مدیران،برای کدام اهداف از کارت امتیازی متوازن استفاده می کنند؟ یک مطالعه تجربی

The accounting literature frequently publishes articles that establish the adoption rates of accounting information systems, such as the Balanced Scorecard (BSC) or Activity-Based Costing, and subsequently examines the factors that drive this adoption. However, much less is known about the specific purposes for which these systems are used. In this paper, I examine the purposes for which managers use the Balanced Scorecard. Data was collected from a survey administered in 19 Dutch firms which had indicated that they used a BSC. The survey resulted in 224 responses from individual managers. Using exploratory factor analysis on Doll and Torkzadeh's [Doll, W.J., Torkzadeh, G., 1998. Developing a multidimensional measure of system-use in an organizational context. Information and Management 33, 171–185.] instrument of multidimensional MIS usage, I find that managers use the BSC for: (1) decision-making and decision-rationalizing; (2) coordination; and (3) self-monitoring. In the second step, I consider drivers of BSC usage for the three different purposes. These drivers are dimensions of evaluation style, alternative controls that are used in the organizational unit, and the receptiveness of managers to new types of information. I find that BSC usage for decision-making and decision-rationalizing purposes is driven by the degree of action controls used and manager's receptiveness to new information types. BSC usage for coordination purposes is driven by the emphasis placed on managerial evaluation of subordinates and the manager's receptiveness to new types of information. Finally, BSC usage for self-monitoring purposes is driven by the emphasis placed on managerial evaluation.

A number of innovative management accounting techniques have been developed over the past two decades, the most widely known of which are Activity-Based Costing and the Balanced Scorecard (BSC). Prior literature evaluates the success of these techniques by estimating its adoption rates and testing relationships between use and different types of outcome measures. Outcome measures used are, amongst others, use of the system, impact on decisions made based on information from these systems, dollar improvements following implementation of the system, and management evaluation of system success (Foster and Swenson, 1997). This study not only assesses the level of use, but also focuses on the purposes for which managers use the BSC. Many factors influence individual BSC usage. First, usage is influenced by the way the organization intends to use the BSC, and therefore by the design of the scorecard. Second, individual usage is influenced by the opinions of top management, supervisors, and other colleagues of BSC users, and by other elements of the control system available in the firm. Together, these factors lead to a varying degree of pressure on individual BSC users to use the scorecard, i.e. the subjective norm. As Hartwick and Barki (1994) argued, however, when system usage is mandatory, i.e. the subjective norm to use the system is strong, the intensiveness of use might still vary. Some managers will use the system all the time, whereas others will use it selectively, e.g. when they consider using it to be effective. This implies that even if BSC usage were mandatory in a firm, both the intensiveness and purposes of use might vary among managers. In this paper, I explore this individual BSC usage in the context of organizational usage (firm effects). I address a number of avenues for further research raised by Malmi (2001). First, he suggests that future research should study whether we can explain various types of BSC usage by looking at firm characteristics. Second, he suggests that we know little about the impact of using the BSC on other control mechanisms in the firm. I provide additional insights into these questions for BSC usage at the individual manager level. First, I explore the purposes for which managers use the BSC. For this analysis, I adopt an instrument developed by Doll and Torkzadeh (1998) which measures the multidimensional usage of a management information system (MIS). The instrument mainly captures BSC usage for decision-making and is less well developed for measuring usage for control purposes and strategy communication.1 Provided that individual usage is examined in this study, and that the communication of strategy and control of employees are primarily firm purposes, these limits do not seem to be too