Thematic Analysis

Conceptual Overview and Discussion

Although widely used, thematic analysis generally has not been well described. It is not a research method in itself but rather an analytic approach and synthesizing strategy used as part of the meaning-making process of many methods, including case study research. Richard Boyatzis describes five purposes of thematic analysis: it is a means (1) of seeing, (2) of finding relationships, (3) of analyzing, (4) of systematically observing a case, and (5) of quantifying qualitative data. As a sensemaking approach, thematic analysis is a tactic for reducing and managing large volumes of data without losing the context, for getting close to or immersing oneself in the data, for organizing and summarizing, and for focusing the interpretation.

A wide range of data sources may be used in a thematic analysis, including interview transcripts, field notes, information written by participants (e.g., diaries or journals), research memos, historical or site documents, photographs, drawings, maps, digital audio files, and video files. Historically, researchers have applied thematic analysis primarily to textual data and have transformed audio or video records to text via transcription prior to analyzing for themes. However, some computerassisted qualitative data analysis software now offers the possibility of coding themes directly within digital audio and video files. NVIVO is an example of Computerassisted qualitative data analysis software specifically designed for thematic analysis of qualitative data, with theory-building capabilities. The features of Computerassisted database management, including coding, linking, searching, and model building, facilitate rigorous and sophisticated thematic analyses, even for large, unstructured data sets and across sites and research teams.

The basic analytic strategy used in thematic analysis is coding, a process of closely inspecting text to look for recurrent themes, topics, or relationships, and marking similar passages with a code or label to categorize them for later retrieval and theory-building. Identification of themes can be done deductively, on the basis of theoretical constructs that the case study researcher wishes to investigate. Researchers might use their research questions, interview questions, or theory-derived categories as a start list of a priori themes for coding data documents, an approach that can facilitate within- or cross-case comparisons.

However, an inductive approach to coding is more typical of thematic analysis. Themes emerge from and are grounded in the data. Through a process of noticing patterns, attending to how participants label events, defining emergent themes, constantly comparing data against codes and categories, cycling back through documents to revise coding, recording interpretive insights in research memos, and developing data displays that reveal overarching patterns, the researcher builds a complex exploratory, descriptive, or explanatory case analysis grounded in the particulars of the case or multiple cases. Inductive thematic analysis avoids the rigidity and premature closure that are risks of a deductive approach.

...