falon deimler methodological workshop presentation
TRANSCRIPT
CASE STUDY AND CROSS-CASE
ANALYSIS
Methodological WorkshopFalon Deimler
Definition of a case:An intensive description and analysis of a single individual (also known as life histories), group, community, or event within its real-world context
Case Study Methodology
CS Key Components: Design
Types & uses of case study based on purpose strategic selection explaining the case
vs elaborating theory Frame/design
feature Single or
multipleHolistic or embedded
Researcher’s role
CS Key Components: Theory Development
Developing a theoretical perspective in conjunction with other tasks
Build, extend, or challenge perspectives – but may limit ability to make discoveries
Protocol: mental framework/line of inquiry
CS Key Components: Data Collection Mixed methods opportunities
Direct or participant observation, survey, interview, statistics, archival record/physical artifact analysis, etc.
Triangulation Three or more independent sources point to
same set of events, facts, or interpretations Rival explanations
Are events and actions as they appear to be? Are participants giving candid responses?
CS Key Components: Data Analysis Pattern-Matching
Internal validity Explanation-Building
Causal links – “how” and “why” questions Time-series Analysis
Simple or Complex Chronology
Logic Models Chain of events; repeated cause-effect-cause-
effect patterns Cross-Case Synthesis/Replication Logic
(multiple case studies)
CS Key Components: Communicating Findings
Avoid mixing evidence and interpretation Social usefulness
Historical context, implications for practice
Case Study Methodology
Exemplars Theoretical and
methodological robustness Innovation Challenge theoretical
assumptions Study rare phenomena Explanatory power “Doubleness” Social usefulness/praxis Ethical/value-laden
Difficulty drawing definite cause-effect conclusions
“N of 1” problem = Generalizing from a single case?
Biases in collection and interpretation of data
Researcher’s own “methodological savvy”
Advantages Limitations
Definition:A scientifically rigorous methodology adapted
from cross-case synthesis (Yin) that seeks to “[mobilize] case knowledge” through the use of theory, methodological and data source triangulation, rules of evidence, and systematic case study protocols and databases to ensure consistent/reliable data collection
Cross-Case Analysis Methodology
CCA Key Components: Techniques
Variable-oriented cross-case comparison Outcomes observed vary in cases Randomness & representation
Mills’ Methods (1843) Comparative method of agreement/disagreement
Case Survey (Yin 1994, 2003) Evidence from large set of cases for statistical
analyses Before-After research design
Case divided into two sub-cases; before/after investigation
CCA Key Components: Techniques cont.
Case-oriented cross-case comparison Commonalities across multiple instances of phenomenon
contributes to conditional generalizations Most Different Design (Przeworski and Teune, 1982
Similar processes/outcomes in diverse sets of cases Typologies (Denzin, 1989; George and Bennett, 2005)
Clusters or families of phenomena that share patterns Multicase Methods (Stake, 2006)
Quintain: a common focus for a set of case studies with both common and unique issues
Process-Tracing (George & Bennett, 2005) Trace progression of events that may have led to
outcome in a single case to map out one or more causal paths
CCA Key Components: Techniques cont.
Mixed data display and analysis techniques Combination of variable and case-oriented approach Visualize sets of cases, help relationships surface
Stacking (Miles and Huberman, 1994) Series of cases displayed in meta-matrix for
systematic visualization and comparison Qualitative Comparative Analysis-QCA (Ragin,
1993) Relationships arranged in “truth table” by variable to
study common causes/outcomes; conjunctions used to local relationships
Constructing Narrative Models (Goldstone, 1997) Encapsulates case as a storyline to preserve essence
Five Steps for Cross-Case Study Methodology
Developing a theory of change Establishing a measurement framework Developing a cross-case study protocol
and building a database Analyzing and interpreting findings Communicating results
Five Steps for Cross-Case Study Methodology: Developing a Theory of Change
Represents a complex hypothesis to be tested
Measure of internal validity Analytic generalization
Conceptual claim of relationships btw concepts, constructs, event sequences
Apply theoretical propositions to outside case study
Community and systems change efforts Which strategies lead to what outcomes at
which levels?
