____________________________________________________________________________________________________

Subject PSYCHOLOGY

Paper No and Title Paper 3: Qualitative Methods

Module No and Title Module 23: Conversational Analysis

Module Tag PSY_P3_M23

TABLE OF CONTENTS
1. Learning Outcomes
2. What is Conversation Analysis?
3. History of Conversation Analysis
4. Basic Concepts

4.1 Turn Taking and Turn Construction Units

4.2 Adjacency Pairs
4.3 Preference and Dispreference Organizaion
4.4 Repairs

5. Steps Involved in Conversation Analysis

6. When to Use Conversation Analysis

7. Evaluation of Conversation Analysis

PSYCHOLOGY PAPER No. : PSY_P3: Qualitative Methods
MODULE No. : M23: Conversation Analysis
____________________________________________________________________________________________________

7.1 Advantages

7.2 Disadvantages

7.3 Comparison with

7.3.1 Discourse Analysis

7.3.2 Thematic Analysis

7.3.3 Grounded Theory

8. Conclusion

9. Summary

1. Learning Outcomes (Times New Roman , size 14)

After studying this module, you shall be able to

· Know the origins of Conversation Analysis

· Learn the steps involved in Conversation Analysis
· Identify the pros and cons of this method
· Evaluate by comparison with other approaches

2. What is Conversation Analysis?

A Conversation refers to a talk, especially an informal one, between two or more people, in
which news and ideas are exchanged. It is an impromptu, spontaneous talk between two or more
people on an everyday basis.

Conversation Analysis hence is the study of the talk; it is the approach to the study of the natural
conversation carried out in a particular manner. It is a naturalistic, observation-based science of
actual (verbal and non-verbal) behavior, which uses audio and video recordings of naturally
occurring interactions, meetings, therapy sessions, telephone calls and the like, as the basic form
of data. Conversation analysis assumes that the conversation is sequential and its structure is rule-
bound and meaningful to the participants in the conversation. It studies the conversation as a
behavior and the thrust of conversation analysis lies in the belief that the conversation itself is
sufficient to understand the principles of what happens when people converse. Simply, it takes the
conversation at its “face value” rather than taking the hidden motives, temperaments, personality
traits etc into consideration.

3. A Brief History

PSYCHOLOGY PAPER No. : PSY_P3: Qualitative Methods

MODULE No. : M23: Conversation Analysis
____________________________________________________________________________________________________

Harvey Sacks (1935-1975), the American sociologist, is

considered to be the pioneer in the field of conversation
analysis. Sacks believed that sociology could be a `natural observational science', one that would
be able to handle the details of actual events, `formally and informatively' (Sacks,
1984). This is the context in which he began to work with tape-recorded conversations. The
following extract helps to clarify his position:
“Such materials had a single virtue, that I could replay them. I could transcribe them somewhat
and study them extendedly ± however long it might take. It was not from any large interest in
language or from some theoretical formulation of what should be studied that I started with tape-
recorded conversations, but simply because I could get my
hands on it and I could study it again and again, and also,
consequentially, because others could look at
what I had studied and make of it what they could, if, for
example, they wanted to be able to disagree with me.” (1984:
21-7)

Sacks’ work in the field of conversation analysis has been

influenced by the two strands of socio-logical thinking – social
interactionism as explained by the sociologist Erving Goffman
and ethnomethodology of Harold Garfinkel.
Erving Goffman considered social interaction as a social
institution which has its own organisation in terms of norms
and moral obligation like other social institutions such as the
family, economy, religion etc. this is not reducible to the
psychology of any individual involved in the interaction-
Fig.1 Harvey Sacks (1967)
Source: www.google.com

an assumption of conversation analysis too. On the other hand Harold Garfinkel’s key concern
was to understand the social interaction in everyday life. The focus of ethnomethodology was
the ordinary conversation. The word ethnomethodology indicates Garfinkel’s method of studying
the common sense methodology used by people to conduct their social interactions.

In the 1960s, Sacks along with Emanuel Schegloff (his associate and friend from postgraduate
days) began to investigate conversations in a Suicide Prevention Centre in Los Angeles by
analysing telephone calls by suicidal people or people who were involved in some way with a
suicidal person. Sacks’ began to investigate how callers’ accounts of their troubles were
produced in the course of their conversations with the suicide prevention centre counsellors. This
provided the stimulus for what was to become Conversation Analysis.
After Sacks’ died in a car accident, his associates Gail Jefferson and Emanuel Schegloff took over
and continued his work in this field. Jefferson‘s major development was the system of
conversation transcription which is virtually, nowadays, universal in conversation analysis work.

