality
4 1
expert/novice collaborative
low equality high equality 3 2
dominant/passive dominant/dominant
low mutuality
Equality is used to describe the degree to which each participant shares in controlling the direction of the task. In a situation with high equality, both people are able to accept the authority and direction of the other. Mutuality describes the students’ engagement with one another’s contributions and is recognized by reciprocal feedback and the sharing of ideas. The four types of pair relationships described by this model were labeled collaborative, dominant/dominant, dominant/passive, and expert/novice.
During Storch’s second stage of data analysis, the students’ performance on the task was compared to the pattern of their dyadic interaction. Three categories were identified regarding task performance: 1) instances suggesting a transfer of knowledge, 2) instances showing no transfer of knowledge, and 3) instances suggesting missed opportunities. Storch found a collaborative or expert/novice pairing to be the most productive in terms of transferring knowledge from pair work into individual performance on the tasks. dominant/dominant relationships were found to be the most inefficient in this regard, yielding the least transfer of knowledge. Finally, dominant/passive relationships resulted in the highest number of missed opportunities for knowledge transfer. These findings highlight the need to take into consideration the nature of pair interactions, and how they are influenced by the personalities of those involved. Storch’s study highlights how the relationships between interlocutors can play a role in the amount and type of negotiation that takes place in dyadic interactions. Students’ relative second language proficiency can also be seen as playing an important part in influencing the type of relationship that is forged in pair work. These findings again highlight the interactive nature of the variables that can influence negotiation in an L2 classroom.
Summary
Work investigating negotiation and interaction in the ESL classroom represents a vibrant and ever-expanding body of knowledge. Numerous factors work together to determine the experience that a learner will encounter in the second language classroom. A great deal has been learned as previous studies have investigated everything from the role of interlocutor to that of task. Also within each major area of influence on negotiation there are many intertwined, contributing variables. Although much has been learned there remains much more to discover.
Previous studies investigating negotiation and interaction have been undertaken from an experimental setting. These laboratory studies provide an important basis for research in classroom second language interaction. Missing however, are sufficient numbers of classroom-based studies that offer an in-depth view of classroom interaction in naturalistic settings.
Historically second language acquisition research has been conducted with several groups of participants including: native speakers working with non-native speakers, young learners, language learners in foreign language settings, and intermediate or advanced English learners in academic settings. A relatively unstudied group is adults who are beginning English language learners. Little work has been done to describe the negotiation that takes place between these learners and determine what factors influence it. In order to fill this gap in SLA interaction research this study will look at student-student negotiations in authentic, ongoing adult low-level, non-academic classrooms. More specifically this study will describe how the quantity and quality of negotiation work is influenced by the amount of language structure in the task and by the aspect of interlocutor.
CHAPTER 3
METHODOLOGY
Research Questions
This research involves an examination of an individual learner in order to address two questions regarding negotiation in pair interactions. First, how does one adult low-level ESL learner’s negotiation change when working on two different task types? Secondly, how does one adult low-level ESL learner’s negotiation change when working with two different interlocutors?
Design
This project is a descriptive case study focusing on how one learner’s negotiation during pair interactions varies with regard to interlocutor and task types. This qualitative analysis takes place within the context of a low-level adult ESL classroom.
Setting
The data for this research project were taken from the Multimedia Adult ESL Learner Corpus (Reder et al., 2003). MAELC is part of an innovative corpus research project being conducted at a Lab School environment located on the Portland State University (PSU) campus. The Lab School () is a collaborative effort on the part of the Department of Applied Linguistics at PSU and Portland Community College (PCC). PSU provides the facilities and research team while PCC provides the ESL instructors and students[1]. The project recently reached the end of the first five years of research and data collection (August 1, 2001—July 31, 2006). MAELC contains approximately 4,000 hours of classroom language and instruction involving roughly 750 learners. Students attending the Lab School are beginning-level adult ESL students of either level A, B, C, or D as defined by the PCC course catalog. The PCC ESL program’s student population consists of refugee residents, immigrants, and U.S. citizens. Four levels of reading, writing, speaking, listening and pronunciation classes are offered. Table 1 below defines these levels in further detail.
Table 1. Definitions for PCC ESL Class Levels
|Level A: |This level is for beginners. Students at this level usually can say their names and addresses. They need |
| |help to conduct day-to-day business and usually have trouble giving or writing personal information |
| |independently. |
|Level B: |This level is for high beginners. Students at this level usually can give information about themselves. They|
| |can use common greetings but usually cannot engage in fluent conversation. |
|Level C: |This level is for low intermediate students. At this level, students can satisfy common communication needs |
| |in daily life. They can ask and respond to questions and initiate conversations. They may need repetition |
| |for unfamiliar topics or when talking about abstractions. |
|Level D: |This level is for the intermediate students. Students at this level can initiate conversations on a variety |
| |of topics. They can express their opinion about immediate surroundings and about more abstract ideas and |
| |concepts. |
Source:
Each Lab School class meets twice a week for three hours each session. Data from all classes is collected for the corpus. A corpus with data from students of these beginning levels is of particular importance as it represents a demographic of ESL learners for which very little SLA research data concerning pair interactions exists.
MAELC is a unique corpus, in that it is comprised of digital and video footage taken from multiple cameras and microphones, which are continuously recording during the course of the ESL classes. The term data clips or simply clips is used to refer to the audio/video recordings of the students’ classroom work; these are recorded using four fixed-mount ceiling cameras. Additionally, two remotely controlled cameras are trained on two select pairs of students whose interactions are also recorded using lavaliere microphones. Activity codes, participation patterns, and transcriptions are applied to this multimedia corpus using specialized software, called Class Action, designed specifically for the Lab School (Reder et al., 2003). Half of the classes are coded for activity and participation pattern. Additionally, two ten minute conversation “bubbles” (Reder et al., 2003) are transcribed, representing work taking place between the student wearing the microphone and the student’s interlocutors. The two student lavaliere microphones record samples from two student pairs in every class. The student pairs that are the focus of recording rotate every class session. Adding further value to MAELC is its extensibility and searchability. Researchers are able to add further layers of coding and transcription to the existing corpus data using the Class Action software.
The Classroom Recording Context
One goal of this research is to strengthen the connection between theory and practice. The naturalistic setting of the classroom environment provides an ideal place for analysis of authentic L2 learning. Previous studies have noted the added authenticity of results that can be gained through classroom-based research (Foster, 1993; Jacob et al., 1996; Swain & Lapkin, 1998). Swain and Lapkin (1998) explain that conducting research in an ongoing classroom context provides a different perspective on the implementation of theoretical principles than research conducted in a laboratory-type setting.
There are, however, drawbacks to classroom-based research, the most common of which is loss of data due to incomprehensibility. As any seasoned teacher will attest, the classroom can often be a noisy and chaotic environment, creating potential problems with recordings of audio data. In the Swain and Lapkin (1998) study, out of a class of 35 students, the data from five pairs were lost as a result of background noise and students’ tampering with the recording equipment. Foster (1993) described having similar issues with data loss. In this present study, data loss did not pose the same threat due to better equipment than that used in previous studies. Additionally, the availability of video of the students’ interactions was necessary for accurate transcription. The Lab School was designed with the recording of pair interactions in mind, and the students and staff have been specifically instructed on the proper use of the lavaliere microphone recording systems. These factors greatly reduced the amount of data loss in this particular study.
Participant Selection
Lab School students are described as low-level adult ESL learners. Student ages range from 17 to 77. The students also represent a multitude of ethnic backgrounds and come from more than 44 different countries. Students also come from a variety of educational and occupational backgrounds. Additionally, length of time in the United States varies from student to student. A unifying factor of these students is their desire to learn English. To answer my research questions, I searched the corpus to find a student who had been at the Lab School for several terms, who had success learning, who was represented in quality recordings of pair work, who was recommended as a focal student by Lab School professionals, and who was represented in a large number of pair work recordings.
1) Length of Attendance: The subject had to have attended the Lab School for several terms (3+). This helped to ensure that the subject had contributed sufficient data for use in my study.
2) Learning Success: The learner’s success was determined by his or her timely progression through the program’s level system. This was a flexible variable given that adult ESL students often have mitigating circumstances (i.e., work and family obligations) that affect the amount of time devoted to academic endeavors, but the desire was to find a student that did not spend more than two terms in one level before advancing.
3) Auditory Quality: To analyze the data accurately it was important to find recordings which were free of recording errors and in which the participants spoke clearly and understandably.
4) Professional Advice: Advice and insights that I received from research coordinators, teacher-researchers, and coder-transcribers at the Lab School were given considerable weight in the selection of a subject, as these people all possessed significant information on the students with regard to my other selection criteria.
5) Recorded Frequency: This was the most significant criterion in the choice of a candidate for research analysis. The student had to have been recorded an ample number of times wearing the microphone or sitting next to someone wearing the microphone in order to have provided a sufficient collection of data clips for analysis. A large number of recorded interactions would likely mean a greater number of interlocutors that the student had worked with during a greater variety of tasks. Greater variation in task and interlocutor would mean that I would have more flexibility in data selection for my final analysis.
After taking into account all the factors listed above, a focal student was selected. The student is a middle-aged Chinese woman who was deemed as a successful student and had attended the Lab School for several terms. A pseudonym will be used to refer to the Lab School students to protect anonymity. The focal student will be referred to as Jin.
Data Selection Criteria
For this research, I chose to focus on the negotiation work and dyadic interaction pattern of one student and how that did or did not change in relation to partner and task. To define these relationships, I looked at Lab School data to find separate recordings of Jin working with same partner but in different types of tasks. I also searched for instances when Jin was working on the same kind of task, but with different partners.
Several variables limited and guided the data selection process. First, I chose to look at clips only from level B classes, there were three reasons for doing this. First, by taking clips from only one level, I was able to ensure that the clips showed students with roughly the same proficiency levels. Comparing clips from different skill levels would have been problematic because it would have introduced skill level as a potential factor in influencing the negotiation and interaction type. By taking clips from only one level, I attempted to avoid this issue. Secondly, the majority of the pair work analyzed thus far with Lab School data has been with level A therefore, research performed on level B data would provide a more useful addition to the existing body of work. Finally, level B contained the greatest number of days when Jin, the focal student, was wearing the microphone or seated next to another student who was. This factor allowed for the largest amount of potential data for research. The number of days that Jin wore the microphone or sat next to a partner who did throughout her time in the three Lab School levels are as follows: Level A = 14 days, Level B = 24 days, and Level C = 12. Selecting level B data for analysis maximized the number of useable clips.