Five Steps for Cross-Case Study Methodology: Establishing a Measurement Framework
Mixed methods and multiple data sources Reveal converging/diverging patterns Increase confidence in interpretation Tell the story
Triangulation to address construct validity “Sophisticated and rigorous”
Five Steps for Cross-Case Study Methodology: Developing a Cross-Case Study Protocol and Building a Database
Protocol Ensures systematic collection and analysis procedures
across cases Makes explicit
How and when data will be collected How data will be stored, managed, analyzed, interpreted,
and reported Who will review drafts The format for reporting the findings The rules of evidence Procedures
Database System for organizing and storing data
Five Steps for Cross-Case Study Methodology: Analyzing and Interpreting Findings
Attend to all evidence Codify data, identify a pattern, determine a
theme, draw inferences within a case Capture emergent themes Document frequency (triangulation) Chain of evidence Address major rival interpretations
Five Steps for Cross-Case Study Methodology: Communicating the Results
Use own prior knowledge to articulate awareness
Usefulness for wide audience Present data in stakeholders’ terms Preserve uniqueness of cases; avoid
contextual stripping
Cross Case Analysis
More intensive examination of phenomenon across a number of cases
Telling a story Visual display
facilitates comparison: “comparability” or “translatability”
Advancing community and systems change
Tensions between particularistic case knowledge and multiple case study research
Destruction of integrity Generalizations as
half-lives Fluid environments
Advantages Limitations
Cross-Case Analysis: Example
CORE ELEMENTS:
• Systems Approach
• Community Collaboration
• Prevention• Intervention
Safe Start Demonstration Project A community and systems
change effort to address issues related to young children’s exposure to violence and the adverse outcomes on their development
Eleven communities (urban, tribal, rural)
Example: Theory of Change
Example: Establish Measurement Framework
Example: Developing Protocol and Database
Protocol How and when Which instrument to use How data will be stored
and managed How data will be
analyzed and interpreted
How findings will be reported
How findings will be used
Protocol• Letter to proj director, email
and call to interviewees• Checklist, interview guides• Interviews transcribed and
stored within 5 days• Codes assigned to data
sources• Glossary of terms, outlines• Inform devo of training
agendasDatabase• Built for each site, every document given a
unique label, list of all documents and identifiers generated
Example: Analysis and Interpretation of Findings
Codified every piece of data relevant to components and linkages in theory of change
Themes validated by repeated comparison Case became a unit of analysis for cross-case
with analytical strategy and pattern-matching technique
Identified common strategies, activities, and community conditions that led to higher numbers of children served
Identified unique elements that affected outcomes in different sites
Example: Communicating Results
Organized to demonstrate how change process proceeded – interlaced data to support themes and conclusions
Cross case as core of final report, 11 case studies included as an appendix
What are some of the defined boundaries (historical, spatial, etc.) of the context/issue?What theories (theory of change, techniques/approaches) were used to frame the study? Does this frame seem at all limiting? What methods of data collection and analysis were used? Were these methods best suited for maximizing the “social usefulness” and reliability of the study?What differences are there in the potential “uses” of the single case studies conducted initially vs. the cross-case analysis?Does the cross-case analysis appear to threaten the integrity of the cases, strip the cases of their context, or over-generalize?Additional questions regarding this design?
Exercise 1 & Discussion
http://www.qualitative-research.net/index.php/fqs/article/view/334/729#g5
Exercise 2 & Reflection
References Denzin, N. K. (1989). Interpretive interactionism. In Gareth Morgan (Ed.), Beyond methods:
Strategies for social research (pp.129-146). Beverly Hills, CA: Sage Publications. Flyvbjerg, B. (2001). Making social science matter: Why social inquiry fails and how it can succeed
again. Cambridge: Cambridge University Press. George, A.L. & Bennett, A. (2005). Case studies and theory development in the social
sciences. Cambridge, MA: MIT Press. Goldstone, J.A. (1997). Methodological issues in comparative macrosociology. Great Britain: JAI
Press. Hodgetts, D. J., & Stolte, O. E. E. (2012). Case-based research in community and social
psychology:Introduction to the special issue. Journal of Community & Applied Social Psychology, 22(5), 379-389.
Kahn, S. & VanWynsberghe, R. (2009). Cultivating the under-mined: Cross-Case Analysis as Knowledge Mobilization. Forum: Qualitative Social Research, 9(1).
Kegler, M.C., Rigler, J., and Honeycutt, S. (2011). The role of community context in planning and implementing community-based health promotion projects. Evaluation and Program Planning, 34(3): 246-253.
Lee, K. S., & Chavis, D. M. (2012). Cross-case methodology: Bringing rigour to community and systems change research and evaluation. Journal of Community & Applied Social Psychology, 22(5), 428- 438. doi:10.1002/casp.1131
Miles, M.B. & Huberman, A.M. (1994). Qualitative data analysis. Thousand Oaks, CA: Sage Publications.
Mills, J.S. (1843). A system of logic. London: John W. Parker. Przeworski, A. & Teune, H. (1982). The logic of comparative social inquiry. Malabar, FL: Robert E.
Krieger Publishing Co. Ragin, C. (1993). Introduction to qualitative comparative analysis. In Thomas Janoski & Alexander
Hicks (Eds.), The comparative political economy of the welfare state (pp.299-319). New York: Cambridge University Press.
Stake, R. (2006). Multiple case study analysis. New York, NY: Guilford Press. Stoecker, R. (1991). Evaluating and rethinking the case study. The Sociological Review, 39(1), 88-
112. Yin, R. (2012). A very brief refresher on the case study method. Ch. 1 in Applications of Case Study
Research, 3rd ed. Sage Publications, Inc. Yin, R. (2014). Analyzing case study evidence: How to start your analysis, your analytic choices,
and how they work. Ch 5 in Case Study Research: Design and Methods, 5th Ed. Sage Publications, Inc.
Yin, R. (1994). Case study research: Design and methods (2nd ed.). Thousand Oaks, CA: Sage Publications.
Yin, R. (2003). Case study research: Design and methods (3rd ed.). Thousand Oaks, CA: Sage Publications.