Research methods for analyzing second language classroom interaction

SLA Second Language Aquisition

As the world has become increasingly more global, interests in second language acquisition has also increased. More specifically second language acquisition as it pertains to the second language classroom has become a focal point. The following research will examine three methods that are utilized in Second Language research including conversation analysis, stimulated recall and the Think Aloud Method. For the purposes of this discussion, we will explore the relationship of each approach to theory and discuss the advantages and disadvantages of each system. Illustrate your argument through the use of transcribed extracts from second language classroom interaction. Let us begin the discussion with some background information about second language acquisition and second language classrooms.

Second language acquisition and second language classrooms

McLaughlin (1984) explains that a first language is defined as the language that a person speaks that is first chronologically. The author explains that this language can be present for only a brief time in a child's development and may even be forgotten or never used (McLaughlin 1984). In such occurrences, it is often the case that the second chronological language becomes a person's primary language and only language in the course of daily conversation (McLaughlin 1984).

McLaughlin (1984) also asserts that the distinctions between first and second language acquisition do not apply to bilingual children (McLaughlin 1984). A bilingual child is defined as an individual who from birth has been exposed to two different languages. For instance, if a mother speaks Spanish and the father speaks English, the child will learn both languages simultaneously, making them bilingual (McLaughlin 1984).

However, differing processes occur when one language is acquired first and then a subsequent language is acquired. McLaughlin (1984), points put that researchers have differing opinions about the age at which a first language is said to be establish. McLaughlin (1984) asserts that this age is around three and that any language introduced after this time is a second language.

The author contends the distinction between simultaneous acquisition of two languages and successive acquisition of a second language is not always easy to make. Children differ considerably in the rate at which they acquire a first language. Hence a cutoff point based on linguistic and cognitive developmental criteria is preferable to one based on chronological age (Here, however, one confronts the intractable problem of what it means to say that a child (or chimp, for that matter) possesses a language. It seems the better part of wisdom to avoid this quagmire and stay with the three-year criterion (McLaughlin 1984)."

Finally, the author draws a distinction between second language learning and second language acquisition. The author explains that second language acquisition is defined as "the sub- conscious acquisition of a second language in a natural environment (McLaughlin 1984)."On the other hand, second-language learning "refers to conscious learning in a formal classroom situation with feedback, error correction, rule learning, and an artificial linguistic environment that introduces one aspect of the grammar at a time (McLaughlin 1984)." However, the author also explains that the distinction between conscious and subconscious language development is difficult to make because one cannot always determine to what degree learners are aware of what they are acquiring (McLaughlin 1984).

This discussion will address the issues of researching second language acquisition as it relates to the second language classroom. It is important to understand that second language classroom research can be problematic because of the parameters that are placed on researchers. According to Gass and Schachter (1996), since applied linguistics is an interdisciplinary field, the research paradigms have been varied, with even the subset of second language classroom research spanning a range of research types. Furthermore, classroom research has also drawn simultaneously from several paradigms, often making it difficult to classify, especially as many researchers do not explicitly state their paradigm (Gass and Schachter 1996)."

The author further assert that second language classroom research works best if researchers have carefully planned the type of research they will conduct. This means the formation of protocols and the role (if any) that the teacher or student will play in research process. The authors assert that it works best if there is a collaborative effort between the teacher and the researcher. Over the next few pages of this discussion, we will explain in detail the theories that are utilized in second language classrooms as it relates to interaction.

Conversation Analysis

Amongst the various theories used to analyze second language interaction in the classroom Conversation analysis (CA) is amongst the most prominent. Conversation analysis is defined as "a form of Analysis of conversational data ACD that accounts for the sequential structure of talk-in-interaction in terms of interlocutors' real-time orientations to the preferential practices that underlie, for participants and consequently also for analysts, the conversational behaviors of turn-taking and repair in different speech exchange systems (Markee 2000)." According to Markee (2000), conversation analysis is a term that is used in many different disciplinary perspectives including interactional sociolinguistics, speech act theory and ethnomethodology and social psychology. The term is synonymous with ethnomethodological orientation or the analysis of conversational data (Markee 2000). The author asserts

The methodology of CA is qualitative and thus subject to the usual evaluation criteria for such research. Beyond this, however, CA attempts to explicate in emic terms the conversational practices that speakers orient to (i.e., the rules of talk they deploy for each other and, by extension, for analysts) by "unpacking" the structure of either single cases or collections of talk-in-interaction. Such cases provide the primary evidence for the asserted existence of particular conversational mechanisms identified by analysts. In short, a case is only convincing to the extent that it is directly motivated by the conversational data presented for analysis (Markee 2000)."

