Multiple Case-Based Conversation Analysis
Also known as: multi-case CA, cross-case conversation analysis, comparative conversation analysis, multiple-instance CA
Multiple case-based conversation analysis applies the fine-grained sequential methods of Conversation Analysis (CA) across two or more distinct cases — settings, groups, or interactions — to identify both case-specific patterns and cross-case regularities in naturally occurring talk. By examining how participants organise turn-taking, repair, and action sequences in multiple contexts, the approach strengthens claims about interactional phenomena beyond what a single-case study can establish.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use multiple case-based CA when your research question asks about an interactional phenomenon that may vary across settings or participant groups and you need more than one case to assess its generality. It is appropriate when you have access to recorded naturally occurring talk from at least two distinct contexts and the goal is to describe systematic patterns of talk-in-interaction rather than to measure frequencies or test hypotheses. Do not use this approach when talk is elicited or scripted (e.g., survey interviews or read-aloud tasks), when the research question concerns attitudes or beliefs rather than interactional organisation, or when only a single setting is available — in that case a single-case CA is more honest about its scope.
Strengths & limitations
- Enables cautious cross-contextual claims about interactional phenomena without sacrificing CA's sequential precision.
- Deviant case analysis across multiple settings sharpens theoretical understanding of when and how a phenomenon occurs.
- Anchors claims in naturally occurring talk, preserving ecological validity that laboratory or survey methods cannot match.
- Purposive case selection allows the researcher to build theoretical arguments about contextual variation rather than merely describing it.
- Replicated patterns across independent cases strengthen the credibility of CA findings beyond a single-instance study.
- Transcription to full Jefferson notation across multiple cases is extremely time-consuming and requires specialist training.
- Findings describe interactional organisation, not participants' intentions, attitudes, or experiences — complementary methods are needed for those questions.
- Purposive case selection does not support statistical generalisation; the logic is theoretical, not population-based.
- The number of cases is constrained by the feasibility of full CA-level analysis; large-n comparison is not practical within this approach.
Frequently asked
How many cases are needed?
There is no fixed minimum, but at least two cases are required by definition. In practice, three to six cases allow meaningful comparison while remaining analytically tractable. The selection criterion is theoretical relevance, not statistical power — each case should be chosen because it can illuminate something about the phenomenon that the others cannot.
Is Jefferson notation mandatory?
For rigorous CA, yes — Jefferson notation captures the sequential and prosodic details that CA analysis depends on (overlap onset, micro-pauses, turn-final intonation, etc.). Simplified transcripts may be sufficient for coarser descriptive purposes, but they limit the depth of sequential analysis and can mislead comparison across cases.
How does this differ from comparative discourse analysis?
Conversation analysis focuses exclusively on naturally occurring talk, examines interaction at the level of individual turns and sequences, and treats participants' own orientations as the primary evidence. Comparative discourse analysis works with a broader range of text types, uses more varied analytic frameworks, and typically asks about meaning or ideology rather than interactional organisation.
Can I combine multiple-case CA with quantitative methods?
Yes — mixed-design studies that combine qualitative CA of sequential structure with quantitative description of frequency or distribution are established in interactional linguistics. The key is to keep the CA analysis primary and use frequency data only after the sequential pattern has been established qualitatively; reversing that order risks producing misleading counts of poorly defined categories.
What software supports CA transcription?
CLAN (from the CHILDES project), ELAN (for video-linked transcription), and Transana are widely used tools that support Jefferson notation and link transcript to media. They assist with data management but do not automate the analytic work, which requires the researcher's sequential and interactional expertise.
Sources
- Sacks, H., Schegloff, E. A., & Jefferson, G. (1974). A simplest systematics for the organization of turn-taking for conversation. Language, 50(4), 696–735. DOI: 10.2307/412243 ↗
- ten Have, P. (2007). Doing Conversation Analysis: A Practical Guide (2nd ed.). Sage. ISBN: 978-1412922579
How to cite this page
ScholarGate. (2026, June 3). Multiple Case-Based Conversation Analysis. ScholarGate. https://scholargate.app/en/qualitative/multiple-case-based-conversation-analysis
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Comparative Conversation AnalysisQualitative↔ compare
- Conversation AnalysisQualitative↔ compare
- Critical Discourse AnalysisQualitative↔ compare
- Discourse AnalysisQualitative Research↔ compare
- Multiple-Case StudyQualitative↔ compare