There were four additional factors determining the selection of clips used in the final analysis. These factors were coding, transcription, variability in task type/interlocutor, and length of the pair interaction during the task.
Of all classes recorded at the Lab School, one-half receives activity and participation pattern codes. Every recorded class session also gets two ten-minute segments transcribed, one for each student wearing a microphone. For the purpose of this research, I was interested in pair work; therefore, I looked at work that was or would be given the participation pattern code of Pair. The activity codes are used to help describe what the majority of the class is actually doing at any given time during the class period. These activity codes are broken into three groups: prompt, information, and language. Regarding activity codes, my main concern was with the language coding as this described the amount of language support provided by the teacher for the students to complete the task. In addition, they describe how much language the students are expected to contribute for each task. Language codes are given for the language that comes from both the teacher and the students and can be thought of as the degree to which the task contains a script for the students to follow. The language codes, which help describe the degree of support within an activity, can be thought of as existing on a continuum.
In determining which language codes to seek, I considered which represented the greatest difference in the scriptedness of the task and took into account the other factors in my data selection process. This represents variability in the task and interlocutor because it seeks occasions when the focal student was working with other students within two task types, which were as dissimilar as possible. The combination yielding the highest number of examples of transcribed interaction with distance in language codes was Language Frame/Target Item (LF/TI) and Language Frame/Question Answer (LF/QA). A task coded as Language Frame/Target Item has a high degree of embedded language support and is less open-ended with regard to the potential correct responses by the students. A language frame is some discourse, provided by the teacher, which has missing information (target items) that is to be produced by the student. An example of this would be an error correction exercise, in which the teacher writes sentences with errors on the board and asks the students to identify and correct those errors. Another example of this would be a task in which the teacher provides a dialogue for a role-play conversation that includes missing pieces of information that the student must fill in as they use the script to practice conversation. In a task coded Language Frame/Question Answer the teacher also provides support with a language frame for questioning, but the students have a wider range of possible correct answers. The teacher provides examples that give a language frame for asking questions while the students have little scriptedness regarding their possible answers. An example of this would be a task practicing giving and receiving directions in which the teacher provides a language frame for asking about the location of different buildings (i.e., “Is there a bookstore near here?”). The students need only provide name of different locations to complete the language frame question, however the language provided by the students in answering is more open-ended. The students use a map drawn on paper to answer the questions. The students give their partner an answer that can vary depending on the location and route they choose. An example answer may be; “Yes, there is. Walk four blocks and turn left.”, but the important part to remember is that there may be many different, yet correct answers. Therefore, in this coding arrangement, the teacher supplies language that is highly scripted, but the language that the students use to answer is far less scripted and much more variable with regard to possible correct answers.
Ultimately, of primary importance in my data selection was the difference between the two task types of Language Frame/Target Item and Language Frame/Question Answer. These task types are different in the freedom that the students have in possible responses. When asking how negotiation might vary depending on task I am more specifically questioning the disparity in this particular aspect of the tasks.
Pilot Study
A pilot study was performed, with a learner from the Lab School, in preparation for this research project. This work was used to get an idea about what features I might want to consider for the second phase of my data analysis. I examined how a student (not the focal student) interacted with her partner in completing tasks and noted the important features their pair work. While watching the recordings I categorized the more frequently observed of these features. Below is a brief summary of these.
Negotiation: Negotiation moves were seldom explicitly stated, but instead executed through body language, repetition, and partial repetition.
Body Language: The focal student and her interlocutor often used body language to signal understanding, agreement, and lack of understanding. Additionally, the students often pointed to language located elsewhere in the classroom, for example the whiteboard, to help clarify the task. Body language often serves as an important function of negotiation for the beginning language learner.
Engagement & Mutuality: In this category, I observed the focus and attention the partners provided each other, as well as how responsive they were to each other’s input.
Cooperation: It seemed that some pairs were, in general, more cooperative than others. The interactions in which the learners were more cooperative and engaged with each other were more productive and efficient.
Task Roles: Most of the activities in the pilot study were unbalanced. Frequently one student exhibited more control over the completion and direction of the task and provided more corrections of her partner’s utterances.
Data Analysis
Information gained from the pilot study led to a two-part data analysis. The pilot study showed that both the number of negotiation features and how the partners related to each other were essential in evaluating students’ pair work. The following section will explicate the two phases of data analysis used in this research study, Negotiation and Collaboration.
Phase One: Negotiation Framework
The clips of interaction in my study underwent two phases of analysis. The data was initially analyzed using the framework for Negotiation of Meaning put forth by Long (1983) and later elaborated on by Gass and Varonis (1985). Comprehension checks, confirmation checks, clarification requests, self repetitions, and other repetitions were tallied for comparison. Untriggered other repairs and recasts were also counted. A comprehension check is something said by the speaker to ensure that the listener has understood the previous statement. A confirmation check is something said by the listener to verify that they have correctly understood their partner’s previous utterance. A clarification/request is used by the listener to seek further elucidation of their partner’s previous utterance. In a self repetition, a speaker repeats himself or herself and in an other repetition, one speaker repeats another. An untriggered other repair happens when one person corrects another without a being asked to do so. A recast is a type of untriggered other repair and is performed when one person corrects another’s mistake by repeating it, but in the correct form. Recasts don’t necessarily contain an explicit statement regarding the presence of an error.
Data analysis began by simply counting the number and types of negotiation. I looked for noticeable differences in the number and types of negotiation performed with each interlocutor.
Phase Two: Mutuality and Equality
In constructing the second phase of my data analysis, I reviewed the information gathered from my pilot study, reflected on previous studies which had examined similar interactions, and considered how I might best provide answers for my research questions. The first phase of my analysis was meant to account for the amount and type of negotiation features; therefore, in the second phase I sought a framework that would help me evaluate the relationship between the interlocutors.
To further describe how the student in my study varies her negotiation according to her partner when performing the same type of task, I utilized a model put forth by Storch (2002). I chose this model of dyadic interaction because it parallels much of the same phenomenon encountered during the course of my pilot study. Additionally, the research that employed this model was guided by a sociocultural framework and focuses on what takes place between the learners as part of the learning process. This was valuable in determining how the role of interlocutor influenced the negotiation, because it focuses on the dynamics of the interaction between the learners as part of the learning process.
As discussed earlier, Storch’s (2002) model describes four distinct patterns of dyadic interaction which are identifiable in terms of equality and mutuality. The four patterns are labeled collaborative, dominant/dominant, dominant/passive, and expert/novice. In collaborative pair work the students work together on all parts of the task. The students respect and engage each other’s contributions. Neither of the students dominates or takes an unequal amount of control over the direction and completion of the task. A dominant/dominant interaction happens when the both students desire to take control and is marked with disagreements and an inability to reach consensus. This pattern may have high equality if it contains a division of labor. A dominant/passive pattern is one in which one student gives directions and assumes an authoritative role while the other is passive and subservient. There is little disagreement in this kind of a relationship; however, there is also little negotiation. The final pattern of expert/novice occurs when a more proficient student and a less proficient student are coupled and the more proficient assists and guides their partner during task completion. This scenario may be lower in terms of equality, but higher in terms of mutuality.
Combined with a description of the negotiations of meaning and their focus on input to the learner (Phase 1), an analysis of observed interaction patterns (Phase 2) yielded a description of the learner’s behavior used to answer my research questions as well as suggest avenues for further research.
CHAPTER 4
RESULTS
The following section contains the results found during the data analysis stage of my research. I will give a brief description of each clip including the participants and task parameters. After giving pertinent information regarding the clips I will display the numbers of features of negotiation, the location of the interaction along the collaboration continua, as well as evidence supporting its placement along the collaboration continua. Finally, I will make comparisons between clips with regard to the language support in the task and the interlocutor.
The data selection criteria listed above yielded four clips for final analysis. The comparisons of these clips allowed for a sufficient depth of descriptive analysis to speak to my specific research questions. Figure 2 below shows the different clip combinations that will be compared in this study
Figure 2. Clip Comparison Combinations
|Interlocutor/ |Aracelia |Inez |
|Task Type | | |
| |Sibling Similarities | |
|Language Frame/ |collaborative | |
|Target Item |--------------------------------- | |
| |Personal Info Exchange | |
| |collaborative | |
| | | |
|Language Frame/ Question-Answer | | |
| |Giving/Receiving Directions |Favorite Family |
| |expert/novice |Member |
| | |collaborative |
Clip #1: Personal Info Exchange
Info Exchange
The first clip was recorded on the second class session of the summer quarter in 2003. Given the nature of the task, this clip has been given the title of Personal Info Exchange. In this clip, Jin is working with Aracelia, a middle-aged Mexican woman whose native language is Spanish. The task that the two individuals are undertaking has been coded as Language Frame for the language provided by the teacher and Target Item for the language provided by the students. Remember a language frame is a form that the teacher constructs in which the students are to supply the answer, or target items. A task coded as language frame/target item has a high degree of internal language support (amount of scriptedness). The student language for this type of activity comes from a restricted set of possibilities and therefore limits the students need to produce language independently. The length of this task is 7 minutes and 19 seconds. The task consists of the students using questions written on a worksheet to guide them through the act of verbally obtaining basic personal information from their partners. The students have been instructed to ask their partners a list of questions from the worksheet, and when their partners respond they are to record the information on the paper. Both students are expected to ask the questions, and both are expected to record the other person’s answers. The students begin this activity almost immediately after the teacher finishes giving the instructions. There is no deliberation regarding how to perform the activity or who will begin. The students appear as if they both understand well what is to be done. Below is a list of the questions from worksheet that the students asked each other during this task.
