Date of Award

2015

Document Type

Thesis

Publisher

Edith Cowan University

Degree Name

Doctor of Philosophy

School

School of Business

Faculty

Faculty of Business and Law

First Supervisor

Associate Professor Hadrian Djajadikerta

Second Supervisor

Professor Craig Standing

Abstract

Insider deviant behaviour in Accounting Information Systems (AIS) has long been recognised as a threat to organisational AIS assets. The literature abounds with a plethora of perspectives in attempts to better understand the phenomenon, however, practitioners and researchers have traditionally focussed on technical approaches, which, although they form part of the solution, are insufficient to address the problem holistically. Managing insider threats requires an understanding of the interconnectedness between the human and contextual factors in which individuals operate, since technical methodologies in isolation have the potential to increase rather than reduce insider threats. This dilemma led many scholars to examine the behaviour of individuals, to further their understanding of the issues and in turn, control insider threats. Despite promising findings, some of these behavioural studies have inherent methodological limitations, and no attempt has been made to differentiate between apparently similar, yet fundamentally different, negative behaviours.

Using the theory of planned behaviour (TPB) and actor network theory (ANT) as a foundation, the current study addresses the first concern by integrating AIS complexity and organisational culture, and identifies the contextual factors influencing behaviours that lead to insider threats. Secondly, the study addresses concerns regarding methodological approaches, by categorising various deviant insider behaviours using the concept of dysfunctional behaviour, based on two-dimensional behaviour taxonomy.

Partial least square structural equation modelling (PLS-SEM) revealed that TPB‘s predictor variables: attitude (ATT), subjective norm (SN) and perceived behavioural control (PBC), together with the moderator variables of organisational culture (CULTURE) and AIS complexity (COMPLEX), accounted for substantial variations in intention (INTENT) to engage in dysfunctional behaviour. The findings also indicated that PBC is a dual-factor construct. Changes in predictors at the behavioural subset level were highlighted, and the findings of previous studies, that ATT is a salient predictor of intention, were confirmed. This was significant across all four dysfunctional behaviour categories.

These findings add to the body of knowledge by contributing a theory that explains insider threats in AIS by deciphering dysfunctional behaviour using a predictive model. The study also provides a methodological foundation for future research to account for behavioural factors. Moreover, the findings have implications for managerial practices who want to reduce insider threats to an acceptable level by strengthening organisational culture, moderating AIS complexity, and focussing on management programs with sufficient momentum to impact attitudinal change.

Included in

Accounting Commons

Share

 
COinS