Digital Conversation Analysis — Studying Talk in Online Environments
Digital Conversation Analysis · Also known as: DCA, online conversation analysis, digital CA, computer-mediated conversation analysis
Digital Conversation Analysis (DCA) applies the systematic, turn-by-turn analytical procedures of Conversation Analysis (CA) to digital and computer-mediated interactions — including chat logs, social media threads, instant messages, and online forums. Rooted in the foundational CA framework of Sacks, Schegloff, and Jefferson, DCA adapts classical concepts such as turn-taking, adjacency pairs, and sequential organisation to account for the asynchronous, multimodal, and textual character of online communication.
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 Digital Conversation Analysis when the research question concerns how social actions — requesting, agreeing, rejecting, joking, supporting — are accomplished sequentially in digital text-based interaction, and when naturally occurring data are available. It is well-suited to studying online communities, customer service chats, social media comment threads, and instant messaging. It is not appropriate when the data are researcher-generated (e.g., survey responses or interview transcripts), when the goal is to count or compare frequencies across a large corpus (use quantitative content analysis instead), or when the research question concerns attitudes, beliefs, or themes rather than interactional sequence and social action.
Strengths & limitations
- Keeps analysis tightly tethered to what participants actually do in real digital exchanges, avoiding researcher-imposed interpretive frameworks.
- Reveals how platform-specific affordances — threading, reply-quoting, emoji, edit history — actively shape interactional organisation.
- Applicable to large digital corpora that are already textually transcribed, removing the transcription burden of spoken CA.
- Generates fine-grained, practice-level findings about online social interaction that survey or interview methods cannot produce.
- Works across diverse platforms and genres — customer support, online forums, social media, group chats.
- Naturally occurring digital data raise significant ethical and access challenges, particularly on closed platforms or private messaging.
- The absence of prosody, gesture, and facial expression removes analytic resources that classical CA relies on; digital CA must develop alternative criteria for sequence boundaries.
- Analysis is labour-intensive and detail-focused; large-scale corpora cannot be fully analysed using CA's close-reading approach alone.
- Asynchrony and the possibility of simultaneous posts complicate the sequential logic that CA was designed to describe.
- Findings are context-specific and platform-dependent; generalisability across platforms or communities requires comparative work.
Frequently asked
How is Digital Conversation Analysis different from standard Conversation Analysis?
Standard CA was developed for spoken, face-to-face interaction and relies on prosodic and embodied cues to identify turn boundaries and sequence structures. Digital CA applies the same sequential logic to written, computer-mediated interaction, which is often asynchronous, lacks prosody, and is shaped by platform-specific affordances such as threading, reply-quoting, and emoji. The analytic procedures are adapted accordingly, but the fundamental focus on sequential action remains the same.
Can I use DCA on social media data I collect via an API?
Yes, API-collected data from platforms such as Twitter/X, Reddit, or Discord can serve as a DCA corpus, provided the thread structure and timestamps are preserved so that the sequential ordering of turns can be reconstructed. Ethical considerations — including platform terms of service, participant anonymisation, and informed consent norms for public versus private data — must be addressed before analysis.
How large should a DCA corpus be?
There is no fixed rule. Because DCA analyses individual sequences in close detail, even a small number of carefully selected exchanges can generate substantive findings. Many published DCA studies work with dozens to a few hundred turns rather than thousands. The goal is to identify and demonstrate patterns across multiple instances drawn from naturally occurring data, not to achieve statistical representativeness.
Does DCA require a theoretical background in linguistics?
A working familiarity with CA's core concepts — turn-taking, adjacency pairs, preference organisation, sequential implicativeness — is essential. These concepts are grounded in sociology and linguistics but are explained accessibly in introductory CA texts. Formal linguistic training is not required, but analysts who are unfamiliar with CA's analytical vocabulary should invest time in foundational reading before attempting a study.
Is DCA compatible with other qualitative methods?
DCA can be combined with digital ethnography to provide contextual background for interpreting sequences, or with discourse analysis for broader discourse-level patterns beyond individual exchanges. However, its sequential, action-oriented logic does not integrate straightforwardly with thematic or coding-based methods. When mixed with other approaches, the researcher should be explicit about which analytic layer each method addresses.
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 ↗
- Herring, S. C. (2007). A faceted classification scheme for computer-mediated discourse. Language@Internet, 4(1). link ↗
How to cite this page
ScholarGate. (2026, June 3). Digital Conversation Analysis. ScholarGate. https://scholargate.app/en/qualitative/digital-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.
- Conversation AnalysisQualitative↔ compare
- Critical Discourse AnalysisQualitative↔ compare
- Digital EthnographyQualitative↔ compare
- Discourse AnalysisQualitative Research↔ compare
- Thematic AnalysisQualitative Research↔ compare