Article Preview
TopIntroduction
The creation and accumulation of knowledge is affected by a variety of innovations on the technological and societal levels. One opportunity to gain, develop, and record this knowledge lies in its preservation in the form of publications as well as in the analysis of the publications’ content. Within the knowledge management (KM) domain, the literature provides an increasing number of publications (Lloria, 2008). Accordingly, recent research highlights the importance of applying text-analytical approaches to the KM domain and affirms their role as an enabler of effective and qualitative knowledge (Khan & Vorley, 2017).
Considering the increasing number of publications, content analysis is a popular and empirically established research method to address context-sensitive information and achieve an improved understanding of its content (Mayring, 2010). Content analysis can be applied both manually (Bontis, 2003; Heisig, 2009) and automatically by using software solutions (Ribière & Walter, 2013). The latter type provides advantages over the former, such as decreased analysis time and increased objectivity of the outcomes (O’Flaherty & Whalley, 2004). Accordingly, automated content analysis and other text-analytical approaches are continuously gaining importance (Fisher, Garnsey, Goel, & Tam, 2010).
Despite the widespread use of text-analytical approaches, a problem prevails concerning the use and analysis of compound concepts. Automated solutions typically yield frequency counts for single words only (e.g., organizational learning vs. organizational and learning). While the use of a dictionary (i.e., a list of relevant single concepts as well as compound concepts such as organizational learning) might solve this problem (Boritz, Hayes, & Lim, 2013; Ceci & Iubatti, 2012; Gottschalk & Bechtel, 2005; Vasalou, Gill, Mazanderani, Papoutsi, & Joinson, 2011); however, recent studies analyzing KM research (e.g., Ribière & Walter, 2013; Wallace, Van Fleet, & Downs, 2011) do not apply dictionaries. One reason for this shortcoming might be the effort required to develop an adequate dictionary, which does not exist for the KM domain so far.
This study pursues the two objectives of (1) developing a dictionary of single and compound concepts relevant to the KM domain and (2) using the dictionary for automated content analysis of KM research. For this purpose, this article uses 4,290 journal and conference papers published in the top ten KM journals (Serenko & Bontis, 2013) between 2002 and 2014 and the proceedings of the European Conference on Knowledge Management between 2006 and 2013. As a result, this work proposes an overview of key research topics and trends within selected premier outlets in the KM domain. The trends can be derived by observing the frequencies of concepts that appear throughout the entire publication sample. The higher the frequency count, the higher its relevance for the community appears to be. With the results, the study thus aims to address the following research questions:
How does the research landscape for the KM discipline look, and what are main topical foci of the underlying publication efforts?
The remainder of this paper is structured as follows. Next, the evolution of applying content analysis, in particular in the KM domain, is reflected. The study proceeds by describing the development of the dictionary and the approach to automated content analysis. In the subsequent section, the results of the content analysis concerning the articles published between 2002 and 2014 in selected KM outlets are described. Then, implications for research and practice as well as limitations of the study are discussed. The paper ends with a short conclusion.