Social Protocol Analysis: Reflections on a Methodology

1
Social Protocol AnalysisReflections on a
Methodology

• David Rossin

2
Reflective not Reflexive

• Paper is not about being reflexive i.e.
  offering alternative critical perspectives on
  methods and outcomes.
• But about giving a reflective account in terms of
  where the methodology fits into the wider
  research arena.

3
Historical Context of Social Protocol Analysis

• Protocol Analysis developed out of two fields of
  research Cognitive Psychology and Behavioural
  Decision Making theory.
• The main contributors were Newell Simon (1972)
  used volunteers to solve problems especially
  chess and cryptarithmetic involves thinking out
  loud, audio-recording and coding.
• Protocol Analysis has also been used to capture
  the cognitive processes of experts within
  business e.g. auditor, stockbroker, loan
  officer.

4
Theoretical Underpinning of Protocol Analysis

• Newell and Simons theory is based on the notion
  that a human is a processor of information (a
  computer!).
• The theory concentrates on the manipulation of
  symbols and symbol structures in the Information
  Processing System (IPS).
• Characteristics of a Information Processing
  System (IPS)
• Receptors (inputs)
• Effectors (outputs)
• Processor
• Short Term Memory
• Long Term Memory
• The behaviour of the person is said to be
  adaptive motivated towards a goal. (Problem
  solving is always a matter of search)
• The problem space is the internal symbolic
  representation of the task environment. It is
  comprised of
• Knowledge Elements - initial states (inputs) and
  desired outcomes (outputs)
• Operators obtaining new information from the
  old (process)

Knowledge Elements
Knowledge Elements
Operators
5
Adaption of Protocol Analysis to Social Problem
Solving Settings

• The notion of the problem space remained the same
  i.e. operators, knowledge elements and goals
• The analogy of the computer now relates to a
  network.
• Symbols can be transferred between individual
  IPSs.
• Greater emphasis is placed on sensory perception.
• Greater emphasis is placed on language (group
  interactivity).
• Non-verbal communication is important in group
  settings.

6
Why use Social Protocol Analysis (SPA)?

• Satisfies the call for more rigour and depth in
  IS case study research that produce thick
  descriptions and deeper insights. (Walsham,
  1995)
• It was argued that interviewing and other
  retrospective techniques are not sufficient by
  themselves
• SPA is good at
• Identifying the information requirements of
  individual members
• Identifying the information requirements of the
  group as a whole
• Understanding the political aspects of their
  information requirements

7
The Case Study

• SA Ltd (pallet manufacturer) annual turnover 4
  million
• Sponsor Managing Director
• Improve the information for the steering
  committee meetings (traditional MIS)
• Semi-informal interviews with participating
  managers
• Two Managing Directors, Finance Director,
  Production Manager, and Sales Marketing Manager
• Video and audio recording 6 steering committee
  meetings over a 6 months period (over 12 hours of
  recordings)
• The role of the researcher Putters Was I a
  fly on the Wall?

8
(No Transcript)
9
Transcribing the Social Protocol
Segment Transcript (June Meeting)

• The coding of the transcript is based on
• Fishers (1970) units of interaction
• Potter Wetherells (1987) Discourse Analysis
  (in the transcript)
• Bales (1950) Interaction Processing Analysis
  (Transaction Code)
• Newell Simons (1972) Protocol Analysis
  (Operator Code)
• Key II Information Item
  APJ Judgement on Actual
  Performance
• CAP Calculation on Actual
  Performance IA Information Assessment

10
Wider Research Framework

• Roots in Cognitive Psychology and Behavioural
  Decision Making theory ideographic research.
• Strong case for it to be regarded as an
  ethological research method a naturalistic
  context.
• The research could be regarded as an ethnography
  BUT I was NOT a participant observer.
• The SPA Methodology can be regarded as an
  ethnomethodology (Garfinkel, 1967) where
  interaction is seen as a natural process of
  everyday living (social reality).
• Ethnomethodology has its links with Conversation
  Analysis (turn taking) rather than Discourse
  Analysis.