شناخت مزایای بالقوه مثلث بندی با مرور قدرت و ضعف روش های تحقیق آرشیوی، رفتاری و کیفی

ABSTRACT A review of the strengths and weaknesses of archival, behavioral, and qualitative research methods: recognizing the potential benefits of triangulation

Accounting is an applied discipline, within which researchers rely on theories from a wide array of root disciplines to investigate research questions. Perhaps because of the variety of root disciplines that have been influential in helping to motivate accounting research (particularly economics, psychology, and sociology), accounting researchers employ a variety of research methods to conduct empirical studies of accounting phenomena. Some of the most common research methods can be broadly classified as archival (secondary sources of, mostly, numeric data at the organizational unit of analysis), behavioral (primary sources of, mostly, numeric data3 at individual, group and organizational units of analysis), and qualitative (secondary and primary sources of, mostly, non-numeric data at individual, group and organizational units of analysis).

Currently, researchers using archival research methods vastly out-number those who use behavioral or qualitative techniques. A survey by Koonce and Mercer (2005) of five of the top accounting journals showed that from 1993-2004, over 94% of financial accounting studies used archival research methods, compared to less than 6% that employed behavioral methods.

Similarly, a more comprehensive survey of 14 accounting journals from 1981-2004 by Merchant and Van der Stede (2006) found that very little accounting research used field research and other qualitative techniques. Part of the reason for the dominance of archival research methods concerns the teaching philosophies of accounting many doctoral programs – whereas virtually all U.S. accounting doctoral students take at least one course emphasizing archival research methods, fewer than half are typically exposed to behavioral methods (Koonce and Mercer, 2005) and only a small number are trained in qualitative research techniques (Merchant and Van der Stede, 2006).

Equally troubling is that doctoral students’ training often emphasizes the research method rather than the research question. As stated by Koonce and Mercer (2005, p. 177), “Accounting doctoral students typically choose one of two, largely non-overlapping, fields of specialization…this choice is often determined by the type of method the student expects to use in his or her coursework [archival or behavioral].” Appropriate research methods should certainly match a study’s research questions. However, given that most future accounting researchers are trained to examine a problem within the confines of a particular research method, most researchers rely on a singular method in examining their research questions, and limit their views of their particular paradigm based on the research method used (see Burrell and Morgan,4 1979). This limitation is perpetuated by the inherent difficulties in developing and maintaining skill sets in multiple methods.

The purpose of this paper is to discuss the benefits, limitations and challenges in developing a well-designed study using any single research method in isolation. In particular, raising accounting researchers’ awareness of the limitations of their “dominant” research method is important, as it can to help foster deeper understanding of the potential contributions of other techniques. While researchers within the management accounting discipline have long recognized the importance of using integrative research methods (e.g. the call for cross-sectional field study research in Lillis and Mundy, 2005), this paper strives to show how triangulation1of research methods (using two or more research methods to study a given topic) can produce a richer, more complete understanding of all accounting phenomena. Currently, the academic accounting community emphasizes the use of archival research methods at the expense of behavioral and qualitative research. By discussing the shortcomings in each of these types of research methods, this paper aims to illustrative how multiple papers on the same research topic should use differing methods in order to overcome the weaknesses in collectively relying upon a single approach to research.

This paper is especially aimed at less-experienced academics (i.e. doctoral students and new faculty members) as such individuals can benefit from understanding how each research method presents its own set of challenges, and why investigations of accounting phenomena through multiple methods helps present a more comprehensive view of a specific phenomenon. The discussion of alternative research techniques within this paper can also be informative to more experienced researchers who wish to investigate research questions using methods that are outside of the focus of their initial training.5

Many previous authors have certainly noted the strengths and weaknesses of archival, behavioral, and qualitative research methods in the social sciences in general (e.g. Kerlinger and Lee, 2000; Shadish et al. 2002; Berg, 2003; Greene, 2003) and in accounting research in particular (e.g. Kothari, 2001; Libby et al. 2002; Merchant and Van der Stede, 2006). While all of these articles and textbooks constitute important examinations of specific research methods, none provides a combined overview of all of these different methods in the context of accounting research. This paper therefore aims to synthesize information on the benefits and limitations of research methods as used in accounting research within the framework of a single paper, which overviews the research methods field and provides new scholars with a single resource for beginning their investigations of how to conduct research.

The remainder of this paper is organized as follows: The next three sections detail the challenges involved in conducting a well-designed study using archival, behavioral, and qualitative research methods. The subsequent section provides a brief example of how examining a research topic with multiple methods can help produce a more complete understanding of the underlying accounting phenomena. The final section concludes with a discussion of the importance of triangulation in accounting research.