A bibliometric analysis of corporate governance and ownership research.

• Author: Joao, Belmiro N.
• Position: Report

1. INTRODUCTION

   Corporate governance and ownership (CG&O) is exposed to criticism with political and corporate failure. Similarly, the field of CG&O research is subject to criticism as well. While encompassing contributions from many disciplines including economics, finance, accounting, management, and law, some doubt arises whether CG&O research is in fact a discipline in and itself. In this sense, CG&O research is the subject of a multi-disciplinary research rather than a discipline.

   This paper aims to identify the current state of this academic literature regarding CG&O, and thus focus and analyze its knowledge base by using bibliometric analysis which meets the purposes of description, evaluation and scientific monitoring. In previous studies, bibliometric analyses were used to investigate, for example, the knowledge base of corporate governance (Durisin and Puzone, 2009), the trends and contributions in International management research Acedo and Casillas (2005), the research in international business (Griffith, Cavusgil & Xu, 2008) and dynamic capabilities (Stefano, Peteraf & Verona, 2010). This inquiry is motivated by two research questions:

   (1) Which recent contributions have been driving the research agenda for CG&O?

   (2) Which emerging themes in the literature are likely to set the stage for future work?

   This paper is structured in the following manner beyond this introduction: In the next section sources of data and method are described, with the search terms used for retrieving the data and social network analysis tools used for analyzing it. The section 3 provides the results from the analyses as a discussion and limitations as well. Finally, findings of the paper and discussing viable alternatives for further research.

2. METHOD

   Based on the bibliometric principle that knowledge of disciplines is concentrated in only a small proportion of important journals, citation data from ISI (Institute for Scientific Information) Web of Science (2010) was retrieved and used here to carry out the searches and retrieve publication data. All three available citation index databases were used in the search: Science Citation Index Expanded (SCI- Expanded), Social Sciences Citation Index (SSCI) and Arts & Humanities Citation Index (A&HCI). Although ISI's Web of Science nowadays covers around 7,500 journals from all fields of scholarship, it does not claim to provide a complete coverage of all journals that are used in scholarly research. Instead, it claims to include the most important or useful ones. These citation indexes also contain a record of the references cited by the authors of the covered publications. This enables the use of cited reference searches and various citation analyses.

   In ISI Web of Science, using a search string to search from Topic includes in the results hits in publication title, abstract, author keyword. Therefore, the final search string for Topic with "Corporate Governance" and "Ownership" was formulated as:

   Topic=("Corporate Governance") AND Topic=(Ownership) Refined by: Document Type=( ARTICLE) AND Subject Areas=( BUSINESS, FINANCE OR ECONOMICS OR BUSINESS OR MANAGEMENT) Timespan=All Years. Databases=SCI-EXPANDED, SSCI, A&HCI. [1]

   Only articles were included. Book reviews, editorial material, meeting abstract, proceedings were put aside and the search was not restrained to any specific years, thus retrieving everything from 1945 to 2010. Other restrictions were applied for subject area. Only papers in "business, finance" or economics or business or management. All searches were performed on September 1st, 2010.

   This resulted in 1092 hits, describing the whole body of publications mentioning the Topic. The 1092 publications contained over 28003 citations (4,572 citations discarded with inconsistencies). In this analysis, we elaborated a co-citation network using the so called Jaccard index as a normalized co-citation strength measure (S) in order to emphasize close relations between similar references that are not cited as often as the most common references (Gmur, 2003).

   S = [N.sub.AB]/([N.sub.A] + [N.sub.B] - [N.sub.AB]) [2]

   where [N.sub.AB] = the number of common citations to articles A and B; [N.sub.A] = the number of citations to A, and [N.sub.B] = the number of citations to B.

   A dense sub-network grouping algorithm, yielding a number of independent densely connected groups and a list of disconnected nodes (Schildt and Mattsson 2006), was used to distinguish highly cited groups of references. It is assumed that these groups represent the different intellectual bases that participate in the discussion evolving around the concept of knowledge work. The dense sub- network grouping analysis is implemented in a bibliometric software tool that was used to produce the co- citation network data. Network analysis were conducted using Ucinet (Borgatti et al., 2002) a social network analysis software. These text files were then imported to...