Case Study Literature Methods Yin 2003

From COST Action FP0804: FORSYS
Jump to: navigation, search

Sean Gordon's notes from this textbook:

Yin, R.K. 2003. Case study research : design and methods. Thousand Oaks, CA: Sage Publications.

Foreword

  • the core of the scientific method is not experimentation per se but rather the strategy "plausible rival hypotheses."

Introduction

  • types of case studies: exploratory, descriptive, explanatory
  • how and why questions are more explanatory and likely to lead to the use of case studies
  • deal with operational links needing to be traced over time, rather than mere frequencies or incidents
  • research questions have both substance (what is my study about?) and form (am I asking a who, what, where, etc. question)
  • how or why questions being asked about the contemporary sets of events, over which the investigator has little or no control
  • traditional prejudices
  • lack of rigor, generalization (but generalizable to theoretical propositions like experiments, not to populations or universes); analytic generalization not statistical generalization
  • investigates a contemporary phenomenon within its real-life context, especially when the boundaries between phenomenon and context are not clearly evident
  • many more variables of interest than data points
  • relies on multiple sources of evidence, triangulation
  • benefits from the prior development of theoretical propositions to guide data collection and analysis
  • in the evaluation research: explain, describe, illustrate, explore, meta evaluation

Designing Case Studies

  • study design: a logical model of proof
  • components: study questions, propositions (theories), units of analysis, logic linking the data to the propositions, criteria for interpreting findings
  • descriptive theories 29
  • criteria for quality
    • sound operational procedure (construct validity), proof of causal relations (internal), replicability (reliability), and generalization possible (external validity).
    • ..\..\Research\Prelims\Gordon prelim - Bliss questions.doc
  • case study designs
    • single: critical, unique, representative, revelatory, longitudinal
    • holistic versus embedded
    • multiple: literal versus theoretical replication; replication backed sampling logic, feedback can alter sampling (but be clear about reasons)

Conducting case studies: preparing for data collection

  • case study research is among the hardest types of research to do because of the absence of routine formulas
  • researcher training and case study protocol especially important if multiple researchers or case studies involved
  • skills: ask good questions, good listener, adaptive, understand issues, unbiased
  • read between the lines
  • immediate interpretation
  • knowledge needs: why the study is being done, what evidence is being sought, what variations are anticipated (and how to accommodate), what constitutes supportive and contrary evidence
  • protocol outline
    • purpose, theory and background
    • data collection procedures
    • outline of the case study report
    • case study questions (interviewees, individual case, pattern across cases, entire study, recommendations and conclusions) and sources of evidence
  • case screening: ask key informants, avoid complex screening
  • pilot case study

Conducting case studies: collecting the evidence

  • six sources: documents, archival records, interviews, direct observation, participant observation, physical artifacts
  • principles:
    • using multiple sources of evidence (triangulation: did I sources, investigators, theories, methods of),
    • creating a case study database: Notes documents and interviews
    • maintaining a chain of evidence: report -- database -- protocol -- questions
  • interviews: open, focused, survey; recording

Analyzing Case Study Evidence

Miles & Huberman (1994)

  • putting information into different arrays
  • making a matrix of categories and placing the evidence within such categories
  • creating data displays -- flowcharts and other graphics
  • tabulating the frequency of different events
  • examining the complexity of such tabulations and their relationships by calculating second-quarter numbers such as means and variances
  • putting information in chronological order

Three General Strategies

  • relying on theoretical propositions
  • thinking about rival explanations (with or without an initial theory)
    • nine types of rivals
  • developing a case description (descriptive research)

Specific Analytic Techniques

  • pattern matching (internal validity)
    • non-equivalent dependent variables as a pattern
    • rival explanations as patterns (pattern of independent variables that is mutually exclusive)
  • explanation building (special type of pattern matching, more difficult)
    • presumed set of causal links
    • iterative nature
  • time series analysis
  • logic models
    • pattern matching with sequential stages
    • individual, organizational, or program level
  • Cross case synthesis

Pressing for a high-quality analysis

  • show that you attended to all the evidence
  • address all major rival interpretations
  • address the most significant aspect of your case study
  • use your own prior, expert knowledge

Reporting case studies

  • compositional structures: linear-analytic, comparative, chronological, theory-building, suspense, unsequenced
  • targeting audience: academic, nonspecialists, funders
  • formats: single case, multiple case (Cross case analysis), question-answer, Cross case analysis
  • integration with other methods
  • anonymity
  • review and validation (participants)
  • exemplary case studies: significance, completeness, consideration of alternative perspectives, sufficient evidence, engaging composition