INF5220 - Autumn 2006

Lecture 6 (October 4th )

First lecture hour:  A brief introduction to Grounded Theory, Ethnomethodology, Conversation Analysis, Discourse Analysis

Second lecture hour: A brief introduction to Action Research (to be continued)

Third lecture hour: Discussion of 3rd and 4th assignment
 

GROUNDED THEORY

Central book: Glaser and Strauss (1967): ”The Discovery of Grounded Theory. Strategies for Qualitative Research”. Aim: Generation rather than verification of theory. Testing of theories means rather: determine the empirical areas of application of theory. Middle-range theories

Central concepts of GT:
Coding of data: create categories which have properties

  • Open coding (substantive coding, first level of abstraction) Concept, category, properties
  • Axial coding (relationships between phenomena, concepts). Causal conditions, context, intervening conditions, consequences
  • Selective coding . Focus on core variable/concept

Theoretical sampling: purposively sample new data with the core concept in mind
Substantive versus formal theory:

  • Theory for a substantive (i.e. empirical) area
  • Theory for a formal (i.e. conceptual) area

Memo-writing: intermediate products from  ongoing analysis
"Memos are the theorizing write-up of ideas about substantive codes and their theoretically coded relationships as they emerge during coding, collecting and analyzing data, and during memoing" (Glaser 1998).

 
ETHNOMETHODOLOGY

Ethnomethodology = the study of people’s methods.  (Central persons: Harvey Sacks and Harold Garfinkel). The study of people’s everyday ways to produce orderly social interaction: How do people give sense to and accomplish their daily actions (communicating, making decisions, reasoning)? The skills and artful practices through which people come to develop an understanding of each other and of social situations. Ethnomethodology has an attention to details of talk-in-interaction. Focus on common-sense practices. Observable and reportable (speech and face-to-face behaviour).Answers the how-questions rather than the what-questions (of contextual givens).

Central concepts:

·        Indexicality, reflexivity, sequentiality

·        Membership categorization devices (MCD)

·        (Moral) accountability

·        Local practices – social order

·        Scenic display (the incongruity experiments, the technique of disrupting the taken-for-granted)

 

From  www.wikipedia.org:

Two central differences between traditional sociology and ethnomethodology are:

–        (1) While traditional sociology usually offers an analysis of society which takes the facticity of the social order for granted, ethnomethodology is concerned with the "how" (the methods) by which that social order is produced, and shared.

–        (2) While traditional sociology usually provides descriptions of social settings which compete with the actual descriptions offered by the individuals who are party to those settings, ethnomethodology seeks to describe the practices these individuals use in their actual descriptions of those settings.


A central method for ethnomethodologists is conversation analysis:

 

CONVERSATION ANALYSIS

Coherent discourse is produced according to some rules, the aim is to discover these rules, and describe the conversational structures they generate (The sequential and structural organisation of talk). How the participants structure their talk, how they use various resources.

 

Conversation analysis goes beyond a grammatical analysis of statements. Relies on detailed transcripts of conversation (naturally occurring or interviews). See symbols in appendix p. 376.

 

“There are recurring structural features in ordinary conversation, irrespective of the psychological characteristics of the participants. Conversation in context-bound, such that it is both productive and reflects the circumstances of its production. These two properties characterise all interactions, so that no detail can be dismissed as out of order, accidental or irrelevant to the ongoing interaction” (in Alvesson and Skøldberg, 2000, citing Holstein and Gubrium, 1994).

 

Some basic concepts: Turntaking and topic

•         General rule regulating turntaking: at least one and not more than one at a time

•         Utterances or turns as basic unit of analysis.

•         Conversation openings.

•         Adjacency pairs (e.g. greeting-greeting, question-answer, complaing-apology/justification). (NB. Phone greetings are differently structured from everyday greetings).

•         Utterance incompletors (‘but, and, however’).

•         Incompletion markers (‘if, since’).

•         Fillers (‘um, er, y’know’).

•         Where do interruptions occur?

•         Insertion sequences, side sequences.

•         Topic (tellability, newsworthiness).

•         Topic change, how does it occur?

•         Topic conflict

•         Story prefaces or floor seekers

•         How repairs are done. (to clear up misunderstandings, resolve disagreement etc).

•         The role of silences.

 

•         Symbols in appendix p. 376.


DISCOURSE ANALYSIS

•         Analysis of texts, talk, interviews: Concerned with what is performed in talk or writing, with the rhetorical and argumentative organisation of talk and texts, how text and talk is part of social practices.

•         Compared with CA, DA is often more concerned with topic closer to conventional social science, e.g. gender relations, social control etc. as well as more heterogeneous kinds of data.

