Digital Reflexive Thematic Analysis

The analytic logic and six-phase process are identical. The difference lies in the data source and the contextual considerations it demands: digital data require attention to platform affordances, the public-private spectrum of online spaces, multimodal signals, and platform-specific ethical norms that do not arise with conventional interview transcripts.

There is no single rule. Public posts are generally considered accessible, but verbatim quoting can make participants identifiable via search. Best practice is to paraphrase or lightly alter quotes when direct identification is a risk, disclose the platform and community in the paper, and consult your institutional ethics board. Private groups, DMs, or forums requiring login require active consent.

Unlike interview-based methods, saturation in the sampling sense is not the target: in reflexive thematic analysis, depth of engagement matters more than volume. A corpus of 50–500 posts is typical for a single-platform study, but what matters is that the dataset is rich enough to generate analytically meaningful themes relative to the research question. Extremely large datasets are often reduced purposively before analysis.

Yes — qualitative data software can assist with organising, coding, and retrieving digital text. However, the software does not generate themes; the analyst does. Automated text-mining or sentiment analysis is a different paradigm and should not be conflated with Digital RTA.

It can be either. An inductive approach lets themes emerge from the data without a prior coding frame, suitable for exploratory studies. A deductive approach applies a theoretical or conceptual lens to the data from the outset. Braun and Clarke recommend being explicit about which stance you adopt and why.