1. What’s your name?
2. Where are you from?
3. What country do you live in now?
4. What state do you live in now?
5. What city do you live in now?
6. What’s your address?
7. What’s your telephone number?
8. What’s your first name?
9. What’s your last name?
10. What’s your social security number?
11. What’s your job?
12. Are you married?
13. Do you have children?
14. Do you like living in Oregon?
Clip #2: Sibling Similarities
Similarities
The second clip involves the same two students working on a task coded as Language Frame/Target Item. This interaction was recorded on the ninth class session of the Summer quarter in 2003. This clip was named Sibling Similarities in reference to the task. Prior to this activity, the students listened to a taped monologue in which a woman described the similarities and differences between herself and her sister. Students listened to the recording two times and then, as directed by the teacher, they discussed some of the information given on the tape. The recorded passage was also supported by a written version appearing in the students’ textbook. The instructions given by the teacher were; “Tell your partner about you and your brother or sister. Are you the same or different?” Although the question provided by the teacher seems somewhat open-ended it received the code of language frame because of the written support provided by the book and the audio recording. The book contained the same information that was in the taped monologue, which was a woman describing the similarities and differences between her and her sister. The following is a portion taken from the taped monologue: “My sister and I are very different. I’m a teacher and she’s a journalist. I live in Chicago. She lives in Paris. I have a small house in the suburbs. She has a large apartment in the city.” This language gave the students support by modeling an example and providing a framework within which they could insert the appropriate language that applied to themselves and their siblings. The students did not seem to have any difficulty getting underway with this activity and there was no discussion as to who would share information first. This kind of a beginning to the activity helps to show equality as neither student is assigning roles to the other. Jin began by telling about herself and her sister, and Aracelia accepted this opening without question. When Jin had exhausted her information regarding her sister, she prompted Aracelia to begin sharing information by looking at her and saying “you?” This activity was slightly shorter than the others, lasting only 3 minutes and 38 seconds. Towards the end of the activity the students seemed to deviate slightly from the teacher’s assignment and began speaking about themselves without reference to how these things compared to their siblings. Although the students were not following the planned activity exactly the way the teacher had intended, the tangent that they took led to more verbal interaction and negotiation.
Clip #3: Giving/Receiving Directions
Receiving Directions
The third clip was recorded on the 27th class session of the Summer quarter in 2003 and is approximately nine and a half minutes in length. In reference to the task involved this clip was labeled Giving/Receiving Directions. This video clip shows the Jin and Aracelia working together again. In this activity, the two partners have handouts of paper maps with common businesses located on them. The students are told to practice asking for and giving directions to one another. The Language codes given to this activity are Language Frame for the teacher and Question/Answer for the students. Compared to Target Item, the students have to use more of their own language and less of a script in a Question/Answer task. The students do not deviate from the Language Frame question provided by the teacher. The teacher provided the language frame question: “Is there a _____ around here?” An example answer given by the teacher was, “Yes, there is. Walk three blocks and turn left.” The teacher modeled the process of asking for and giving directions several times in front of the class before the partners begin work. The modeling performed by the teacher prompted the students to ask about the location of different places, thereby leading the teacher to model further proper responses. The language frame to be used in this exercise was also presented in a jazz chant activity immediately preceding this activity. Additionally the students have a handout with the language frame. Both Jin and Aracelia are called upon by the teacher to ask a question during the modeling period thereby assisting in the demonstration of the activity. The total time the two partners spent working on this activity is 6 minutes and 14 seconds. It is noteworthy that the pair work was interrupted at two different points, once by the teacher and once by another student. This time did not count towards the total interaction time documented for this research. As seems to be the case with many student activities at this level, there appears to be little to no preplanning of the activity by the partners with regard to who will ask first and who will answer first. This shows equality in the sense that neither student is controlling task authority. In this activity, Jin begins by asking the first question. In total, Jin asks four questions of Aracelia, and Aracelia asks five questions of Jin. Prompts for the questions that the students were supposed to ask each other were provided on a handout. Below are listed the questions that the students ask each other throughout the course of this task.
|Jin asking Aracelia |Aracelia asking Jin |
|1. Is there a bookstore near here? |1. Is there a Chinese restaurant near here? |
|2. Is there a parking lot near here? |2. Is there a Dairy Queen near here? |
|3. Is there a gas station near here? |3. Is there an Italian restaurant near here? |
|4. Is there a post office near here? |4. Is there a school near here? |
| |5. Is there a park near here? |
Clip #4: Favorite Family Member
Family Member
The fourth clip analyzed took place on the 12th class session of the Winter quarter in 2003. This clip was given a name in reference to the activity the students were performing. The name for this clip is Favorite Family Member. This clip involves a task in which Jin is working with a different partner, a middle-aged Mexican woman named Inez. The native language of Jin’s partner is Spanish. Language codes given to this exercise are Language Frame/Question-Answer. This clip shows the second part of a multi-part activity. The total time for this clip is 9 minutes and 49 seconds. In this activity, the students are supposed to talk to another student and discuss their favorite family member. During this first part of the task, the students are not supposed to take notes and are only to interact verbally. The students are told to remember the things that their partner tells them, because they will then retell these things to a different student during the second part of the activity. This clip shows the work done by these partners during the second part of this activity. The students in this clip are supposed to be telling each other what they heard their first partner say about a favorite family member. Jin and Inez begin with this, but very quickly they change direction and begin asking and telling each other about their respective favorite family members rather than what they had heard the other student describe. In other words, they revert to the initial part of the activity rather than complete the secondary retelling phase. Another thing these students change about the activity is that they are writing down the answers that their partner gives rather than just listening, as they were instructed. The following questions are those that the students are supposed to ask each other about their favorite family member.
1. Who is your favorite family member?
2. What is his or her name?
3. Where does he or she live?
4. What does he or she look like?
5. What is his or her personality?
Results from Phase 1: Negotiation Features
All of the clips were analyzed with regard to the amount and type of negotiation performed therein. The following features were recorded in Table 2 below: clarification requests, confirmation checks, untriggered other repair, self repetitions, other repetitions, and recasts.
Table 2. Negotiation Features
| |L.Frame/ T.Item |L.Frame/ T.Item |L.Frame/ |L.Frame/ |
| |Personal Info. | |Q-A |Q-A |
| |Exchange |Sibling Similarities|Giving/ |Favorite |
| |(Aracelia) |(Aracelia) |Receiving Directions|Family |
| | | |(Aracelia) |Member |
|Jin and ( | | | |(Inez) |
|Clarification Requests | 8 |2 |5 |14 |
|Confirmation Check |19 |11 |6 |19 |
|Comprehension Check | 1 |1 |1 |0 |
|Untriggered Other Repair | 0 |0 |8 |3 |
|Self Repetition (# of words) | 36 (63) |23 (36) |44 (83) |65 (86) |
|Other Repetition | 44 (81) |18 (25) |39 (64) |36 (46) |
|(# of words) | | | | |
|Recast | 1 |0 |1 |0 |
|Total Words |A=178 |A=131 |A=139 |I=224 |
| |J=338 |J=143 |J=256 |J=271 |
| |(516) |(274) |(395) |(495) |
|Total Time |7 m. 19 s. |3 m. 38 s. |7 m. 30 s. |9 m. 49 s. |
* For definitions of negotiation features see pages 18 & 19.
Results from Phase 2: Collaboration Continua Placement
The second phase of analysis was to determine where each clip should be placed along the collaboration continua. Table 3 summarizes this information and gives a dyadic interaction pattern label for each clip.
Table 3. Degree of Collaboration
| |L.Frame/ T.Item |L.Frame/ T.Item |L.Frame/ |L.Frame/ |
| |Personal Info. Exchange |Sibling Similarities |Q-A |Q-A |
| |(Aracelia) |(Aracelia) |Giving/ |Favorite |
| | | |Receiving Directions|Family |
|Jin and ( | | |(Aracelia) |Member |
| | | | |(Inez) |
|Collaboration Continua | collaborative |collaborative |expert/ |collaborative |
|Placement | | |novice | |
Support for Placement along the Collaboration Continua
The next section provides the evidence supporting each clips specific placement on the collaboration continua. The framework from Storch (2002) was used as a guide in making these determinations. These lists are not to be viewed as exhaustive inventories of the features found in each clip, but rather as primary points supporting their categorization.
Personal Info Exchange
Personal Info Exchange was determined to be collaborative based upon its negotiation features and the quality of the interaction between the interlocutors. This interaction showed moderate to high levels of equality and mutuality between the participants. Table 4 below gives the evidence found to support this clip’s categorization as collaborative.
Table 4. Highlights from Personal Info Exchange
|Personal Info Exchange |
|Language Frame/Target Item |
|(Jin and Aracelia) |
|collaborative |
|data from clip |evidence regarding collaboration |
|High number of negotiation features, specifically in the |This shows that the students are engaged with one another’s |
|form of confirmation checks and repetitions. |contributions and therefore have a high level of mutuality |
| |in their interaction. |
|Equal control over task completion and direction. Neither |This mutual control gives evidence of a high degree of |
|participant is forcing her will over the direction of the |equality, therefore precludes a dominant/passive pattern. |
|task. | |
|Lack of argumentative or confrontational dialogue on the |The absence of these features rules out this clip’s |
|part of the participants. |placement in the dominant/dominant realm of the |
| |collaboration continua. |
|Seemingly comparable English proficiency levels. |These students cannot be categorized as expert/novice |
| |because of their greatly similar language ability. |
Sibling Similarities
Sibling Similarities was also categorized as collaborative. Table 5 below lists the key features leading to this categorization and the evidence that each provides.
Table 5. Highlights from Sibling Similarities
|Sibling Similarities |
|Language Frame/Target Item |
|(Jin and Aracelia) |
|collaborative |
|data from clip |evidence regarding collaboration |
|High number of negotiation features, specifically in |This shows that the students are engaged with one another’s |
|the form of confirmation checks, repetitions, |contributions and therefore demonstrating a significant degree |
|reformulations, and phatic utterances. |of mutuality. |
|Equivalent control over task completion and direction. |Demonstrates a high degree of equality, therefore rules out a |
|This was shown by neither student assigning roles or |dominant/passive categorization. |
|taking control over task direction or completion. | |
|Lack of argumentative or confrontational dialogue on |The absence of this type of speech in the interaction helps |
|the part of the participants. |rule out a dominant/dominant categorization. |
|Comparable English proficiency level. |The students’ similar proficiency levels disqualify this |
| |interaction from being a representation of an expert/novice |
| |interaction. Neither student appears to be doing more work in |
| |this interaction. |
|Large amounts of related information exchanged. |This is another example of the mutuality found in this |
| |interaction. The students are engaged with each other’s |
| |responses and respond accordingly. |
Giving and Receiving Directions
“Giving and Receiving Directions” was determined to be expert/novice with regard to the collaboration continua. Below, table 6 illustrates the evidence provided for making this categorization.