•         Examples of focus in DA studies:

–        How are ‘versions of the world’ produced in discourse?

–        How are claims and versions constructed?

–        How are alternatives undermined?

–        How are each participants constructions accomplished and/or undermined?

–        How does a text tell the story and work up coherence and incoherence?

•         Some DA concepts: interpretive repertoires, stake and scripts.

–        Interpretive repertoires (classifications and category systems). How does participants define their moral status and identity? “Interpretive repertoires are systematically related sets of terms that are often used with stylistic and grammatical coherence and often organized around one or more central metaphors.” (e.g. motherhood)

–        Stake: how is a particular type of blaming achieved?  People treat each other as entities with desires, motives, institutional allegiances and so on, as having a stake in their actions. Referencing stake is one principal way of discounting the significance of an action or reworking its nature.

–        Script refers to the way participants construct events as scripted as instances of a general pattern, or as anomalies and exceptions. A script is a way of invoking a routine character of described events.

 

Example form Silverman 2001:

David Frost, interview with Salman Rushdie on Public Broadcasting Service, November 26th 1993 on the fatwa:

Frost: “And how could they cancel it now? Can they cancel it –

            they say they can’t?”

 

Rushdie: “Yeah, but you know, they would, wouldn’t they, as

            somebody once said. The thing is, without going into

            the kind of arcana of theology, there is no technical problem.

            The problem is not technical. The problem is that they

            don’t want to.”

 

Potter (1997): ”The familiar phrase ’they would, wouldn’t they’ treats the Iranians’ claim

as something to be expected: it is the sort of thing that people with that

background, those interests, that set of attitudes would say; and it formulated that

predictability as  shared knowledge. This extract illustrates the potential for

invoking stake to discount claims.”

 

Example from (Edwards, 1997), a couple in marital counselling:

Mary: ”I went out Friday night (.) and Jeff was working (.) on call (.)

         and (.) um (2.2) the place that I went to (.) like (.) closed at half past

         twelve and I got home about one o’clock.”

 

Mary mentions the time the place closed and when she got home.

It helps her to presented this as a routine set of events (a script) in which

nothing extraordinary or morally reprehensible took place.

 

Mary also explains the fact that she went out without Jeff: her evening out is not

to be heard as some wilful action of a woman ignoring her partner but as

something that was unavoidable (’scripted’) and therefore, morally acceptable.

 

DISCUSSION:

Can these approaches be useful to study and/or design information systems/information and communicaiton technology?

Responses: Yes, useful to study and learn how humans communicate in order to build communication technologies. + these perspectives may be resources for studying ”human – machine dialogues”: (Example: Lucy Suchman (1987): Plans and Situated Actions)


ACTION RESEARCH

Post-WWII: researchers should understand and remedy ”social illnesses” through intervention (USA: Kurt Lewin and UK: the Tavistock Institute)

Action research has been typified as a way to build theory, knowledge, and practical action by engagement with the world in the context of practice itself (Whyte et al. 1991)

Intervention as prerequisite for knowledge acquisition

Aims at achieving both practical and research objectives. Central principle: collaborative

Establishment of a ‘client-system infrastructure’ or research environment:

Specification of agreement between researcher and organization, e.g.   :

  • With what authority may the researcher/involved practitioners specify action?
  • What is the boundaries?
  • When and where do the researcher enter and exit?
  • How widely can findings be disseminated?

 

The five phases:

  • Diagnosis,  The joint identification of the primary problems
  • Action planning,  The interventions are planned
  • Action taking, The planned actions are implemented
  • Evaluation Evaluating the outcomes
  • Specifying learning  Specifying knowledge gained

 

Example of AR

The HISP project (www.hisp.info) started in South Africa in 1994

Open Source Software: DHIS - District Health Information System

Version 1.4 MS Access-based

Version 2.0 Fully platform-independent

Capacity building, training, research. Training of health workers at all levels (Health care managers, health staff). Building local technical competence in OSS technologies. Several countries in Africa and Asia. Large, > 70 Master and PhD students. Learnings gained from this AR project would not have been gained unless action had been taken, i.e. depend on the intervention.

Homework:

Readings:

Avison et al (Download from the Digital Library at www.acm.org, select "Magazines", find Communications of the ACM)

Baskerville & Wood-Harper (Handout) NB! Changed from reading list

Checkland & Holwell (Here: http://heim.ifi.uio.no/~petterog/Discussion/Checkland%20-%20Action_Research_Its_Nature_and_validity.pdf)

Optional reading: Kalleberg (Handout, in Norwegian)