The author explains that the theory of conversational analysis began as a subdiscipline in the 1960's and 1970's (Markee 2000). In the beginning, those researching CA placed most of the emphasis on the organizational structure of normal conversation (Markee 2000). The researchers defined "normal conversation" in terms of the routine conversations that amongst acquaintances and friends in social situations when people were either on the phone or face-to-face (Markee 2000). The early researchers referred to the structure of such conversations in relation to repair practices and turn taking (Markee 2000).

Markee (2000) asserts that from an epistemologically standpoint conversation analysis and ethnography are similar approaches. Both of thes "e approaches focus on the particular rather than the general and also seek to develop a participant's rather than a researcher's perspective on whatever phenomenon is being studied. Developing a participant's perspective involves developing a rich description of context (Markee 2000)." The author explains that the main difference between the two is the manner in which they comprehend context differs (Markee 2000).

For instance, some conversation analysts tend to combine ethnographic information into their analyses because they contend that this information is essential for a complete understanding of talk-in-interaction (Markee 2000). However, those researchers that follow a strictly purist approach to conversational analyst do not incorporate ethnographic account cultures or biographies to make an argument (Markee 2000). With the exception of certain cases in which internal evidence exists in the conversational data to provide a warrant for the inclusion of information related to culture or biographies (Markee 2000).

Analysis utilizing conversational analysis

According to Markee (2000) the minimal tools needed in such an anlysis are audio or video recordings of the individual's talk. These tools are needed because it is the only way to preserve the complexity of conversational behavior. The author also reports that the audio and video data are the primary sources that conversational analyst use (Markee 2000). Markee (2000) also explains that in the past audio recordings were preferred because of the expense and awkwardness of video cameras. For this reason, many early researchers used telephone conversations to transcribe (Markee 2000). In fact, the most well-known transcription utilizing conversational analysis is the "two girls" telephone talk (Markee 2000). The author asserts

By focusing on talk that occurs during telephone calls, researchers were able to circumvent two problems rather neatly. First, the expense involved in conducting research was reduced to manageable levels because audio was already a mature recording technology. Second, because telephone partners do not have access to each others' facial expressions and gestures, audio recordings are well suited to capturing how participants display their mutual understandings to each other by voice alone. However, with the advent of cheaper and better video equipment, video recordings are now the medium of choice, as they allow researchers to see how phenomena such as the direction of participants' eye gaze, facial expressions, and gestures are coordinated with, and indeed are part of, the structure of talk-in-interaction (Markee 2000).

As it relates to the second language classroom a transcript in which conversation analysis was utilized can be found below. In this particular instance, the participants were four Turkish-Danish bilingual 14-year-old 8th grade students. (Steensig 2004) the students were sitting in a classroom performing a task where they were asked cut out pictures from magazines, glue them onto a poster, and create a story about the pictures (Steensig 2004).

The researcher observed the following conclusions about conversation analysis

The use of a conversation-analytical transcription is important because it pinpoints details which are essential for understanding code-switches and the negotiation of roles and relations (Steensig 2004).

The method also provided a detailed analysis of what it is pertinent for each participant to do at precise points in the interaction (Steensig 2004). This is critical to comprehending the context in which events such as code-switches, occur (Steensig 2004).

The conversation-analysis theory can also aid in understanding how Participants make alliances and afford "power wielding" in the interaction (Steensig 2004).The author asserts that "Although this point was only cursorily developed in Steensig (2000a) it was claimed that detailed analyses using conversation analytical methods may be a clue to a better understanding of the social relations between the participants (Steensig 2004)."

Advantages and Disadvantages of Conversation analysis

The primary advantage of conversation analysis is that it relies upon the participants' actual conversational process. According to Steensig (n.d.) "Conversation Analysis only uses data from recordings of situations in people's daily lives where nothing has been done to favor certain types of behaviour or otherwise experimentally control what is going on (Steensig n.d.)." Because analyst only use the participants actual conversations they can garner an accurate view of the participants actual organizational structure as it relates to second language acquisition. Steensig (2004) also notes that the conversation analysis method is advantageous because the detailed manner in which the data is gathered makes it easier to understand and draw conclusions from.

The primary disadvantage of conversation analysis is the sheer complexity interpreting the data once it is collected. There are a myriad of factors researchers take into consideration, including body language and the tone of the conversation. In some cases, the conversations that are researched are relatively short and it may be difficult for the researcher to gather an accurate analysis of the conversation.

Stimulated recall methodology

Stimulated recall is also an approach that is utilized in analyzing second language interaction in the classroom. Stimulated recall methodology is unique in that it is an introspective approach. Gass and Mackey (2000) assert that the stimulated recall methodology is one subset of a range of introspective methods that represent a means of eliciting data about thought processes involved in carrying out a task or activity. The assumption underlying introspection is that it is possible to observe internal processes in much the same way as one can observe external real-world events. Another assumption is that humans have access to their internal thought processes at some level and can verbalize those processes (Gass and Mackey 2000)."