Table 6. Highlights from Giving and Receiving Directions
|Giving and Receiving Directions Language Frame/Target Item|
|(Jin and Aracelia) |
|expert/ novice |
|data from clip |evidence regarding collaboration |
|Unequal student English proficiency levels in relation to |This supports the expert/novice classification on the |
|this activity as evidence by disparate numbers of |grounds that one person (the expert) would be noticeably |
|clarification requests and untriggered other repairs. |more advanced than the other (the novice). During this |
| |interaction Aracelia clearly provides more help to Jin than |
| |Jin does to Aracelia. |
|Supportive and assistive role taken on by Aracelia which is|An expert/novice relationship is partially defined by the |
|seen in the ways she provides clarification and untriggered|helpful and supportive way the expert guides the novice |
|other repairs. |through the task. |
|Lack of features (i.e., long monologues by one student who |A higher degree of mutuality is seen in this interaction |
|assumes near-total ownership of the task) that would show a|than would be observed in a dominant/passive pattern. In |
|dominant/passive pattern which is in many ways the most |other words, the more advanced student draws in the novice |
|similar to that of the expert/novice. |student rather than completing the task on her own without |
| |input from the student of lesser proficiency. |
Favorite Family Member
Favorite Family Member was categorized as collaborative as evidenced by the following features. Table 7 below shows evidence supporting this categorization.
Table 7. Highlights from Favorite Family Member
|Favorite Family Member |
|Language Frame/Question-Answer |
|(Jin and Inez) |
|collaborative |
|data from clip |evidence regarding collaboration |
|Prevalence of laughter throughout the interaction. |A sociable/friendly dialogue is often seen in collaboration.|
|Equal proficiency levels of the partners as shown by equal |Lack of imbalanced proficiency levels helps to rule out |
|amounts of errors and corrections. |expert/novice and, to a lesser extent, dominant/passive |
| |categorizations. |
|High degree of mutuality as demonstrated by frequency of |Moderate to high degrees of mutuality are essential in a |
|repetitions, clarification requests, confirmation checks, |collaborative interaction and show that the interlocutors |
|and extension/reformulation of previous utterances. |are actively engaged with each other’s contributions. |
|Neither student exhibits unequal control over the |Moderate to high degrees of equality are always found in a |
|progression and/or completion of the activity. Neither |collaborative dialogue. Presence of this level of equality |
|student demands the task be done a certain way or tells the|rules out categorization as dominant/passive. |
|other how to complete the task. | |
Discussion of Results Found & Comparisons between Clips
The research questions under consideration in these comparisons are “How does one adult low-level ESL learner’s negotiation vary with regard to task when working with the same interlocutor?” and “How does one adult low-level ESL learner’s negotiation vary with regard to interlocutor when working on a similar task with the same degree of language structure?” In the following section, I will answer these questions by comparing the results found from different clips in which one student is working with the same interlocutor on different tasks, with different interlocutors on very similar tasks, and with the same interlocutor on very similar tasks. I will briefly describe each of the activities, make comparisons with regard to negotiation features, make comparisons regarding the amount of collaboration found, and discuss ideas that may explain the differences found.
Similar Interlocutor / Dissimilar Task
The first two clips to be discussed here contain tasks in which Jin is working with the same interlocutor named Aracelia. The clip titled Personal Info Exchange has been coded Language Frame/Target Item for the teacher and students respectively. This coding represents a high degree of language structure in which the students must provide a specific, finite answer. In this clip, students are practicing asking each other a list of questions about basic personal information from a worksheet. After asking for the information, the students are to record their partner’s responses on the worksheet. An example from this task is question #4 which asks “What state do you live in now?”. Below the question is the language frame “I live in _______ .”. In completing this task the person answering only needs to supply the target item (in this example the word Oregon) which is then inserted into by the partner into the language frame on the worksheet. The language frame and the concreteness of the answers help to support the students in completing this task.
The second clip labeled Giving/Receiving Directions has been coded Language Frame/Question-Answer. This language code represents a lesser degree of language structure for the students with regard to the language they must provide for task completion. In this clip the students are practicing giving and receiving directions using a paper map. As in Personal Info Exchange, this task uses a language frame provided by the teacher. In this example the language frame is “Is there a _____ near here?”. The student asking the questions is required to input the name of a building when asking. The student being asked is then supposed to answer by stating “Yes, there is.” or “No, there isn’t.” If the answer is yes then the student must also verbally give directions on how to get there from the starting location. An example question and answer taken from this task are: Q: “Is there a Chinese Restaurant near here?” A: “Chinese restaurant? … Chinese oh yeah … go straight two block … two block um two block and uh half block turn left right?”. It becomes obvious in looking at this answer, that the student experienced some degree of difficulty and uncertainty in answering. In this activity, a lesser degree of language support is found in the open-endedness of the questions and the answers that the students provide. The students should be asking about buildings that exist on the map that they have in front of them therefore the answer to the first part of the question should always be yes. The second part of the question however, may have many correct answers because there are many different ways to move from the starting location to the different buildings located on the map. It is the fact that Giving/Receiving Directions creates the possibility for multiple correct answers that makes it less structured and thereby more difficult than Personal Info Exchange, which asks the students to provide only one concrete and concise answer.
Differences in Negotiation
The first evaluation will be of the features of negotiation. Below, Table 8 allows for side-by-side comparison of these as they appear in the two clips. The features of clarification requests, confirmation checks, comprehension checks, self repetition, and other repetition have been widely discussed in previous research. In this study I have also counted untriggered other repairs and recasts. A better sense of the students’ interaction with each other’s responses and of their relative proficiency levels are gained by accounting for these additional features.
Table 8. Personal Info Exchange Compared with Giving/Receiving Directions
|Aracelia & Jin |Lang. Frame/ Target Item |Language Frame / Question |Difference |
| | |Answer | |
| | |Giving/Receiving | |
| |Personal Info Exchange |Directions | |
|Total Words |A = 178 |A = 139 |A = 39 |
|A = Aracelia |J = 338 |J = 256 |J = 82 |
|J = Jin |(516) |(395) |(121) |
|(combined words) | | | |
|Clarification Requests | 8 |5 |+ 3 Personal |
| | | |Info Exchange |
|Confirmation Check |19 |6 |+ 13 Personal Info Exchange |
|Comprehension Check | 1 |1 |0 |
|Untriggered Other Repair | 0 |8 |+ 8 Giving/Receiving |
| | | |Directions |
|Self Repetition / | 36 (63) |44 (83) |+ 8 (20) Giving/Receiving |
|(# of words) | | |Directions |
|Other Repetition / | 44 (81) |39 (64) |+ 5 (17) |
|(# of words) | | |Personal Info Exchange |
|Recast | 1 |1 |0 |
|Total Time |7 m. 19 s. |7 m. 30 s. |N/A due to interruptions |
|Interaction Pattern |collaborative |expert/novice | |
* (information taken from Tables 3 & 4, page 67-68)
When comparing the amount of negotiation, many differences can be observed. Some of the most noticeable differences are found in the amount of confirmation checks and untriggered other repairs. Other differences of lesser magnitude also appeared in clarification requests and repetitions (self and other).
The overall number of confirmation checks was greater in Personal Info Exchange than in Giving/Receiving Directions. The reasons for this can be explained by the distinctions between the two tasks. In Personal Info Exchange the information being exchanged has two noteworthy qualities. The information the students are asking each other for is: 1) of a personal nature, known by the person being asked; and 2) must be recorded in written form by the person asking during the course of the activity. Due to these two factors, there is greater need for the students to ensure accuracy in their answers. They do this by performing confirmation checks. Due to the personal nature of much of the information, the students are not able to breach any gaps in their comprehension with an educated guess. Therefore, when the students are not quite sure that they have the answer correct, they often repeat it back, as they are writing, with a rising intonation to indicate a confirmation check. In addition, the need to record the answer in writing forces the students to attempt to be more accurate in their reception of the answers. The increase in this type of confirmation check in turn increases the amount of repetition occurring in this interaction.
In contrast to the task in Personal Info Exchange, the task in Giving/Receiving Directions does not require the students to ask for and receive personal information nor to record the answers in written form. This results in less need to confirm the accurate comprehension of the answers. In Giving/Receiving Directions, the students are using a map of a town center to practice asking for and giving directions. The students have some common knowledge between them because they both have copies of the map and the list of buildings/businesses. The information given and received here is different from that in Personal Info Exchange in that it is not personal to one student and foreign to another student, such as a home address or a telephone number. In addition, during the task in Giving/Receiving Directions the students are only required to exchange the information verbally and are not asked to record anything in writing. This allows the students to have more tolerance for a less than complete understanding and thereby leads to fewer confirmation checks. This leaves the opportunity for the students to make an educated guess rather than seeking certainty in the answer because they are not required to put it in writing. The requirement for a less structured and exact answer in Giving/Receiving Directions explains the occurrence of fewer confirmation checks and repetitions.
In the category of untriggered other repairs, there is a striking difference between Personal Info Exchange and Giving/Receiving Directions. Giving/Receiving Directions contains eight untriggered other repairs while Personal Info Exchange contains none. When looking at Giving/Receiving Directions it seems that Aracelia’s proficiency in English, concerning the demands of this task, is higher than that of Jin. It is likely that a slight difference in proficiency levels also exists in the students during Personal Info Exchange, but the demands of the task and the personal nature of the questions are highly structured enough to hide these differences. The language provided by the students in Personal Info Exchange consists of finite answers. In Giving/Receiving Directions there is less language support provided because the students are asked to produce questions and answers, though there may be several ways of doing this, or multiple correct answers. Overall Personal Info Exchange provides more language support than Giving/Receiving Directions. Of the eight untriggered other repairs found in Giving/Receiving Directions, seven of these occur when Aracelia is correcting Jin and only one occurs when Jin is correcting Aracelia. The fact that Aracelia is responsible for the majority of the untriggered other repairs suggests that her proficiency level is greater, at least with regard to the task in Giving/Receiving Directions. These untriggered other repairs on the part of Aracelia help define this interaction pattern. Excerpts 1 & 2 below are some examples of untriggered other repairs found in this interaction. Highlighted text identifies untriggered other repairs. Additionally Aracelia also performs one recast. In recasting Aracelia mimics the format that a teacher would use in speaking to a student who does not understand. She does this by adjusting her rate of speech to a much slower one in order to aid in comprehension.
Excerpt 1.
Jin: uh is they_ is there uh (+) car parking (+) car parking (1) parking
Aracelia: parking lot
Jin: parking lot.
Excerpt 2.
Jin: uh Dairy Queen uh go straight one block and uh turn right uh go one block and uh turn (+) half turn (1)
Aracelia: turn left
Jin: turn left uh ha_ and half (+) walk uh half (+) wa_ walk (1) xxx?
