A Heuristic Method for Composing a Literature Review

Abstract

The process of doing and documenting a literature review is usually quite idiosyncratic. Luft and Shields (2003) presented a compelling process to map theory-consistent empirical research in management accounting. However, their mapping process depends on a reliable selection of works. Although Luft and Shields (2003) present some criteria of selection of empirical works, we propose a formal heuristic method for the process of literature reviewing that integrates their mapping process. The heuristic is based on the concept of layers, and it is aimed at mitigating inclusion errors and exclusion errors which might occur when selecting papers for a literature review. These errors impact internal validity and construct validity, since a literature review defines operationalization of variables, choices of proxies and selection of methods. By mitigating errors, we contend that threats to internal validity and construct validity that can happen in a study's literature review phase are controlled. The process proposed here consists of a cycle with three phases: building a bibliographic database; mapping extant works and writing up the literature review. By using the heuristic process, a researcher can clearly demonstrate that he or she has made the best efforts to mitigate errors that could reduce a study's validity. Thus, the literature review of an empirical work could be evaluated in addition to the empirical section. Routine use of literature review protocols with open criteria for inclusion or exclusion of works, along with the use of processes of mapping extant research, will contribute to the celerity of consolidation of theory and methods in management accounting.