severe. Second, I explore the drivers of these different purposes of BSC usage. These drivers are the evaluation style of managers, other control mechanisms used in the organizational units, and the receptiveness of managers to information from modern performance measurement systems. Use of performance measurement systems is influenced by how managers evaluate their subordinates (Otley and Fakiolas, 2000). The BSC consists of a number of different types of measures that make a distinction between financial and non-financial measures, objective versus subjective measures, leading indicators versus lagging indicators, and measures that are easily quantifiable as opposed to measures that are more difficult to quantify. Managers can therefore evaluate performance on many different dimensions. The way managers evaluate their subordinates’ performance is therefore expected to influence the intensiveness of BSC usage and type of BSC usage. Based on an exploratory factor analysis, I identify three different dimensions of evaluation style. These dimensions include whether managers have a rigid or a flexible evaluation style to evaluate their subordinates; whether they value financial or non-financial performance measures more strongly; and whether they value quantitative or qualitative performance measures more strongly. Further, usage of the BSC takes place amid the usage of other control mechanisms, because firms often use a package of controls (Otley, 1999). The design of this control package is also expected to influence intensiveness and type of BSC usage. Finally, the adoption diffusion literature suggests that some individuals are more innovative and more eager to experiment with new systems than others (Rogers and Shoemaker, 1971). Therefore, greater receptiveness on the part of managers to use new types of information sources is expected to result in higher BSC usage. The different dimensions of BSC usage and evaluation style are both identified from an exploratory factor analysis. Therefore, the nature of this study is exploratory. The survey was conducted in 19 firms identified as BSC users. The sample consists of 224 managers responsible for the BSC of a department or business unit in these firms. From an exploratory factor analysis, I find three different purposes of BSC usage: (1) a decision-making and decision-rationalizing dimension; (2) a coordination dimension; and (3) a self-monitoring dimension. The three resulting dimensions have a moderately high correlation with each other, suggesting that these purposes of use are complementary. BSC usage for decision-making and decision-rationalizing is positively related to the number of action controls used and manager's receptiveness to new types of information. BSC usage for coordination purposes is also positively related with manager's receptiveness to new types of information and positively related with the emphasis placed on managerial evaluations of subordinates. Finally, BSC usage for self-monitoring is positively related with the emphasis placed on managerial evaluations. I make three contributions to the literature. First, instead of simply measuring the degree of BSC usage, I explore whether managers use BSCs for different purposes. Thus, I focus on BSC usage at the individual management level. Second, in line with the comments on the narrow and outdated measurement of the concept of evaluation style (Otley and Fakiolas, 2000), I identify additional dimensions of this concept. Otley and Fakiolas (2000) argue that the original ‘reliance on accounting performance measures’ (RAPM) instrument was built in an era when budgets were the only control mechanism used in firms. In the context of more contemporaneous control mechanisms, such as the BSC, additional dimensions of managerial evaluation style can be expected. Finally, I explore whether these different dimensions of evaluation style, other controls used and the receptiveness of managers to new information types are associated with BSC usage for different purposes. The remainder of the paper is structured as follows. In the next section, I describe purposes of BSC usage, Doll and Torkzadeh's (1998) instrument that measures multiple dimensional usage of management information systems, and potential drivers of these different purposes. In Section 3, I outline the sample and the procedures used to administer the survey. In Section 4, I provide the empirical results of the study. Following the discussion of the results in Section 5, I conclude the paper with a summary in Section 6.