Aracelia: um hm ((nods head)) (1) is there um (+) Italian (+) restaurant near here
Jin: Italian Italian res_ *restauran *restauran (1) Italy Italy oh oh (1) go straight one block turn le_ right (+) turn right uh two block uh one block and uh turn right (2) and uh
Aracelia: walk ((points at board))
Jin: walk yeah
Aracelia: half a block
Jin: half block yeah turn right
As noted earlier, other differences found were more clarification requests in Personal Info Exchange, more other repetitions in Personal Info Exchange, and more self repetitions in Giving/Receiving Directions. The numbers for these were relatively small compared to confirmation checks and untriggered other repairs, and therefore will not be talked about in great length. It is worth mentioning, however, that when we look at the features of negotiation as a whole there seems to be slightly more negotiation taking place in Personal Info Exchange. It must be considered when making this claim that although there are more untriggered other repairs in Giving/Receiving Directions, these are not a dialogic process in the way that checking for and providing confirmation are.
Comparative Placement on the Collaboration Continua
Aside from differences in the amount and types of negotiation, it was also found that these interactions differed in their degree of collaboration. It should be noted that there is a strong relationship between the placement along the collaboration continua and the quantity and quality of negotiation features found in each of the clips. These two levels of analysis, by their nature, contain overlapping categories. Personal Info Exchange was given the label of collaborative with regard to its placement along the collaboration continua, whereas Giving/Receiving Directions was labeled as expert/novice (Aracelia being the expert and Jin being the novice). In thinking about the issue of overlapping categories of analysis, it should be noted that Personal Info Exchange contained a greater amount of negotiation and also contained more collaboration. Personal Info Exchange was judged as collaborative because moderate to high mutuality and equality can be observed between the two partners. High mutuality was observed through the large numbers of confirmation checks and repetitions. Both partners seemed to be equally involved with each other’s answers. In addition, equality was observed by the fact that neither partner seemed to exert her will any more than the other as to how the task should be completed. This task progressed rather smoothly in an “ask, record, and repeat” fashion. In contrast, Giving/Receiving Directions was categorized as expert/novice. In Giving/Receiving Directions both students possessed relatively equal control and authority over the progression of the task, however it is evident the students weren’t equal in terms of proficiency and thereby mutuality. When viewing this clip it becomes obvious that the task was much more difficult for Jin than it was for Aracelia. In response to Jin’s many problems during the task, Aracelia was very helpful and supportive. Aracelia’s role as expert is evidenced by her many untriggered other repairs to correct Jin’s mistakes. Additionally, the majority of clarification requests are due to breakdowns in the communication from Jin’s side of the dialogue. The way in which Aracelia reacts to Jin’s many difficulties is notably different from the type of reactions that may appear in a dominant/passive scenario. Aracelia is patient and encouraging. It can be seen in several instances that Aracelia patiently waits and gives Jin plenty of time to find the answer on her own rather than rushing in and quickly telling her the answer and moving on. In addition to being patient, Aracelia is supportive and encouraging. Aracelia often uses phatic utterances to praise Jin when she has answered correctly. Excerpts 3-5 below are a few examples of Aracelia using phatic utterances to support Jin after answering correctly. Highlighted text identifies phatic utterances.
Excerpt 3.
Jin: turn left uh ha_ and half (+) walk uh half (+) wa_ walk (1) xxx?
Aracelia: um hm ((nods head)) (1) is there um (+) Italian (+) restaurant near here
Excerpt 4.
Jin: half block yeah turn right
Aracelia: um hm is there a school near here?
Excerpt 5.
Aracelia: is there a school near here?
Jin: school near here *cool near here (5) school? (+) um xxx!
Aracelia: um hm ((nods head)) walk_ yes there is walk half a block
It is the difference in proficiency levels which leads to unequal mutuality, and it is Aracelia’s supportive role which defines this interaction as having an expert/novice pattern. It could be the case that the reduced language support embedded in this task (Language Frame/Question-Answer) as compared to Personal Info Exchange (Language Frame/Target Item) allows for preexisting proficiency differences to become more apparent.
Dissimilar Interlocutor / Similar Task
To answer the second research question regarding the role of interlocutor, I will now compare the results from two clips in which Jin is working on tasks with the same amount of language structure, but in which she is working with different partners. Both of the tasks in this second comparison have received the language codes of Language Frame for the teacher and Question/Answer for the students. The clip Giving/Receiving Directions was discussed above and will now be compared with a clip called Favorite Family Member. In the latter of these Jin was working with a partner named Inez. Inez is an outgoing, middle-aged Mexican woman. In this clip, the students are practicing talking with their partner about their favorite family members and sharing information received from talking to a different partner about that person’s favorite family member. The students are guided by a list of questions provided by the teacher. Both of these clips show Jin working on a task with a similar amount of language support, but with different partners.
In Giving/Receiving Directions the students receive the language frame question, “Is there a _____ near here?” and an example answer language frame, “Yes, there is.” or “No, there isn’t.” An example answer is also given which describes how the person would get to the building asked about from the starting location located on the map. The students are supposed to ask each other about different locations and use the language frame provided to do the asking and the answering.
Favorite Family Member contains a similar amount of structure in that the students are given questions and a language frame that they use in talking to each other about their family members. In Favorite Family Member, the students are provided with a list of questions to ask about each other’s favorite family members.
The following questions were given to the students to stimulate the conversation about a family member.
1. Who is your favorite family member?
2. What is his or her name?
3. Where does he or she live?
4. What does he or she look like?
5. What is his or her personality?
The questions along with some example answers, which provide language frames, are written on the board prior to the beginning of the exercise. Some of the language frames the teacher gives as examples for the students to use in their discussion are:
• His name is _______.
• Is his name _________?
• He has ___________?
• Does he have _________?
• He is ____ years old.
• Is he ____ years old?
• She lives ________.
• Does she live _________?
Both of these activities provide language support to the students in the form of a language frame given by the teacher. The student provided language in both of these activities is also similar in that in both it is open-ended in the form of questions. There isn’t simply one discreet answer to the questions in these activities, rather the answers are longer in nature and are variable for each person answering. In these ways the language and language support provided by both the teacher and the students is similar in these two activities.
Differences in Negotiation
Table 9 below allows for comparisons of the features of negotiation found in each of the clips. Following this chart is a brief discussion regarding some of the differences in the quantity of features of negotiation found in these two interactions.
Table 9. Giving/Receiving Directions Compared with Favorite Family Member
|Jin |Aracelia |Inez |Difference |
| |Giving/Receiving |Favorite Family Member | |
|L.Frame/Q-A |Directions | | |
|Total Words |A = 139 |I = 224 |I = 85 |
|A = Aracelia |J = 256 |J = 271 |J = 15 |
|J = Jin |(395) |(495) |(100) |
|I = Inez | | | |
|(combined words) | | | |
|Clarification Requests |5 |14 |+ 9 Favorite Family |
| | | |Member |
|Confirmation Check |6 |19 |+ 13 Favorite Family |
| | | |Member |
|Comprehension Check |1 |0 |+ 1 G./R. Directions |
|Untriggered Other Repair |8 |3 |+ 5 G./R. Directions |
|Self Repetition |44 (83) |65 (86) |+ 21 (3) Favorite Family |
|(# of words) | | |Member |
|Other Repetition (# of words) |39 (64) |36 (46) |+ 3 (18) |
| | | |G./R. Directions |
|Recast |1 |0 |+ 1 G./R. Directions |
|Total Time |7 m. 30 s. |9 m. 49 s. |N/A due to interruptions |
|Interaction Pattern |expert/novice |collaborative | |
* (information taken from Tables 3 & 4, pages 67-68)
In looking at these two interactions, it can be seen that “Favorite Family Member contains considerably more negotiation than Giving/Receiving Directions. Favorite Family Member contains almost three times as many clarification requests and just over three times as many confirmation requests. Additionally, many more self repetitions occurred during Favorite Family Member. In some of the categories Giving/Receiving Directions does contain more features, however the categories in which Giving/Receiving Directions contains more occurrences are comprehension check, recasts, other repetition, and untriggered other repair. The differences in these categories, with the exception of untriggered other repairs and other repetition, are of one occurrence compared to zero occurrences. These differences are too minor to sway the overall determination that Favorite Family Member contains more negotiation. The difference in the amount of other repetitions is also not substantial when viewed relative to the total amount of negotiation. The one category in which Giving/Receiving Directions actually contains a considerably higher number of occurrences than Favorite Family Member is Untriggered Other Repairs. As discussed in the previous section, these repairs take place primarily when Aracelia is correcting Jin’s utterances. This provides evidence supporting the categorization of this interaction as expert/novice, with Jin being the novice and Inez the expert.
Not only does Jin’s interaction with Inez contain more negotiation, there are also clear differences in the way that Jin and Inez interact compared to Jin and Aracelia. Jin and Inez’s interaction contains an almost equal amount of words spoken by each student. In contrast, the interaction between Jin and Aracelia is quite different, with Jin speaking almost twice as many words as Aracelia. Additionally, it is easy to notice a difference in the tone between these two interactions. Jin’s interaction with Inez contains copious amounts of laughter and has many instances of friendly body language, such as patting one another on the back. While Jin’s interaction with Aracelia is by no means unfriendly, it is not as sociable and relaxed as her interaction with Inez. It may not be possible to determine whether a more relaxed atmosphere led to more negotiation and collaboration in this instance, or if more negotiation and collaboration led to a more relaxed atmosphere. Regardless of which came first, it is important to notice the differences in the interpersonal communication styles in these two instances.
Proficiency Level and Task Language Support
In the previous section, elements of the task that might explain the lower degree of negotiation found in Giving/Receiving Directions were described. Taking into account this second comparison however in which the elements of language structure in the task should be equivalent, and therefore not a factor, a more complex explanation must be sought. As mentioned earlier, Aracelia’s English proficiency is slightly higher than that of Jin, which does become a factor. The fact that these two students are in the same class level confirms that the difference in their proficiency levels is not great. It is, however, possible that a more open-ended task with less of a script (less language support) would make slight proficiency differences more apparent. The role of interlocutor and task can work together to create variations in the amount of negotiation and the degree of collaboration within an interaction. When a task contains more language structure, small differences in proficiency become less apparent and therefore may not exert significant influence on the amount of negotiation. When the students receive less language support from the task itself, however, they are then forced to rely on their own language abilities more heavily. This is when a student’s linguistic weaknesses become more evident. Even though Giving/Receiving Directions and Favorite Family Member contain the same amount of language structure in the task, there was no disparity in ability between Jin and Inez as there was between Jin and Aracelia. Without this factor, Favorite Family Member was a collaborative interaction whereas Giving Receiving Directions was an expert/novice interaction. In summary, an underlying proficiency difference highlighted by the lessened degree of language structure in the task was responsible for the expert/novice interaction in Giving/Receiving Directions.
