A thematic analysis process for a structured review was followed (Dixon-Woods et al. 2005), in which relevant literature identified previously was classified as having discussed one or more approaches that contributed to an understanding of disciplinary development. A data-driven approach was adopted, in which the themes emerged from the data. The analysis did not reflect the frequency of the themes, but instead accepted themes that offered a ‘high level of explanatory value’ (Dixon-Woods et al. 2005:47). The relationship between the identified components was also examined. As it was considered necessary to examine the views of researchers from IS and other disciplines, it was acknowledged that any classification of approaches to disciplinary development derived from the literature was unlikely to be complete, due to the quantity of publications available on the topic. It is argued, however, that only sufficient analysis is needed to identify the main issues when reviewing disciplinary development, until theoretical saturation is achieved, as is done when working towards concept development in primary qualitative research (Dixon-Woods et al. 2005). Therefore any omission of literature in the area is unlikely to weaken the analysis and classification process. The purpose of distinguishing the themes identified from the IS literature from those of the wider literature was to allow an evaluation of the degree of consistency between IS researchers and those from other disciplines. If the framework developed matched the components identified from both groups of literature, and the relationship among the components, the robustness of the framework would be strengthened.

As classification is a largely subjective process, two trained IS researchers categorised the literature independently. Discussion took place where the classification differed, until agreement was reached. The major themes derived from the literature, social processes and a core body of knowledge, were used for analysis. As foreshadowed, the latter category was broken down into four subgroups: research and teaching methods and standards (or IS knowledge gain and transfer), a unique symbol set, key research and teaching topics (or IS knowledge domain) and theoretical issues. To acknowledge the importance to IS of providing professional training, any literature that considered teaching methods and quality in the development of the discipline was grouped with the research and teaching methods and standards category, while literature on teaching relevance was grouped with key research and teaching topics. The impact of local influences was also sought.

A record was made for each reference to these themes, by author. The results were delineated further by whether the author came from the IS discipline or elsewhere, as determined by the publication outlet. The relationship between the themes was captured by recording the combination of issues discussed with reference to disciplinary development, for each author.

Table 3.1 sets out the results of the analysis of the literature, where the components explicitly examined in discussion of the nature and development of disciplines were identified and then classified. Themes taken from a review of the IS literature have been distinguished from those that were derived from the more general literature. The order of listing in the table matches that followed in the earlier discussion.

• derived from IS literature

o derived from non-IS literature

As all the categories were found in the IS and the general literature, this finding was interpreted to mean that both groups shared a common view of the components used for explaining the nature and development of a discipline. Consequently, it was assumed that the components identified were robust and appropriate for inclusion in a framework of the development of a discipline. From the number of studies examined, and the results, it was assumed that saturation of topics had been reached. As many authors identified more than one component, a combination of components was considered most appropriate to account for the nature and development of the IS discipline.