Stimulated recall is of particular important in the area of second language acquisition because it seeks to understand the actual thought process of an individual as they acquire a new language (Gass and Mackey 2000). Other methodologies that are used in second language research are sometimes troublesome because "understanding the source of second language production is problematic because often there are multiple explanations for production phenomena that can only be assessed by exploring the process phenomena (Gass and Mackey 2000)."

Utilizing the Stimulated recall method

Gass and Mackey (2000) report that prior to using this method in second language research, there should be a research protocol should be created. The author explained that a detail protocol is needed to ensure that the researchers do not succumb to the pitfalls of the stimulated recall method due to its complex nature (Gass and Mackey 2000). For instance, the authors assert that, "a stimulated recall of oral interaction often involves making at least two separate data recordings, one replay, and two sets of instructions. Thus, the amount of detail specified in the research protocol is important. A detailed protocol helps the researcher to anticipate problems in advance while also acting as a checklist for the many variables and factors the researcher needs to consider and balance while carrying out the procedure (Gass and Mackey 2000)."

In addition, the authors contend that it is crucial that researchers perform a pilot test on all procedures that they present to the learner (Gass and Mackey 2000). Researchers using the stimulated recall method should also pay close attention to the impact of their instructions on the procedure. In this instance pilot testing is critical because it can lead to modifications and clarifications of the protocol (Gass and Mackey 2000). Pilot testing can aid researchers in avoiding expensive and time-consuming problems during the data collection procedure (Gass and Mackey 2000). It can also aid to avoiding the loss of useful and perhaps irreplaceable data (Gass and Mackey 2000).

Second Excerpt

In the second excerpt, Shoko explains her interaction when students read their newspaper article. She asserts "To be honest, I sometimes do not listen to their reports. Their reports are often long and they sound like they are reading out from the newspaper. I cannot understand them. (Recall Session 1; 5 Oct.) (Nabei and Swain 2002)"

Oddly enough, the teachers found that although Shoko repeatedly ignored other students' opinions during teacher-led discussion, she did not ignore the teacher's comments (Nabei and Swain 2002). The researchers concluded that Shoko valued the teacher talk because the comments often reflected Ms. Johnson's intellect and culture (Nabei and Swain 2002). This is evident in Excerpt 3 when Shoko makes the following comment pertaining to the differences in culture

Third Excerpt

Since we are all from Japan, we compared bathing and showering in the Japanese context. But the teacher is not Japanese, and has experienced both Japanese and western bathing. She had different perspectives and ideas. I never thought of comparing western bathing and showering. I am very impressed. (Recall Session 4; 1 Nov.) (Nabei and Swain 2002).

The researchers observed that Shoko was periodically conscious of the teacher providing linguistic feedback in the teacher-fronted contexts (Nabei and Swain 2002). Nevertheless, her awareness of teacher feedback in teacher-fronted REs was often superficial as was seen the final excerpt (Nabei and Swain 2002).

Fourth Excerpt think Hisako said something like 'waste'. I don't remember what Hisako said, but I can recall that I was thinking what the teacher said was right. I mean, Ms Johnson changed the sentence and I thought she said it in a correct manner. (Recall Session 4; 1 Nov.) in this comment, Shoko was not precise about the nature of the teacher's feedback. She commented only that the teacher 'changed' something to make the previous utterance more target-like. Shoko had tuned out Hisako's presentation because it was less related to and urgent for Shoko. Thus, she sensed Hisako's linguistic problem but could not recall what exactly the problem was. A language problem faced by a peer interlocutor in her group, in contrast, was more immediate and demanding; thus, Shoko's attention to form and noticing was more concrete and precise (Nabei and Swain 2002)."

Advantages and Disadvantages

Although the stimulated recall method has been utilized quite often in second language research Gass and Mackey (2000) report that the method has been severely criticized throughout its history. According the authors the most problematic aspect of stimulated recall is whether or not the findings actually reflect the though processes of the participants. For instance the authors explain that the retrospective report compiled by the researcher "is directly accessible and available for verbal reporting. This assumption is better justified with only a small amount of intervening time between the event and the recall. Nonetheless, participants should be able to access some type of memory structures when instructed to report what they can remember about their thought processes during an event (Gass and Mackey (2000)."

Some researchers have asserted that the use of verbal reporting is an inaccurate way of determining thought processes because participants provide researchers with fabricated reports.

These researchers assert that it is nearly impossible to garner an accurate picture of cognitive thought processes through the use of the verbal reporting and introspection. The authors explain when individuals attempt to report on their cognitive processes, "they do not do so on the basis of any true introspection. Instead, their reports are based on a priori, implicit causal theories, or judgments about the extent to which a particular stimulus is a plausible cause of a given response" (p. 231). www.questia.com/PM.qst?action=getPage&docId=23153190&offset=1" 1 in other words, the verbal reports, perhaps tainted by inaccurate memory, contain (unknowingly) fabricated mental events. Part of Nisbett and Wilson's argument rests on the fact that conscious awareness can only relate to the products of mental processes; the processes themselves cannot be reached through introspection (Gass and Mackey 2000)."