Comparative Placement on the Collaboration Continua
The clip titled Giving/Receiving Directions represents an example of an expert/novice pattern of interaction. Both the proficiency level of the students and the lower amount of language support provided worked together in setting the stage for this kind of interaction. The expert/novice pattern was evident in the difference in ability with relation to the task which resulted in Jin relying on Aracelia to provide scaffolding to help complete the task. When looking at Favorite Family Member the same disparity in proficiency levels cannot be observed between Jin and Inez. With this factor absent, the two students worked on the task in a manner best described as collaborative. Inez and Jin worked together during this task in a very, equal and cooperative manner, and performed a relatively high amount of negotiation. The key element that can be held responsible in making one of these tasks expert/novice and the other collaborative, would be the difference in proficiency levels of the pairs of students in these particular interactions.
In conclusion, differences in the nature of these two tasks, in concert with underlying proficiency inequalities, caused the interaction in one task to be collaborative and the other expert/novice. It seems that the more language structure and support that is inherent in a task, the easier it is for the lower-functioning student to get along without reliance on help from her partner.
Similar Interlocutor / Similar Task
In this final comparison, Jin is working with the same partner (Aracelia) during both clips that contain tasks with a similar amount of language structure. Both of these tasks were given the language code of Language Frame for the language provided by the teacher and Target Item for the students. In the clip titled Sibling Similarities, the students are given the instructions “Tell your partner about you and your brother or sister. Are you the same or different?” This occurs after they listen to a recording of a person describing the same thing. A written version of the recording also appears in the textbook as a guide for the students during the activity. This written version of the recording provides a language frame that the students use to talk about their own sibling and how they are similar or different. This clip was compared to Personal Info Exchange, discussed previously. In Personal Info Exchange, the students also have a worksheet giving questions for them to ask their partner along with language frame answers in which they must write down the target item, which is the answer that the partner gives them.
Differences in Negotiation
Table 10 below compares the negotiation occurring in the two interactions, Sibling Similarities and Personal Info Exchange.
Table 10. Sibling Similarities Compared with Personal Info Exchange
|Aracelia & Jin |Lang. Frame/ Target Item |Lang. Frame/ Target Item |Difference |
| | | | |
| | | | |
| |Sibling Similarities |Personal Info Exchange | |
|Total Words |A = 131 |A = 178 |A = 47 |
|A = Aracelia |J = 143 |J = 338 |J = 195 |
|J = Jin |(274) |(516) |(242) |
|(combined words) | | | |
|Clarification Requests | 2 |8 |+ 5 Personal |
| | | |Info Exchange |
|Confirmation Check |11 |19 |+ 8 Personal Info Exchange |
|Comprehension Check | 1 |1 |0 |
|Untriggered Other Repair | 0 |0 |0 |
|Self Repetition | 23 (36) |36 (63) |+ 15 (27) Personal Info |
|(# of words) | | |Exchange |
|Other Repetition | 18 (25) |44 (81) |+ 26 (56) |
|(# of words) | | |Personal Info Exchange |
|Recast | 0 |1 |0 |
|Total Time |3 m. 38 s. |7 m. 19 s. |N/A due to interruptions |
|Interaction Pattern |collaborative |collaborative | |
* (Information taken from Tables 3 & 4, pages 67-68)
These two interactions show Jin working with the same partner in two similar tasks. The initial predictions were that these two clips would contain roughly equal amounts of negation and similar patterns of interaction. Both of these predictions appear to have been accurate. At first glance, it may seem that Sibling Similarities contains less negotiation than Personal Info Exchange, but this is not true. When the total time for each activity is taken into account, the amount of negotiation in the clips is very close. Sibling Similarities is only half as long as Personal Info Exchange and thereby contains less negotiation because it was a shorter activity. These results seem to support the idea that the amount of language structure in the task may influence negotiation and interaction pattern.
Comparative Placement on the Collaboration Continua
Both of these clips were considered collaborative because of the moderate to high levels of mutuality and equality found in each. These two clips also contained a substantial amount of negotiation for their respective lengths. As mentioned earlier, it seems that there is a difference in language proficiency between these two learners, but the high degree of language support found in this type of activity assists the less proficient student to a point where this is not readily apparent. As stated above, these results suggest that the amount of task language support, amongst other variables, can influence the amount of negotiation and pattern of interaction.
Summary
This chapter began by describing the clips that I analyzed and then listed the results found during the first and second phase of data analysis. Next, I discussed the similarities and differences I found between the clips. Many of the conversational features discussed by Storch (2002) were observed in the clips that I reviewed. The results of between clip comparisons suggest that both interlocutor and task language support interact in exerting a complex, multi-dimensional influence on negotiation and interaction. Increased language support inherent in the task seems to mediate slight shortcomings in proficiency.
In the following chapter, I will further discuss the results and their implications. I will summarize major findings and attempt to put them into perspective with regard to other work concerning the variables of task and interlocutor. Limitations of this study will be discussed followed by a brief section suggesting ideas for further research. Lastly, I will discuss the implications that this study and its findings have for second language teaching.
CHAPTER 5
DISCUSSION
Summary of Findings
This qualitative research study had two goals. One was to describe the relationship between interlocutor and negotiation found in the pair interactions of an adult ESL learner. The other goal addressed the relationship between task and negotiation, more specifically the amount of language structure or support inherent in the task.
The results led to the following conclusions:
• The majority of the interactions were collaborative.
• The variables of task and interlocutor, especially relative language proficiency, are interdependent in their influence on negotiation.
• Differences in relative proficiency level have the potential to influence negotiation and interaction pattern in pair work.
• Students can assist other students in ways similar to that of the teacher, especially in expert/novice interactions.
• Students will actively use resources available to get the linguistic support they need.
• The benefits of interaction and negotiation are often still realized even when students are not completing the task as intended or are completely off-task.
Collaborative Majority
When thinking of collaboration as existing on a continuum there is a wide range of possible ways in which students may interact during a classroom task. Storch (2001, 2002) viewed this continuum in terms four patterns; collaborative, dominant/dominant, dominant/passive, and expert/novice. With a range of possible responses, it is important to note that the majority of the pair work recordings from my research showed the same pattern. A collaborative pattern of interaction was found in three of the four interactions analyzed. The predominance of collaboration agrees with the findings of Storch (2001) in an even larger set of interactions. The prevalence of a collaborative pattern could be the result of many factors. Student motivation, similar proficiency levels, and a multi-cultural classroom may all be somewhat responsible.
The classroom in which these interactions occurred consists of students who are there on their own volition and who are not getting credit for the coursework. Many of these students balance the responsibilities of busy work schedules and family duties and still make time to come to class to improve their English. The students in the Lab School classes are often highly motivated. The simple desire to learn on the part of the students likely causes them to strive for collaboration in their learning.
The students in this class are all level B, meaning they all have roughly the same English proficiency. Although there may be some slight relative differences between students, there are not vast differences in English ability. The absence of huge discrepancies in English proficiency level reduces the likelihood of expert/novice scenarios and may make collaboration more likely to occur.
Finally, because this classroom contains students from many different countries, collaboration may be more prevalent. In an ESL class with mixed gender and mixed ethnicity students, a third culture of the classroom is created in which the students’ own culturally biased perceptions of talk do not apply. This levels the verbal playing field, leading to greater equality in contributions by male and female students. This same concept was found at work by Topolski (2004). The high proportion of collaboration in this kind of mixed-gender, mixed-ethnicity ESL classroom might be in part due to the great amount of diversity in the student population coupled with a high degree of motivation to learn English.
Task and Interlocutor Interdependence
In completing this research, I found that the influence of task and interlocutor cannot accurately be viewed in solitude. The results of this study show that the roles of task and interlocutor combine and work together in influencing the amount and kind of negotiation in pair interactions. It is not possible to study the impact of either of these without accounting for the other. An important component of the interlocutor that can be seen to influence negotiation is relative proficiency level. Interactions between Jin and her partners show that increased language support within the task can help mediate differences in relative proficiency level. This suggests that tasks with more language support may more likely be collaborative because the students are participating on a more equal footing, where tasks with less language support may show interaction patterns which reflect a disparity in second language proficiency level, for example expert/novice. This was exactly the case in the data that I analyzed.
The interaction pattern that differed from the rest was expert/novice. It seemed that in the expert/novice clip, the decreased language support in the task exacerbated underlying proficiency differences and caused Jin to rely more heavily on Aracelia for support. In this instance, the relative proficiency levels of the students together with a reduced amount of language support set the stage for an expert/novice interaction. It is important to remember that Aracelia could have responded in a number of ways that would not have resulted in her being labeled as an expert and Jin as a novice. Aracelia provided peer scaffolding to assist Jin during this activity. Teachers often provide structured support, known as scaffolding, that allows students participate in the task and extend current language competence. In pair work, learners can also provide each other with scaffolding to overcome obstacles (Donato, 1994). It was the kind of support and assistance that Jin provided during breakdowns in the communication, which led to designation as expert/novice. In response to Jin’s difficulties, Aracelia could have chosen not to be helpful and could have reacted in a domineering and controlling manner, which may have resulted in a dominant/passive situation. Jin, however, acted in a helpful and guiding manner that resulted in an expert/novice pattern of interaction. Aracelia can be observed tutoring Jin through difficulties in comprehension in a way that seems to allow Jin a better opportunity to grasp the language of the task.
In another interaction with a similar task type in which Jin was working with Inez, it was clear to see that the students were performing on more of an equal footing with regard to their skill in completing the activity. In Favorite Family Member more equivalent proficiency levels together with congenial and cooperative attitudes on the part of the students created a collaborative environment where each had some difficulties, but neither had strikingly more than the other.
In conclusion, tasks with less language structure may cause some students of slightly lesser ability to become more reliant on their more proficient partners for support. The impact of this factor on the negotiation and collaboration is also mediated by how the more proficient student responds to her partner’s needs. When a task contains a greater degree of language support, a less proficient student may not require as much assistance from his or her partner because he or she is able to receive some support from the task itself in the amount of language scriptedness it contains.