4. Basic Concepts

4.1 Turn Taking and Turn Construction Units (TCUs)

In conversation analysis, conversation comprises of turns. A turn refers to a ‘turn to speak’ or a
turn- at- talk. That is, a turn is person’s conversation before a different person takes over the
conversation. The primary focus of conversation analysis is on adjacent turns in order to
PSYCHOLOGY PAPER No. : PSY_P3: Qualitative Methods
MODULE No. : M23: Conversation Analysis
____________________________________________________________________________________________________

understand how the second turn is ‘designed’ to fit with the

previous turn. A speaker who has taken over in the
conversation has both the right and obligation to include a turn construction unit (which is
basically a complete speaker turn) which includes a ‘transition relevance space’ at which the next
speaker will begin to speak. (Howitt, 2010)

4.2 Adjacency Pairs

Adjacency pairs are based on the notion that many turns are actually pairs- the two turns are by
different speakers, they come next to each other in the basic form and the two turns belong to the
same type. Adjacency pairs may seem simple but they have considerable impact within a
conversation because they organize the later turns by setting up expectation about the later turns.
Adjacency pairs can be of many kinds such as, question-answer, invitation-acceptance, request
for favor-granting/rejecting, apology-acceptance, complain-denial, complain-apology etc.

An example of question-answer adjacency pair is:

Q. How are you?

A. I am fine.

4.3 Preference and Dispreference

Organization

In a conversation, turn taking assumes that

there is a preference for particular turns to be
followed by a particular kind of second turn.
These are conversational preferences and not
psychological ones. Like, in the above example
the preferred response to the turn i.e., question
‘how are you?’ is ‘I am fine’.

4.4 Repairs
In conversation analysis, the analyst looks for
evidence of ‘repeated patterns’ which are a
result of the participants’ effort to progress the
conversation. One such pattern is repairs. It is the process by which the participants in a
conversation go about correcting the error made by them during the conversation. It is argued
that the pattern is for the members of a conversation to prefer the repair to be done by the person
who made the error in the first place. (Howitt, 2010) For example:

A: How many subjects do you have in this semester?

B: umm I have six subjects….

A: Six?

B: Oh. Seven subjects.. Fig.2 Cartoon depicting few common words used

PSYCHOLOGY PAPER No. : PSY_P3: Qualitative Methods

MODULE No. : M23: Conversation Analysis
____________________________________________________________________________________________________

Source: www.google.com
There are places in a conversation where it is possible that another speaker will take over, such
places are referred to as ‘transition relevance spaces’ for example in the repair pattern, there
might be a very short gap before the next person speaks which is meant for the correction of the
error. Usually, if there is transition from one speaker to another it occurs smoothly without either
a gap or an overlap of speakers.

5. Steps involved in Conversation Analysis

5.1 Producing the recording

The recording is vital in conversation analysis. Indeed the analyst may not prefer to “collect data”
in the sense of making a recording of conversation themselves .Instead ,it is a part of the tradition
of conversation analysis that the researcher uses extant recordings in the form say, of telephone
conversations routinely recorded by organizations .The ethos in conversation analysis encourages
the choice of naturalistic or naturally occurring conversations whenever available. Of course,
what comprises of a natural conversation is a problematic issue, in itself. The conversation
analyst does not need to be present at the conversation, say, making notes and the analyst would
see no point in doing so. Though the recordings are done by machine but humans decide what is
to be recorded and when it is to be recorded .Apart from this, a recording is basically unselective
and what is recorded is not a product of the conversation analyst’s questions and it has not gone
under a tidying-up process .There is no embargo on the use of other data and you will find good
examples of the researcher-initiated (e.g. focus groups) being used for conversation analysis.
There are choices to be made, and as long as the choices are consistent with what the researcher
wishes to attain, the details of the data are a factor in the analysis and not an obstacle.

5.2 Transcripts
In conversation analysis, the recordings are always transcribed, now-a-days in Jefferson transcription
system. Transcription systems cannot, by definition, completely capture what is on the recording any
more than the recording is an exact reproduction of the original conversation. The transcription
system imposes its own characteristics on the data which usually does not cause any difficulty.
However, the transcript is not complete and the analyst may need to check the transcription against
the original recording where there may be doubts. Still, the individual transcriber may produce a
transcript different from other transcribers of the same recording. Therefore, conventionally the
researchers transcribe their own data .This helps the analyst to generate familiarity with the data and
also makes the analyst familiar with what can be heard on the recording having the transcript in front
of them. Nevertheless, it means that the researcher is encouraged to focus on the rich but minute
details of the conversation.

