مدل احتمالی درک اثربخشی در سیستم های اطلاعاتی حسابداری:اثرات هماهنگی و کنترل سازمانی

A contingency model is advanced that examines sources of requirements for organizational coordination and control as they affect the extent of integration in an accounting information system. Requirements that are contingent on the degree of organizational formalization, information interdependence among functional areas, and dependence in interorganizational information sharing and electronic data interchange links, are examined. The congruence or fit of system integration with those requirements is a key concept that influences beliefs about system effectiveness. Results of the empirical study indicated that, as hypothesized, the fit between the accounting system design and the contingency factors resulted in a more successful system. Specifically, system fit was a significant factor that explained variations in perceived AIS effectiveness, as measured by decision makers' perceived satisfaction with the accuracy and monitoring effectiveness of output information. The effect of system fit on a second factor of perceived AIS effectiveness, as measured by decision-makers' satisfaction with the perceived quality of information content in system outputs, was only marginally significant. The study addresses an important area in accounting systems research that directly relates to the decision facilitation and control objectives of accounting information.

A critical research issue in the fields of accounting and management decision-making concerns the fit of the accounting information system (AIS) with the organizational requirements for information communication and control. An AIS is defined here as a computer-based system that processes financial information and supports decision tasks in the context of coordination and control of organizational activities. Prior accounting research has examined different models of fit between an AIS and an organization's task technology, structure, and environment Chenhall and Morris 1986, Gordon and Miller 1976, Gordon and Narayanan 1984, Kim 1988, Macintosh and Daft 1987 and Mia and Chenhall 1994. Although these earlier models have provided useful directions for AIS research, they have not examined specific system design constructs in relation to system effectiveness. The present study adds to this body of literature by developing a specific system design construct, “AIS Integration,” and by examining its functional relationship with perceived system effectiveness. AIS Integration, in turn, is hypothesized to be a function of a number of contingency constraints that create organizational coordination and control requirements. Sources of contingency constraints on AIS design have been identified through a review of prior accounting studies. In addition, this study introduces constraints due to interorganizational interactions that have not been examined before in the accounting systems literature. The approach in this study is consistent with the information-processing paradigm of organizational design Daft and Lengel 1986, Galbraith 1973 and Tushman and Nadler 1978, which would suggest that AIS design represents a response to the requirements for organizational coordination and control (OCC). AIS integration refers to a particular design state where the system in its implemented form can provide output information that may be effectively used to address OCC problems and requirements. Contingent variables, such as (a) the degree of formalization in the structure of an organization Hage and Aiken 1969 and Simons 1987, (b) interdependencies in information requirements between functional areas within an organization Govindarajan and Fisher 1990 and Thompson 1967, and (c) dependencies due to interorganizational information sharing and electronic data interchange (EDI) links Bakos 1991, Srinivasan et al. 1994 and Zaheer and Venkatraman 1994, influence the extent to which organizations experience different levels of coordination and control problems. AIS integration can resolve difficulties in coordination and control that are created by these contingent variables. The purpose of this study is to empirically examine the relationship between AIS integration and perceptions of system effectiveness. Specifically, it is hypothesized that to the extent that AIS design provides for system integration, as necessitated by the three contingent variables mentioned above, the system would be perceived as effective. This hypothesis is tested with data collected from firms in the United States using the survey research method. Results partially confirm the hypothesis that the degree of fit between AIS integration and the contingent variables predicts AIS effectiveness. There is stronger support for the hypothesis when AIS effectiveness is defined by decision-makers' satisfaction with the accuracy and monitoring effectiveness of output information than by the more traditional definition of satisfaction with quality of information content in system outputs. The remainder of the article is organized as follows: In the next section, research that relates to individual components of the research model is reviewed. The theoretical framework is developed and the research hypothesis for the study is advanced. The research method for the study is presented next, followed by a presentation of the empirical findings. The article concludes with a discussion of the findings and with suggestions for future research.