Relative Proficiency Levels
Even within one level, for example level B in this case, there may be slight yet noteworthy variations in the English proficiency of the students. Students that show similar proficiency test scores may have noticeable difference in ability when observed doing classroom tasks (Canada, 2004). As in Canada’s study it was evident there were differences in second language proficiency between the students despite them being in the same class level. It was not expected that these differences would influence negotiation. In completing a language task, there are a few different places where less proficient learners seek assistance or language support. Students often ask the teacher directly for help with difficulties. Students who have a similar first language may use it to mediate dissimilar second language proficiency levels (Canada, 2004). When the language support is not available in the form of a shared first language, it may be made available within the task. Greater language support (i.e.. scriptedness) in tasks also helps mediate dissimilar second language proficiencies. Tasks with fewer possible correct answers and a fuller script to assist in answering the questions give the less proficient student support and help to conceal the proficiency deficiencies between them and their language partner. However, in a task with less language support, a difference in relative proficiency became a factor. The result was an expert/novice pattern of interaction. In summary, not only do students within one ESL class level possess different relative second language proficiencies, these also exert an influence on the negotiation and interaction pattern of pair work.
Task and Interlocutor as Variables of Negotiation
It has been suggested that increased negotiation may positively influence acquisition (Canada, 2004; Day, 1986; Donato, 1994; Gass & Varonis, 1994; Long, 1996; Smith, 2004; Storch, 2001). This leads to questions about what aspects of the classroom experience may play a role in the amount of negotiation that takes place. Solving that puzzle may allow instructors to design tasks that optimize negotiation.
In this study, I considered how an aspect of task (inherent language support) and interlocutor influence negotiation. Others have attempted to understand the role of task with regard to negotiation and have illuminated many variables within a task that may help determine negotiation. These variables often act together in their influence on negotiation. Whether the information exchanged in a task is one-way or two-way may result in differences in negotiation (Gass & Madden, 1985; Gass & Varonis, 1985). Task type (optional or required information exchange) and the student grouping pattern (dyads or small groups) may influence the amount of negotiation (Foster, 1998). Task type was investigated by comparing a Jigsaw task and a Dictogloss task with the hypothesis that the Dictogloss group would focus more on form (Swain & Lapkin, 2001). Both task content familiarity and task language structure have also been investigated with regard to their influence on negotiation (Hardy & Moore, 2004). These studies and others help to confirm dimensions of task as important avenues of research with regard to their relationship to negotiation and interaction.
As an aspect of task, I analyzed tasks containing different degrees of language structure provided to the students by the teacher. Tasks that were labeled in the Lab School data as Language Frame/Target Item had a higher degree of language structure than those labeled Language Frame/Question-Answer. I found that a higher degree of language structure helped mediate relative differences in the students’ English language proficiency. The end result of this was that students that may have ended up having either expert/novice or dominant/passive interaction patterns ended up being collaborative due to the fact that the less proficient student was able to receive some support from the inherent structure of the task. The findings of my study with regard to task language structure are not similar to the studies mentioned above as they all investigated aspects of task other than language structure. A similarity can be seen to the study done by Canada (2004). In Canada’s study, she found a similar mediating effect on proficiency differences when students shared a first language.
In summary, if there exist differences in the English proficiency of the students, a collaborative interaction can still take place as long as the weaker student is able to get extra support from somewhere whether through their partner and a shared first language or a highly structured task, which lessens the difficulty and ambiguity.
It is also worthwhile to mention again the importance of the modality of the task and the passive or active role of the partners involved as they relate to negotiation. In other words, whether a student is required to record an answer accurately in writing or simply passively listen to a verbal response will influence the negotiation that takes place. As mentioned in the comparison of Personal Info Exchange and Giving/Receiving Directions in chapter 4, if the students are required to actively write their partner’s responses, a greater amount of negotiation may be observed than if there is not such a requirement. Given no requirement to record a partner’s answers in writing it is likely that the student will be as concerned with complete understanding of the message as they would be if they were simply required to listen passively to the message. In short, a requirement of active written recording of their partner’s responses likely will result in a greater focus on accuracy and thereby potentially cause an increase in negotiation.
Another variable thought to influence negotiation in pair work is that of interlocutor. Like the variable of task, interlocutor consists of many interrelated aspects. Researchers have investigated the roles of gender, language backgrounds, and interlocutor familiarity, among others. Of concern in Gass and Varonis (1985) was the role, if any, that gender played during pair interactions. Topolski (2004) also explored its role in mixed-gender dyads. Some have queried the function of similar or different language and education backgrounds (Canada, 2004; Foster, 1993; Swain & Lapkin, 1998). Others have sought to explain the role of working with a familiar or unfamiliar interlocutor (Plough & Gass, 1993). This body of work makes a strong case for further research in determining the influence of interlocutor variables on the quantity and quality of negotiation in pair work activities.
Students as Teachers
One important conclusion that I realized when analyzing and reviewing my data is that students can be invaluable resources to other students during the learning process. This was most evident in the expert/novice interaction in which Aracelia acted as the expert and Jin the novice. It was encouraging to see Aracelia applying many of the same techniques that are often used by teachers. Aracelia can be observed using self repetition coupled with a slower rate of speech to clarify the information that Jin was not able to understand. Aracelia also employed recasts and untriggered other repairs in the way a teacher might in order to help Jin become aware of problems. Encouragement was frequently offered after correct answers in the form of phatic utterances and verbal praise. These features demonstrate that students can offer each much assistance in a way similar to students. This allows the students in essence to be teachers and is an important conclusion for several reasons.
Teachers in large classes may not be able to meet the needs of all the students simply because of time constraints and the teacher/student ratio. For this reason peer assistance like this can be extremely helpful to students who may be struggling. Expert/novice pairings are not only beneficial to the novice, but also to the expert. There is often no better way to ensure that something is thoroughly comprehended than by trying to explain it to another person.
Use of Available Resources
My results show that students will actively use the resources available to get the linguistic support they need. When students do not find support from one source they will seek it out in another. An example of this is found when we compare two of the activities Jin took part in with Aracelia. In Personal Info Exchange, there exists a high degree of linguistic support built into the task. Jin can be observed using this support or scriptedness in the task to help her understand the questions involved and surmount difficulty she has in giving responses. Giving/Receiving Directions on the other hand does not contain a lot of scriptedness and is far more open-ended. For this reason, Jin is not able to appeal to the structure of the task for help and instead looks to her partner as a linguistic resource. Other resources that the students can be seen using are the writing on the whiteboard, the teacher, and other students who speak the same first language. Many resources are available to assist students and it is important to remember that as active learners they will use what they can to help them during task completion. Teachers should help ensure that students have access to as many of these resources as possible.
Benefits of Being Off-Task
Within the tasks that I observed, students would sometimes complete the task in a manner different from that which was intended. On other occasions students would for some reason or another wander off-task and begin talking about some subject that was completely unrelated to the task at hand. In analyzing the times when students were off-task or going about the task in an incorrect manner I found that there was still a great deal of valuable conversation that took place. If we are to say that students receive the benefits of language acquisition through the processes of interaction and negotiation, then it may not always matter that they are completing the task exactly how it was intended or that they are even working on the task at all. It should be noted that when students are using English as the primary language of communication that they are open to receiving the many benefits of interaction and negotiation. Teachers should be aware of this and know that it may not always be necessary to redirect students that are off-task, as they may be engaging in the learning processes through whatever conversation they may be having.
Limitations of this Study
This study was small in scale and qualitative in nature and therefore offers a depth of explanation not available in a large scale, quantitative study. However, it is important to talk about how some of the things that may be virtues for this research project may also represent shortcomings. Three primary limitations are the use of naturally occurring data, the small number of interactions of students involved, and the variable length of the activities analyzed. Additionally, there may be a myriad of environmental variables that may have confounded the results.
The use of naturally occurring data should be viewed as a strength and a weakness. The data gathered from the interactions is all genuine and authentic. These interactions were not planned or constructed, but rather were samples from an ongoing ESL class. Without the strict control of variables found in an experimental environment, it is hard to pinpoint causal factors. The real language classroom is a valuable and viable location to study language acquisition; however, the door is open to confounding variables such as variable length of activities, unpredictable size of the data set, and the potential for the students’ previous familiarity with the activities.
This study took an in-depth look at one student and her work with two different partners across two different types of activities. Data selection yielded four separate interactions and three comparisons between these interactions. The small number of interactions analyzed means that although the results may contain more detail, the results cannot be generalized to interactions involving other students and other activity types. The small number of interactions analyzed in this research is a result of the use of naturally occurring classroom data discussed above. Having chosen to focus on the interactions of one student, I was then limited by the availability of recorded data involving that student. Instead of creating the number of interactions that I would have liked to analyze, as I might have done in an experimental environment, I used what was available in the existing collection of Lab School recordings.
Another aspect of real classroom data is the variable length of tasks. This created difficulty in comparing the raw numbers of negotiation between the different tasks. I attempted to choose tasks that were close in length, but this was not entirely possible. A related aspect of this is the inability in classroom data to control for interruptions by other students and the teacher. These interruptions not only affect the length of the pair interaction, but they often affect the potential for negotiation as well. Most of the time these interruptions involve one or both of the students negotiating with the teacher or another student, either in English or in the student’s native language. This may lower the amount of negotiation that might have taken place between the two students.
Questions for Further Research
The findings from my study suggest questions for further research, which fall into the following categories: 1) size of the data set, 2) content specific relative proficiency 3) acquisition gains and specific student partner roles, and 4) greater difference in the degree of language support within the task.
The small data set used in this study represents a limitation as well as a direction for future study. My results cannot be broadly generalized because of the small data set; therefore, expanding on this study by adding more interactions for analysis would likely add strength, detail, and greater generalizablity to the conclusions. A larger data set could shift the focus from one of a qualitative nature to that of a quantitative one.
Also, because difference in comparative proficiency levels was thought to be part of what led to an expert/novice interaction in Giving/Receiving Directions, it would be important to verify that the difference in proficiency levels was not specific to the content of that particular task. A future study in which Jin is completing a Language Frame/Question Answer task with Aracelia with different task content, for example one not involving directions and prepositions, would help determine if the difference in proficiency levels is content-specific.
My study revealed a majority of collaborative interactions. To complement this information discovered, it would be useful to learn more about the gains of the other student role types other than collaborative. An interesting direction for further research would be to determine the proficiency gains for students in interactions labeled as dominant/dominant, but were expert/novice and dominant/passive instead. Determining if specific roles within these relationships experience greater gains in relative proficiency is a worthy goal of future research.