PSYCHOLOGY PAPER No. : PSY_P3: Qualitative Methods

MODULE No. : M23: Conversation Analysis
____________________________________________________________________________________________________

Fig. 3 A few Symbols used in transcription and their meanings

Source: www.google.com

5.3 Selecting the features of the transcription to study

The researcher should concentrate on the features of the conversation selected for analysis. It is
not analytically useful to try and deal with the entirety of the conversation and, consequently it is
necessary to select a feature (or possible features) to form the starting point of the analysis. There
are some basic principles which can be applied to a conversation in order to identify features of
the research such as:

· Parts of the conversation which do not seem to be going well are a likely area of focus.

PSYCHOLOGY PAPER No. : PSY_P3: Qualitative Methods

MODULE No. : M23: Conversation Analysis
____________________________________________________________________________________________________

· Once some basic understanding of the conversation

analysis theory has been achieved, then this may suggest issues for considerations.
· There may be parts if the conversation which the researcher simply may not understand.
It is unclear what conversational processes are involved.

5.4 Making sense /interpreting the conversational episode

The process of interpretation involves a detailed reading and re-reading of the conversation to
familiarize oneself with what is going on in the conversation. An attempt has to be made to use
the content of the conversation to access how the members of the conversation are making sense
of each of the turns and other elements of the conversation. The researcher’s personal
understanding of the conversational exchanges as a person generally involved in conversations is
significant. The researcher’s general knowledge of research and theory in conversation analysis is
required.

5.5 Explicating the interpretation

The conversation analyst is a member of the broad community whose conversations are the focus
of study. Consequently, the researcher’s native or common sense understanding of what happens
in an episode from a conversation is an important resource. Nevertheless, this is incomplete as an
explication without bringing the links between this resource and the details of the conversation
episode together. That is, the analyst needs to demonstrate how their understanding links to the
details of the conversation. The analyst needs to be sensitive to the fit of their analysis to the data
but prepared for instances where the analysis and the data simply do not work together. It may
also involve revising the explication.

5.6 Elaboration of the analysis

Once a particular episode, has been analyzed, it has to be set in the context of the rest of the
conversation transcription. This may allow a further understanding on the episode in question
though equally, the entire transcription may raise questions about the adequacy of the analysis so
far. Later sequences in the conversation may in some way, directly or indirectly, relate back to
analyst’s chosen and key conventional episode.

5.7 Comparison with episodes from other conversations

A conversation need not end with a particular conversation and the researcher’s analysis of it.
There is the question of whether other instances of conversations, seemingly similar, support or
detract from the analysis .This stage is extremely important because individual episodes are not
considered in conversation analysis to be unique .The mechanisms and ways in which a
conversational is both produced by conversationalists and understood by other conversationalists
are expected to re-occur across different conversational episodes. Comparisons may be made to
find similarities and dissimilarities which may help refine the analysis.

6. When to Use Conversation Analysis

PSYCHOLOGY PAPER No. : PSY_P3: Qualitative Methods

MODULE No. : M23: Conversation Analysis
____________________________________________________________________________________________________

Conversation analysis should not be considered as an approach

to qualitative data analysis, but rather an approach to
understanding language, conversation, and how the latter evolves through the production of
sequences and structures. To understand conversation analysis one must, first, understand the
roots of this approach in ethnomethodology. Ethnomethodology seeks to find the methods used
by ‘members’ in everyday interaction which achieve the sense of order often called ‘society’
(‘ethno’ methods). Note that, it does not emphasize the consideration of psychological notions
like motivation, attitudes, etc. Like any other approach, conversation analysis, too, has a limited
area of usage, beyond which it would be rendered significantly useless.

Conversation analysis can be used when:

a) The data are in the form of a conversation. This automatically means that the researcher does
not refer to other sources of information like written documents (like diaries, journals, etc.),
photographs, etc.

b) The aim is to determine the characteristics of a particular institutional activity, and not to make
sense of larger entities of social action on multiple levels of organization.

A few examples where conversation analysis is used are therapy sessions, medical consultations,
courtroom trials, press conferences, interviews of various kinds, help-line talk, computer talk,
interaction in the airline cockpit, classroom interaction and politicians’ talk etc.

