Coding, Flexible

Flexible coding describes coding approaches that allow categories to emerge from data. It is characterized by a dynamic and nonlinear process in which researchers incorporate and develop new categories, ideas, and concepts from their interactions with the texts. Flexible coding is both exploratory, aiming to identify emerging themes and commonalities, and generic, generating insights and building theories from contextualized raw materials. Theories, instead of playing the role of deducing coding categories as in fixed coding, are inductively formulated and refined through the coding processes. In other words, whereas the goal of fixed coding is to confirm a theory, flexible coding is to build a theory. Flexible coding is mostly used in research broadly situated in an interpretive-hermeneutic approach. Researchers doing thematic analysis or open coding in the grounded theory approach, for example, employ methods of flexible coding. This entry explains three types of flexible coding categories: organization, substantive, and theoretical. Next, a description is provided for how researchers should interpret and code data, followed by a brief review of some criticisms of flexible coding.

In flexible coding, the researcher is no longer supposed to be an objective, neutral observer who follows explicit rules to place unitized data into predefined categories. Instead, categories are induced from the data, and are reconceptualized, merged, partitioned, reorganized, and reconstructed throughout the coding process. Joseph Maxwell describes three types of coding categories that are employed in qualitative research: organization, substantive, and theoretical. Organizational categories are broad areas or issues relevant to the topic for initial data sorting. Substantive categories contain descriptive concepts or labels that tap participants’ beliefs and understandings. Theoretical categories are the researcher’s conceptualization and are building blocks of the theory developed in the study. It is important to move from organizational categories to the latter two categories as researchers accumulate more data for in-depth theoretical insights.

Although the researcher’s interest, background, and initial objective for the study play a role in deriving concepts and developing theory, they are keenly attentive to what the data suggest as the key themes and ideas and are open to incorporating unanticipated results. Researchers are expected to situate their readings and interpretations in the chosen context, carefully analyze the data, and treat coding as a process in which both research questions and answers to these questions arise and inform each other. In flexible coding, understanding is temporary and interpretation is tentative until the texts are saturated with emergent ideas and notes and codes become repetitious. Researchers often tentatively arrive at a certain code or conceptual category and experiment with its boundary and labeling until they fit the data. Researchers are also encouraged to frequently interrupt the coding to record theoretical notes that have been triggered in the process. Research questions, concepts, and categories are constantly subject to alternative interpretations and discordant evidence and are revised accordingly throughout the coding process. With insights obtained later in the coding process, researchers can go back to reanalyze previously coded texts and modify their interpretations. As opposed to the fixed coding process that follows a sequence of analytical steps, the flexible coding process is nonlinear and iterative.

...