The evolution of internal audit research: a bibliometric analysis of published documents (1926–2016)

ABSTRACT

Addressing the rise of internal auditing in the post-SOX era, this study examines the scientific transformation of the topic within current accounting research. In an attempt to shed light on the existing research themes and core works that have been shaping this topic, we combine co-citation and social network analysis to analyse citation patterns of 170 research articles published in five leading accounting journals between 1926 and 2016. The scientific landscape of internal auditing within accounting research is found to be highly fragmented and partly defined by internal auditors' relationships to other parties of the corporate governance framework. Additionally, results reveal the existence of a research nucleus which emphasises the increasingly important construct of internal audit quality.

KEYWORDS:

• Internal audit
• internal audit function
• bibliometrics
• co-citation analysis

Acknowledgments

We thank David Wood and participants of the 2018 Annual Congress of the European Accounting Association, the 2018 Annual Accounting Conference, and the 2018 Auditing Section Midyear Meeting for many helpful comments and suggestions. We would also like to thank two anonymous reviewers and the editor for their useful comments that improved this manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 Worth mentioning is also a broader stream of studies connected to the survey-based Common Body of Knowledge (CBOK) on the current status and perception of internal auditing on different continents (e.g. Hass, Abdolmohammadi, and Burnaby 2006).

2 Namely Accounting, Organizations and Society (AOS), Contemporary Accounting Research (CAR), Journal of Accounting and Economics (JAE), Journal of Accounting Research (JAR) and The Accounting Review (TAR).

3 Hertzel (1987) provides a thorough overview of the evolution of bibliometrics and the usefulness of co-citation analysis in particular. However, White (2011) mentions possible constraints of citation-based network analysis considering the omission of the amount, quality and therefore impact of scientific output of publishing individuals or affiliated institutions. We address this issue by adding a descriptive dimension to our paper.

4 For a more detailed explanation of the methodological steps, see Appendix 1.

5 Although included in both the ABS Journal Guide and the 2016 Financial Times research ranking, evidence for the importance of the journal Review of Accounting Studies (RAST) within scientific studies is highly inconsistent (e.g. Bonner et al. 2006, 666). Since consistency was a central selection criterion for this study, we decided to omit the journal.

6 This revision dates back to one of the largest corporate accounting scandals of this decade, namely the failure of internal controls of the pharmacy company McKesson & Robbins Incorporation in 1938 (Barr and Galpeer 1987; Bealing, Dirsmith, and Fogarty 1996; Baxter 1999).

7 The gap of three years between the actual endorsement of SOX and the observed increase in scientific publications can be explained by the often time-consuming publishing process, ranging from the stage of initial conceptualisation to the final acceptance of an article.

8 We additionally gathered information on the affiliated institutions for most of the authors and were able to match data from 147 of the 170 articles. Overall, nine organisations and 130 academic institutions could be identified; 297 authors or co-authors were affiliated with them across all publications. Our findings, as summarised in Appendix 2, highlight the 20 most frequently mentioned institutions regarding sole or co-authorships.

9 We refer to qualitative research as either field work, interview, or case study approaches (see Smith 2014).

10 This corresponds to common theories related to citation frequencies of scientific evidence. For an example see Zipf-Pareto's law in Yablonski et al. (1980): ‘It is always possible to single out a small number (the “nucleus”) of specialised journals that carry the main informational load of the field under study, and a vast number of peripheral journals that are sometimes quite remote from this scientific discipline. The same applies also to other disciplinary parameters: scientific contacts, citation networks, etc.’ (Yablonski et al. 1980, 4).

11 D. A. Wood, as an exception, includes internal auditing as a central theme in almost 80 per cent of his cited works.

12 The link strength indicates the number of times that two sources occur jointly in a list of references of the 170 articles. Despite the general lack of theoretical guidance for the optimal link strength, our chosen value eliminates random observations as it reduces our network to a consistent core of 230 documents with an average link strength of 3.4.

13 The Final Rule: Revision of the Commission's Auditor Independence (SEC 2000) serves as the regulatory backbone of this discussion as it is found at the centre of the cluster.

14 A majority of the unrelated studies are located in transitional areas of adjacent clusters and are with few exceptions limited in their impact on the network.

15 One reason for the omission of internal auditing is the data availability of internal audit information. For example, Dhaliwal, Naiker, and Navissi (2010) note: ‘internal auditors can also play a role in preserving financial reporting quality by monitoring organisational risks and assessing internal controls. [···] We do not control for internal auditing because of the significant time and cost constraints involved in identifying the presence of an internal auditing function for our large sample’ (Dhaliwal, Naiker, and Navissi 2010, 803). Most of the other archival studies of the cluster follow a similar rationale (e.g. McDaniel, Martin, and Maines 2002; Srinivasan 2005; Krishnan and Visvanathan 2008).

16 Included studies on audit committee effectiveness not cited in the cluster description are DeZoort (1998) and Cohen, Krishnamoorthy, and Wright (2008).

17 A material weakness describes a significant deficiency or combination of significant deficiencies, that results in more than a remote likelihood that a material misstatement of the annual or interim financial statements will not be prevented or detected (PCAOB 2004).

18 We provide a comprehensive list of research questions in Appendix 4.