The results from the empirical analysis suggest that firm effects, organizational unit effects and individual manager effects influence BSC usage. First, the regression results indicate that some firms have a higher average score than other firms. Together with the lower standard deviations of the two firms that have the highest average BSC usage, this indicates that these firms probably stimulate the use of the system, for example, they make BSC use mandatory or introduce managers to its beneficial effects. Although the study suggests that firm effects play a role, it does not provide an understanding of which firm characteristics these are. The literature provides a number of suggestions. First, Ittner et al. (2003) showed that users and non-users of the scorecard did not differ in the type of strategy (i.e. either a maintain, innovate or flexible strategy) they implemented, which seems to be consistent with Kaplan and Norton's argument that all firms, irrespective of their strategy, need a BSC. Second, the firm structure, for example, degree of decentralization and interdependency, might influence the extensiveness and purposes of BSC usage. For example, Abernethy and Bouwens (2005) argue that decentralization might mitigate the resistance of managers to using new accounting information systems, such as the BSC. The reasons provided for this conjecture are that decentralization increases the ability of managers to adapt the organizational unit to these new information systems, and second that decentralization leads to higher involvement of the unit manager in the design of accounting systems and therefore leads to more meaningful information. Third, the design of the BSC and subsequently the quality of the scorecard might influence its usage. For example, a number of studies suggest that not all firms that implemented a BSC also used cause and effect logic (Bedford et al., 2006, Speckbacher et al., 2003 and Ittner et al., 2003), and this might have an impact on the intensiveness and type of usage. Finally, the organizational adoption literature suggests that factors, such as the existence of a project “champion”, the size of the firm, and the usage of consultants all influence the decision to adopt a BSC or not. Further research is needed to explore whether these factors also influence the intensiveness and type of use. As argued before, individual use of BSC is embedded in firm use. Hartwick and Barki (1994) suggest that individual use of information systems differs among managers even when firms adopt information systems and managers are pressed to use the system. The results support this proposition. Although the firm dummies explain a considerable degree of the intensity and purposes of use, there is still a considerable amount of variation in BSC usage, suggesting that managers have considerable discretion as to whether and how they use the system. If managers use the system for one purpose they are often also using it for more purposes, indicated by the strong correlations between the three purposes. Although the usage instrument of Doll and Torkzadeh is not in any way based on the purposes that Chenhall (2003) mentioned for which managers use integrated performance measurement systems, there is considerable overlap between the two classifications. Both stress that the scorecard is used to facilitate decision-making, to communicate information between managers and to provide feedback on the decisions managers make. The empirical results would seem to show that use of the scorecard is higher for individual reasons, i.e. to help make decisions and to receive feedback on these decisions, rather than to communicate with others. Further, the results suggest that the type of controls used in the organizational unit is associated with BSC usage. More specifically, when the manager's unit uses more action controls, the manager has a higher BSC usage for DMR. One potential explanation for the result is that managers who are held action accountable, i.e. work under many action controls, need to document more often why they took the decisions they did. As an example, Merchant (1998) argues that a superior can track employee actions by gathering evidence of actions taken. In this process, managers make more extensive use of BSC information to show what the results of their actions are. In this study, the impact of other accounting and control systems on BSC usage is measured through very broad categories, such as, action controls and personal and cultural controls. However, prior literature indicates that the existence of specific accounting information systems, such as the use of budgets, might influence the intensiveness and type of BSC use. For example, Otley (1999) and Malmi (2001) suggested that using a BSC reduces the extensiveness of budget use. In addition, Malmi (2001), Bedford et al. (2006), and Speckbacher et al. (2003) found that some firms link the BSC to their reward system, whereas others do not, which might also influence intensiveness and type of BSC usage. From the three dimensions of evaluation style, the dimension rigid versus flexible use of evaluations is negatively associated with BSC usage for both COOR and SELFMON, but is not related to DMR. One reason for these results is that usage for DMR is related to the primary tasks of the managers, i.e. whether he makes the right decisions. Usage for COOR and SELFMON however is related to managerial functions of the managers and therefore directly assesses the performance of the manager and his organizational unit. The NFF and QUALQUAN dimensions are not related to BSC usage, suggesting that personal preferences for some types of performance measures have no impact on BSC usage. Finally, the receptiveness of managers to new information is positively associated with BSC usage for both DMR and COOR. This result extends the evidence of Baird et al. (2004), who found that organizational innovativeness explained variation in the adoption of activity management practices of firms, to the individual level. In addition, it suggests that innovativeness not only influences the adoption of new management accounting techniques, but also the use of these techniques. An implication from these results is that whether managers use the BSC is not only a rational choice, based on the value of the system, but also simply a matter of managers’ preferences. This observation is reinforced by the high correlation of the BSC usage for the three different purposes. Thus, when managers use the BSC extensively for one purpose they also use it for the other two purposes. This would imply that when firms think using the BSC is beneficial in their circumstances, that the preferences of managers should be influenced, for example through training, or the firms should select managers who advocate using modern information systems to gather information.