7. Evaluation of Conversation Analysis

7.1 Advantages of conversation analysis

· Scientific rigor: Standard transcription systems allow for comparability and replicability
between researchers.
· Transcripts become a public record: The transcripts are available for other researchers to
study. Conversation analysts often share data and collaborate in developing an
understanding of interactional phenomena.
· Recordings can be replayed: Recordings can be studied many times over.
· Transcripts can be re-analyzed: Transcripts can be re-analyzed in different ways. The
researcher is not limited by the original transcription.
· Different sequences can be selected: A different researcher can choose different
sequences of utterances to the first researcher.

7.2 Disadvantages of Conversation Analysis

· Time-consuming: Transcription of data is time-consuming and it is therefore difficult to

analyze a large body of data.
· Analyzing visual data is difficult: It can be difficult for a researcher using visual data to
see everything, given the limitations of technology. Visual transcription systems are still
being developed.

PSYCHOLOGY PAPER No. : PSY_P3: Qualitative Methods

MODULE No. : M23: Conversation Analysis
____________________________________________________________________________________________________

· Transcription is a subjective process: Transcription

needs to be seen as a ‘kind of mediation between the raw data, the recordings, and the to-
be-constructed images’ (ten Have 1999:28).

Sources: Silverman 2000:49; ten Have 1999

7.3 Comparison With:

7.3.1 Discourse Analysis

Conversation analysis, by Harvey Sacks(1984), emphasizes on the natural conversation that is

carried out in a particular manner, whereas discourse analysis refers to a number of approaches
that aim to analyze written, vocal, or sign language use; it can include discourse, written material
and conversation. Unlike the conversation analysis, discourse analysis studies language beyond
the structure and sequences of conversation. This further ensures that the discourse analysis takes
into account the core sociological concerns (gender, class and power).

The method of conversation analysis aims at determining how the participants produce
conversation and make sense of the conversation produced by the other participants in a
conversation. It is does not seek to examine the circumstances around which the conversation is
produced; and only concentrates on the methods used by the participant to make conversation,
unlike the discourse analysis. In summary, discourse analysis examines broad categories, and
conversation analysis (although more difficult to employ) examines a narrow set of categories.
However both the conversation and discourse analysis consider “talk” as a mode of social
interaction, even though the two approaches handle this mode in different ways.

7.3.2 Thematic Analysis

Thematic analysis has existed as a method for a long time virtually without any known
advocates. As a consequence its methodology has been much deferred until recently Conversation
analysis emerged and developed in the late 1960’s in the context of developments of sociological
theory. Both conversation analysis and thematic analysis are descriptive rather than theory –based
approaches. Both thematic and conversation analysis are built on a great deal of analytical work
on the part of the researcher. The researcher should be familiar with the content and the analysis
is based on detailed, intense hard work of the analyst. Even the smallest of the details are taken
into account and nothing is ruled out.

7.3.3 Grounded Theory

Grounded theory concentrates on deriving a theory from the data, gathered in a systematic way
and analyzed through a research process. It goes in depth and looks at the meaning of the
conversation and tries to develop an understanding of the data to emerge a theory out of it. On the
other hand, conversation analysis just analyses the conversation at its face value and is not
bothered with its meaning. It is the surface study of the conversation like the gaps, words,
punctuations etc. used. No theory is evolved here. Conversation analysis hence is a theoretically
embedded approach to understanding the structuring and sequencing of the conversation.
Therefore, unlike grounded theory, it is difficult to evaluate conversation analysis as a general
method for the qualitative psychologist’s toolbox.
PSYCHOLOGY PAPER No. : PSY_P3: Qualitative Methods
MODULE No. : M23: Conversation Analysis
____________________________________________________________________________________________________

8. Conclusion
Conversation analysis can therefore be seen as theory of language rather than a research method.
As it is a logically consistent and stringent approach to understanding to how conversations
works .The close analysis of the data is the primary focus of conversation analysis. Nothing is
ruled out and every portion of the data is given equal importance. It is an un conventional
approach and does not draw on psychological explanatory concepts such as motives and feelings.

9. References

1. Smith, J. A (2003) (Ed.) Qualitative psychology: A practical guide to research

methods.
2. Howitt, D. (2010) Introduction to Qualitative Methods in Psychology.

PSYCHOLOGY PAPER No. : PSY_P3: Qualitative Methods

MODULE No. : M23: Conversation Analysis