The strategic management as an academic discipline began to take shape in the early years of the 1960s and early authors such as Anshen et al. (1973) and Meyer (1978) noted the need to direct research efforts to shape the theoretical corpus of strategic management. In May 1977, Professors Dan Schendel and Charles Hofer run at the University of Pittsburg a conference entitled 'The State of the Art Business Policy and Planning Research" which represented an important milestone in the consolidation of strategic management as an emergent academic field (Saunders et al, 1980).
Schendel and Hofer (1979) recommended the segmentation of strategic management in research lines, which was fundamental to foster research on these topics: strategy concept, strategy formulation, strategy evaluation, strategy content, goal formulation/structures, social responsibility, environmental analysis, public policy, strategy implementation, formal planning systems, strategic control, strategic management process, board of directors, general management roles, entrepreneurship and new ventures, multibusiness/multicultural firms, strategic management in not-for-profit organizations, and research methods. Then, the Strategic Management Journal appeared in 1980 and quickly became the flagship of the scientific community in this area (Guerras-Martin et al, 2013). The growth of scientific production and the increased number of international journals on strategic management fostered the consolidation of the research lines and also led to the emergence of diverse additional lines of research such as competitive advantage and resource based view of the firm among others. The initial efforts gave birth to the current structure and defined the internal dynamics of the strategic management field.
Many efforts have been devoted to trace the evolution of the discipline in order to contribute to its future development. For example, Ramos-Rodriguez and Ruiz-Navarro (2004) determined the changes in the intellectual structure of the discipline by analyzing co-citations. Later on, Nerur et al. (2008) employing of co-citations with multidimensional scaling and factor analysis expanded the results obtained. They employed publications and authors as the units of analysis. Both works are attempts to define the intellectual structure of the strategic management field using quantitative methods under the assumption that papers or authors who are frequently co-cited are somehow intellectually related. These works present interesting insights of the discipline and contributed to know how strategic management has developed between 1980 and early 2000s as well as defining distinct subfields and their relationships through the citations. However, the results do not address the consistency and interconnectedness between the lines of research that form the theoretical basis for the strategic management field and how advanced is each research line.
Other authors, e.g. Boyd et al. (2005), analyzed the development of the field of strategic management in the first 25 years of life. Interestingly, the authors found that many strategy faculty members never publish in their career and article writing activity, as well as success, may be determined by the orientation and prestige of the institution where faculty is based. Moreover, they found that strategic management has characteristics of a young and advanced discipline simultaneously due to weak consensus and low productivity but the research outcomes are based on merit, respectively. However, the conclusions provided are general and not focused on the lines that conform the discipline, which have different levels of maturity.
Therefore, there is a need to take stock of the development of the research lines that conform the strategic management field nowadays if we want to offer interpretation, evaluation, and challenges to existing areas of strategic management research. This paper will not only reflect the current state of the field but also have the potential to stimulate and guide future research efforts by illuminating and offering insight into the core strategy concepts and theoretical streams. To address what we know and do not know from the existing research and direct attention to development of the theories within and across the different subfields, we perform a content analysis of all articles published in the SMJ from its inception until 2013 by combining bibliometric and social network analysis. In short, this paper aims to answer the following research questions: What are the current research lines that form the theoretical framework of strategic management? How advanced are the research lines that make up the intellectual structure of strategic management as an academic field of research?
We have conducted a content analysis of all articles published in the Strategic Management Journal from its founding to December 2013. The criteria for focusing on the Strategic Management Journal (SMJ) in this review are: 1) SMJ has been employed in source similar studies such as Nag et al (2007); Furrer et al (2008); and Ronda-Pupo et al (2012); 2) SMJ is recognized as the most influential journal on strategic management in the world (Azar et al, 2008; Franke et al, 1990) and is among the most influential journals in the field of business management (Tahai et al, 1999); and 3) the journal has contributed to the formation and development of the research community on strategic management in the world (Ronda-Pupo et al, 2010). Other journals, such as Academy of Management Journal or Academy of Management Review, are not included because not all articles published in these journals address issues related to the discipline (Nag et al, 2007).
The collection, preparation and analysis of the data followed a five steps process.
First, we downloaded all articles and research notes published in SMJ from the database ISI Web of Knowledge. Then, we obtained the citation count using Social Sciences Citation Index. Since SMJ covers a time span of 33 years we decided to segment this time frame into 3 stages of approximately 11 years to trace the dynamics of the evolution of each line of investigation.
Second, using content analysis, we created a matrix of co-occurrence with concepts that appeared two or more times in the titles and/or abstracts of each paper. We defined a lower boundary of ten appearances for each term since a term appearing less than ten times in 33 years is not relevant. To avoid the possible omission of important concepts and to ensure the reliability of the terms we used the lexicon of the field of strategic management from previous work such as Nag et al. (2007: 942, table 2); Furrer et al. (2008: 7, table 2) and Ronda-Pupo et al. (2012: 187, Appendix 1a). We have added to the analysis 249 key words that co-appear in the abovementioned papers. The terms corresponding to each research lines can be observed in the appendix.
Third, the co-occurrence matrix was normalized using cosine. We employed VOS (van Eck et al, 2010), which is a multi-dimensional scaling technique, to cluster the terms, and manually ensure the correct assignation of terms to clusters. The objective of clustering terms was to identify groups that are tightly linked and may correspond to areas of interest or display strong levels of activity (Callon et al 1991). The terns in the matrix of co -occurrences were grouped into 14 research lines that make up the current field of strategic management (see appendix 1). The number of times each term appeared can be observed in appendix 2.
Fourth, using social network analysis, we calculated the internal cohesion of each research line considering the density of the network that conform each line. Density is the strength of the links that tie together the terms that makes up an investigation line. Then we compare the centrality of the research line in relation with other research lines to develop a map of the strategic management field. Centrality measures the strength of an investigation line's interaction with other lines of investigation. Both operations were performed with Pajek (Batagelj et al, 1998). Then, we created a map clustering the terms belonging to a particular line of research and defining their importance in terms of centrality and density (Callon et al 1991). The x axis corresponds to the centrality of the research line and the y axis to its density, thereby forming four quadrants. The intersection point corresponds to the median of the centrality value, and the median of the density value. The top right quadrant shows lines which are central and with high density. The quadrant on bottom right side depicts lines that are central but they are still underdeveloped (low density). The top left quadrant presents peripheral but highly developed (high density) lines. Finally, the quadrant on the bottom left displays peripheral and under developed research lines.