This study examined the relationship between the degree of fit between organizational requirements for coordination and control with the design of an AIS and perceptions of effectiveness about the system. Contingency theory served as the basis for the development of the hypothesis in the study. The results of the study indicated that internal dependence due to (a) required information sharing across organizational functions, (b) organizational formalization, and (c) interorganizational dependence in terms of both resource importance and accessibility, had a significant effect upon the requirements for organizational coordination and control that should be met by the design of the system. The fit between AIS design and those requirements significantly contributed to perceptions of monitoring effectiveness and to perceptions about the accuracy of information outputs. System fit, however, failed to exhibit a strong effect on user information satisfaction, that is, on the perceived quality of information content available in system outputs. The overall results are consistent with the theoretical perspectives underlying the concepts examined in the study. The theory of resource dependence (Pfeffer and Salancik, 1978) explains the behavior of organizations in cases where one organization is dependent upon another for critical resources. As interorganizational EDI links become more prevalent and organizations depend on them for their survival, the requirements to coordinate and control internal activities become more intense if continuous flow of resources is to be ensured. The more extensive the use of EDI and the more dependent organizations are on firm-specific communication formats, the greater the dependence and the need for internal coordination. The fit of the system with those requirements has been shown to contribute to satisfaction with the accuracy of system-generated information, as well as with satisfaction with monitoring effectiveness. Decisions and actions giving rise to costs at the different stages of the ordering, delivery, receiving, inspection, storing, and payment processes can be more effectively controlled through the use of integrated systems. Such benefits could reduce the total cost of ownership of raw materials and provide strategic advantages to the organization (Cooper and Kaplan, 1991). The results of the study are also consistent with information processing theory Daft and Lengel 1986, Galbraith 1973 and Tushman and Nadler 1978, which posits that, in the face of uncertainty, organizations will incorporate information-processing capacities in their structures in order to match information processing requirements. Interdependent processes among functional areas can create requirements for information sharing and the design of integrated AISs that will satisfy such requirements. Adherence to rules and procedures may lead to an increased need for ongoing evaluations and these results have shown that, in such cases, AIS design must be responsive to those requirements in order for the system to be perceived as effective. The design of integrated systems, therefore, as a response to such requirements, may satisfy the need for improved accuracy in information sharing and continuous monitoring, and thus result in more effective systems. The lack of significant results between system fit and the concept of “user information satisfaction” might be due to the choice of the research instrument used to measure UIS. Doll and Torkzateh's (1988) instrument was chosen because of evidence of its superior psychometric validity in the specific context of accounting information systems (Seddon and Yip, 1992). More recent conceptualizations of the system effectiveness construct advance a concept of “task/technology fit,” that is, information systems will have a positive impact on performance only when there is a correspondence between their functionality and the task requirements of users (Goodhue, 1995). The Doll and Torkzateh (1988) instrument might have provided too limited a measurement for an overall evaluation of system functionality. It has provided user evaluations for a wide range of information characteristics, but a different focus might have been necessary. For example, a measure of system functionality might focus on the level of support that is provided to users in carrying out their managerial functions of planning, investigating, coordinating, and evaluating, as they have been specified in the traditional management literature (Mahoney et al., 1965) or in any management accounting text (e.g., Hilton, 1997). Like all studies, the current study also has its limitations due to the methodology employed. For example, the study measured all research variables at a single point in time and used co relational analysis. This approach limits statements about causation. Use of summated responses to questionnaire items that appear on the same instrument always entails some risks. Responses could be biased because of the common method used for the collection of all data. Though care was taken to extensively validate these data through psychometric analyses, which have not indicated any violations of scale validity, this criticism of the survey method can never be completely ignored and should be taken into account. Nonresponse has been a problem in the data collection phase of this study. Nevertheless, all tests for nonresponse bias have provided satisfactory results. Despite the above limitations, this study has contributed useful results for both professionals and researchers in AIS. Professionals can utilize the scales for AIS integration and AIS effectiveness that were examined in this study in order to measure and evaluate the level of satisfaction with an existing system. This evaluation could also be performed within the context of a post-implementation review, where the objective might be to identify areas of concern that should be addressed in future system development projects. The importance of using standardized scales in a post-implementation evaluation context is that they promote learning about the system. This effect is often difficult to measure, even though it represents a major purpose for post-implementation reviews Chenhall and Morris 1993 and Miller and Dunn 1997. For researchers, the concepts developed here, particularly the concept of system integration and its fit with contingent variables, offer empirically observable indicators that could be examined in a post-implementation context. The relationship between the evaluation of post-implementation perceptions of effectiveness and firm performance has not been examined in past empirical research, despite its importance to professionals in the field. Although this study has also not investigated this important relationship, it provides some insights into this complex question. Any future research effort addressing this relationship must consider the issue of using AIS evaluation scales that were validated in different contexts in order to isolate the effect of system learning on a firm's operating performance. The complexity of this issue is manifested by the fact that firm performance could not only be defined in different ways, but it could also be influenced by other factors confounding the effects of interest. These concepts could also be incorporated in research models that examine the effect of contingencies relating to internal cost structure and interorganizational cooperation through just-in-time exchanges on the implementation of cost management systems. The results of this study have demonstrated the feasibility as well as the necessity of examining multiple contingencies in future research models of AIS design. There are also a number of other useful directions for future research. First, the study has relied on qualitative or perceptual measures of AIS effectiveness. Future research could examine the effects of accounting system design choices on quantifiable measures of firm performance, such as operational measures of performance (e.g., inventory turns) as well as profitability measures (e.g., return on assets employed). Second, the measurement of interorganizational variables needs to be refined and further developed. Finally, future research could also further refine the concept of organizational coordination and control and directly define and measure different types of costs that relate to such requirements. Consequently, the relationship between system fit and AIS effectiveness could be made more explicit and could incorporate direct types of costs that could be reduced, eliminated, or displaced through the use of appropriate accounting information systems. In conclusion, the central issue of system integration that was examined in this study has been demonstrated to be an important concern, both empirically as well as conceptually. The respondents to this study have indicated their reliance on the measure of system integration in forming their opinions about the effectiveness of an AIS. This study is one of a very few studies to empirically examine the concept of AIS integration and its relationship with perceptions of AIS effectiveness. At the conceptual level, AIS integration successfully captures the effect of specific contextual influences on AIS design. The concept of AIS integration as a specific state of system design, therefore, is validly related to an organization's specific context, as this may be determined by interorganizational agreements, internal teamwork, or cross-functional decision-making units. This study only represents the start of a series of research investigations into these important issues that are faced by current businesses.Committee of Sponsoring Organizations of the Treadway Commission 1992, Company Profiles 1995, Davis Davis Moore 80 and Deloitte and Touche Hyperion Software 1995