Towards an effective design of management control systems : a contingency approach

摘要

In an attempt to provide a better understanding of MCS design, this study has utilised the contingency theory approach to investigate the contingent relationships between perceived environmental uncertainty (PEU), business strategy, organisational structure and various attributes of MCS simultaneously. This study has also investigated the mediating effect of two structural dimensions (centralisation and formalisation) on the relationship between PEU, business strategy and MCS design. In addition, this study has investigated the implications of fit, internal consistency or coalignment between the contextual variables and MCS attributes on organisational effectiveness, thus adopting a systems approach to fit recommended in the contingency literature. This study also builds on the works of Simons (1987) and Chenhall and Morris (1986) in terms of developing a wider and comprehensive view of MCS design. Seven MCS attributes have been investigated in this study including budgetary control system practices, budgetary control system usage, cost control systems, broad scope information, managerial evaluation and rewards system, aggregation and timeliness of information. A cross-sectional survey employing a questionnaire method of data collection was adopted. A total of 274 usable responses were received representing a response rate of 28%. For purposes of analysis, the research utilised structural equation modeling (SEM) multivariate statistical technique enabled by EQS 5.7 version software (Bentler, 1995). Thus, this study is one of the first studies in MC contingency literature to utilise SEM for validating the research constructs, controlling measurement error and for testing the structural relationships between the constructs simultaneously. The results of this study confirmed the multi-dimensional nature of PEU, business strategy and organisational structure and the differential effects each dimension has on MCS design. Also the results indicated that the different approaches to fit (i. e., bivariate and systems approaches) utilized in this research resulted in insightful findings relating the contingent relationships between the contextual variables and MCS attributes. In addition, the structural dimension of formalisation was found to have significant positive relationships with many of the MCS attributes and also mediated the relationship between differentiation strategy and MCS attributes of budgetary importance, cost control systems and non-financial performance measures. The results of this study also raise several implications for future MC contingency researchers and fill some gaps in the existing MC contingency literature. This study also contributes to the body of knowledge by providing guidance for future MC contingency researchers to implement the SEM method. SEM has a greater potential for testing theories, controlling measurement error, validating research constructs and testing structural relationships among multiple contextual variables and multiple MCS attributes simultaneously.