Finally, the clips analyzed in my study contained different amounts of language support. Although different, these two task types provide amounts of language support that aren’t greatly dissimilar. In response to that issue, and to better understand the relationship between negotiation and task language support, it would be useful to examine interactions with greater difference in their amount of language support. A follow-up study should analyze more interactions in general and should take them from the categories of Language Frame/Target Item and Question-Answer/Question-Answer. These two types of interactions contain drastically different amounts of language support in the tasks. Tasks given the code of Question-Answer/Question-Answer contain even less language structure and support than those coded Language Frame/Question-Answer and therefore may deliver a better explanation of the role of task language support when compared to tasks, which are Language Frame/Target Item.
Implications for Teaching
From this study, we can see that adult low-level ESL pair work is generally collaborative in nature. I believe this reflects the students’ motivation to learn. Even when students show a different pattern of interaction (i.e.… expert/novice) they still have many opportunities to learn. As teachers, we can learn that our choices in pedagogy and student pairing build on an already cooperative inclination, but can influence the negotiation and interaction pattern that takes place. In addition, the students’ active role in the learning process can’t be ignored. Linguistic resources in a classroom setting are ideally available from a large variety of sources (other students, textbook, handouts, teacher, task, classroom environment etc.). Motivated students will actively seek out and use the resources available to them.
What I learned during the course of this research study will affect my teaching in a number of ways. I will recognize that most students work actively to cooperate and help each other learn. In addition, during lesson planning, I will be more conscious of the amount of language structure that is provided in the language tasks I include. I will consider the linguistic resources available to assist with the task. It is important to provide a learning environment rich in linguistic resources, therefore I will strive to achieve this. The walls of my ideal classroom would be filled with words and pictures in a variety of forms from grammar charts to realia to student projects and writings.
I will consider the individual ability of the students with regard to the task when I assign partners for pair work. Having students self-select their partners will also be a useful option. I will consider pairing students with potential for expert/novice interactions, when appropriate. I believe in students’ abilities to provide scaffolded assistance to each other and will attempt to promote its occurrence in my classroom. In summary, I feel that I have learned a great deal from this research and know that it will definitely influence my teaching in a positive manner.
References
Canada, S. (2004). Distance and negotiation in the low-level esl classroom., Portland State University, Portland.
Chaudron, C. (1988). Second language classrooms: Research on teaching and learning. New York, NY: Cambridge University Press.
Cumming, A. (1990). Metalinguistic and ideational thinking in second language composing. Written Communication, 7, (482-511).
Day, R. R. (Ed.). (1986). Talking to learn: Conversation in second language acquisition. Cambridge: Newbury House Publishers.
Donato, R. (1994). Collective scaffolding in second language learning. In J. Lantolf & G. Appel (Eds.), Vygotskian approaches to second language research (pp. 33-56). Norwood, NJ: Ablex.
Ellis, R., Tanaka, Y., & Yamazaki, A. (1994). Classroom interaction, comprehension, and the acquisition of l2 word meanings. Language Learning, 44(3), 449-491.
Ferguson, C. A. (1971). Absence of copula and the notion of simplicity: A study of normal speech, baby talk, foreigner talk, and pidgins. In D. Hymes (Ed.), Pidginization and creolization of languages. (pp. 141-150). Cambridge: Cambridge University Press.
Foster, P. (1993). Discoursal outcomes of small group work in an efl classroom: A look at the interaction on non-native speakers. Thames Valley University Working Papers in English Language Teaching, 2, (1-30).
Foster, P. (1998). A classroom perspective on the negotiation of meaning. Applied Linguistics, 19.
Gass, S., & Madden, C. (Eds.). (1985). Input in second language acquisition. Rowley, Massachusetts: Newbury House Publishers, Inc.
Gass, S., & Varonis, E. (1985). Task variation and nonnative/nonnative negotiation of meaning. In S. M. Gass & C. G. Madden (Eds.), Input in second language acquisition (pp. (149 - 161)). Cambridge: Newbury House Publishers.
Gass, S., & Varonis, E. (1994). Input, interaction, and second language production. In Studies in second language acquisition (Vol. 16 cum. index v.1-15, pp. (283-302)). Cambridge, MA: Newbury House.
Hardy, I. M., & Moore, J. L. (2004). Foreign language students' conversational negotiations in different task environments. Applied Linguistics, 25(3), (340-370).
Harris, K., Twenge-Jinings, F., & Lachenmeier, E. (2006). Negotiation across activity types as they occur in adult low-intermediate esl classes: A view from inside the classroom. Paper presented at the Second Language Research Forum, Seattle, Washington.
Hatch, E. (1978a). Acquisition of syntax in a second language. In J. Richards (Ed.), Understanding second and foreign language learning (pp. (34-70)). Rowley, MA: Newbury House.
Hatch, E. (1978b). Discourse analysis and second language acquisition. In E. Hatch (Ed.), Second language acquisition: A book of readings. (pp. 401-435). Rowley, MA: Newbury House.
Hatch, E., & Wagner-Gough, J. (1976). Explaining sequences and variation in second language acquisition. Language Learning, 4, (39-47).
Jacob, E., Rottenberg, L., Patrick, S., & Wheeler, E. (1996). Cooperative learning: Context and opportunities for acquiring academic english. TESOL Quarterly, 30, (253-280).
Krashen, S. (1982). Second language acquisition. New York: Pergamon Press.
Krashen, S. (1985). The input hypothesis: Issues and implications. London: Longman.
Larsen-Freeman, D., & Long, M. (1991). An introduction to second language acquisition research. New York: Addison Wesley Longman Inc.
Long, M. (1980). Input, interaction, and second language acquisition: University of California, Los Angeles.
Long, M. (1983). Native speaker/non-native speaker conversation and the negotiation of comprehensible input. Applied Linguistics, 4, (126-141).
Long, M. (1985). Input and second language acquisition theory. In S. M. Gass & C. G. Madden (Eds.), Input in second language acquisition. (pp. 377-393). Rowley, MA: Newbury House.
Long, M. (1996). The role of the linguistic environment in second language acquisition. In W. Ritchie & T. Bhatia (Eds.), Handbook of second language acquisition (pp. 413-467). San Diego, California: Academic Press.
Long, M., & Crookes, G. (1992). Three approaches to a task-based syllabus design. TESOL Quarterly, 26(1), (27-56).
Ouellette, S. (2004). Making the effort: A study of one student's communication strategies in an esl classroom., Portland State University, Portland, Oregon.
Phillips, J. (1973). Syntax and vocabulary of mothers' speech to young children: Age and sex comparisons. Child Development, 44(182-85), 70,78.
Pica, T., Young, R., & Doughty, C. (1987). The impact of interaction on comprehension. TESOL Quarterly, 21(4), (737-758).
Plough, I., & Gass, S. (1993). Interlocutor and task familiarity: Effects on interactional structure. In G. Crookes & S. M. Gass (Eds.), Tasks and language learning (pp. (35-56)). Clevedon: Multilingual Matters.
Reder, S., Harris, K., & Setzler, K. (2003). The multimedia adult esl learner corpus. TESOL Quarterly, 37(3), 546-557.
Sachs, J., Brown, R., & Salerno, R. (1976). Adult's speech to children. In W. Von Raffler Engel & W. Lebrun (Eds.), Baby talk and infant speech. (pp. p. 70). Lisse: Peter de Ridder Press.
Selinker, L. (1972). Interlanguage. International Review of Applied Linguistics(10), 209-231.
Smith, B. (2004). Computer-mediated negotiated interaction and lexical acquisition. Studies in Second Language Acquisition, 26(3), 365-398.
Smith, B. (2005). The relationship between negotiated interaction, learner uptake, and lexical acquisition in task-based computer mediated communication. TESOL Quarterly, 39(1), 33-58.
Snow, C. E. (1972). Mothers' speech to children learning. Child Development, 43(549-65), 70-71.
Snow, C. E. (1986). Conversations with children. In P. Fletcher & M. Garman (Eds.), Language acquisition: Studies in first language development (pp. 69-89). Cambridge: Cambridge University Press.
Storch, N. (2001). An investigation into the nature of pair work in an esl classroom and its effect on grammatical development. The University of Melbourne, Melbourne.
Storch, N. (2002). Patterns of interaction in esl pair work. Language Learning, 52(1), 119-155.
Swain, M. (1985). Communicative competence: Some roles of comprehensible input and comprehensible output in its development. In S. M. Gass & C. G. Madden (Eds.), Input in second language acquisition (pp. (235-253)). Rowley, MA: Newbury House.
Swain, M. (2000). The output hypothesis and beyond; mediating acquisition through Collaborative dialogue. Sociocultural theory and second language learning, 99-116.
Swain, M., & Lapkin, S. (1995). Problems in output and the cognitive processes they generate: A step towards second language learning. Applied Linguistics, 16, (371-391).
Swain, M., & Lapkin, S. (1998). Interaction and second language learning: Two adolescent French immersion students working together. Modern Language Journal, 82, (327-343).
Swain, M., & Lapkin, S. (2001). Focus on form through collaborative dialogue: Exploring task effects. In M. Bygate, Skehan, P., and Swain, M. (Ed.), Researching pedagogic tasks (pp. 99 - 118). Essex: Pearson Education Limited.
Topolski, C. (2004). Talk time: A study of mixed-gender interaction in an adult esl classroom., Portland State University, Portland, Oregon.
Varonis, E., & Gass, S. (1985). Non-native/non-native conversations: A model for negotiation of meaning. Applied Linguistics, 6(1), 71-90.
Vygotsky, L.S. (1962). Thought and Language. Cambridge, MA: MIT Press.
Vygotsky, L.S. (1978). Mind in Society. Cambridge, MA: Harvard University Press.
Wertsch, J. V. (Ed.). (1979). The concept of activity in social psychology. Armonk, NY: M.E. Sharpe.
-----------------------
[1] The National Labsite for Adult ESOL (known locally as the Lab School) is supported, in part, by grant R309B6002 from the Institute for Education Science, U.S. Department of Education, to the National Center for the Study of Adult Learning and Literacy (NCSALL). The Lab School is a partnership between Portland State University and Portland Community College. The school and research facilities are housed at the university while the registration, curriculum, and teachers of the ESL students are from the community college.
................
................
In order to avoid copyright disputes, this page is only a partial summary.
To fulfill the demand for quickly locating and searching documents.
It is intelligent file search solution for home and business.
Related searches
- co signers rights and responsibilities
- education and success in life
- cultures and co cultures
- measuring and constructing angles calculator
- chemotherapy success and failure rates
- co teaching strategies and examples
- constructing linear functions quizlet
- therapy modifiers co and cq
- modifier co and cq
- co and cq modifiers 2020
- co and cq modifier
- cost of constructing a home