Data Analysis

Abstract

The interviews were analysed in four steps. First, Nvivo, a qualitative software program, was used for coding and categorising of the data. The purpose of this was to develop a hierarchical structure of themes with systematic searches to find common usage of the meaning of commitment amongst respondents. Parallel to this procedure, I read all interviews three times and listened to all audio recordings two times so as to get a better understanding of the interview content as a whole and sense possible inter-relationships between the interviews. Second, narrative reports were compiled. These narratives helped to gather a more structured outline of the meaning of the interview text. Third, comparison of narrative reports, transcripts and field notes were made. This stage involved a first inductive open coding so as to derive a codebook, core categories and subcategories of commitment behavior, and triggered outcomes. Open coding involved “breaking down,” examining, comparing, conceptualizing, and categorizing data“ (Strauss and Corbin 1998). Finally, I started to notice that a categorization scheme was starting to appear from the interviews. The code development was the most time-consuming part of the analysis process. The confirmatory stage was deduced by analyzing if the collected data overlapped with Meyer and Allen’s (1997) OC conceptualization. Multiple methods to increase reliability and validity of the coding procedures were applied and are explained in depth in later sections. Fourth, data was analysed by identifying the means-end chains of meaning for individual respondents.