In fact, the authors, Gass and Mackey (2000), agree that verbal report data may be unsuitable when directed at certain sorts of questions. This is particularly true of questions that involve abstract knowledge (Gass and Mackey 2000). Another researcher asserts that one spsicific issue associated with independent systems derives from requesting learners to retrospect on their performance (Gass and Mackey 2000).

Gass and Mackey (2000) do not disagree with the idea that retrospection is not a valid way of gathering information concerning the position of the proposed acquisition-learning dichotomy. However, the authors contend that the criticisms of retrospection are not always valid.

Gass and Mackey (2000) report that in the case of certain researcher the believe that,

The learner is asked to report what he was thinking at the time the particular language sample was produced, and this is sometimes days after the actual performance" (p. 186, emphasis ours). As we noted in earlier chapters, the timing between the event to be recalled and the recall itself is crucial. The fact that retrospective comments have been gathered inappropriately (e.g., without the recall prompt and with a significant time delay) does not invalidate the general procedure as a means of gathering important information (Gass and Mackey 2000)."

Although the disadvantages of this method are often recognized as being significant, there are also some advantages associated with the use of this method. The advantages of the stimulated recall method include the following.

Revelation of the information attended to during task performance- Learners and researchers are able to review the data in terms of the participants performance of the task (Gass and Mackey 2000).

Data collected is indicative of cognitive events - the data gathered may be correlated with the thought processes of the participants (Gass and Mackey 2000).

Reliability- because reports can be correlated with observed behavior (Gass and Mackey 2000).

Stimulated recall is useful in strategy research (Gass and Mackey 2000).

Stimulated recall is useful in assessing what prior knowledge is utilized in processing texts (Gass and Mackey 2000).

Think Aloud Method

According to Hamilton et al. The Think aloud method has been used extensively in the area of cognitive psychology. The authors assert that the method is respected because "it exposes thoughts while they are still in short-term memory and seems to produce relatively little interference with task requirements (Ericsson & Simon, 1993; Greeno & Simon, 1988). When used with assessment tasks, the procedure can identify misunderstanding of instructions, absence of requisite knowledge, and ways in which misconceptions can actually lead to correct answers (Hamilton et al. 1997)."

According to. Lorch and O'Brien, (1995) the think aloud method was first developed by 1995 Newell and Simon (1972) and was used to assess reading comprehension.

Lorch and O'brien (1995) assert that the think aloud method is of interest in the field of second language acquisition because it gives the researcher an instant picture of the comprehension of the learner (Lorch and O'brien 1995). The authors explain that in past studies researchers have evaluated the merits of protocol analysis in reading comprehension. They carried out systematic analyses of the inferences that occur in think aloud protocols during reading of narrative and expository texts. In their studies, subjects read one sentence at a time and thought aloud about each one as they read and tried to comprehend it. They were explicitly instructed to make inferences or to elaborate, to report on connections between the present sentence and prior ones, to make predictions on what might occur, and to comment on what they felt was the role of the current sentence in the overall organization of the text. In line with these instructions, the subjects frequently made reference to antecedent information, inferences, and predictions during reading. Olson et al. (1984) validated their verbal protocol data against reading time measures. Of interest is their finding that think-aloud inferences accounted for about 4% to 5% of the variance in the normal reading times for sentences Lorch and O'brien (1995)."

This finding suggests that when verbal methods involve questions or distinguished requests for information these processes are impacted when subjects are asked to think aloud (Lorch and O'brien 1995). Researchers have concluded that the instructing learners "to think aloud may force deeper and richer processing of the material than one would ordinarily obtain with normal reading (Lorch and O'brien 1995)."

Advantages and Disadvantages

The think aloud method has existed for many years, but there is still no consensus on the effectiveness of the method. Some tend to believe that the think aloud method paints an accurate portrait of what the learner has understood from the reading. Still others contend that there is no evidence to suggest that the think aloud method is any more effective in evaluating comprehension than recall or other methods. One of the obvious limitations of the method is that it cannot determine whether the information that the learner comprehends in the moment is retained. There is no way to know if they will actually remember what was learned.

Conclusion

The research examined three methods that are utilized in Second Language research including conversation analysis, stimulated recall and the Think Aloud Method. We found that each of these methods play a role in determining the interaction that takes place in a second language classroom. We also explored the relationship of each approach to theory and discussed the advantages and disadvantages of each system. The research supported these arguments through the use of transcribed extracts from second language classroom interaction.