کاهش تضاد در ارزیابی های کارت امتیازی متوازن

Recent studies [Ittner, C., & Larcker, D. (2003). Coming up short on nonfinancial performance measurement. Harvard Business Review(November) 88–95; Ittner, C., Larcker, D., & Randall, T. (2003b). Performance implications of strategic performance measurement in financial services firms. Accounting, Organizations and Society, 28, 715–741] provide evidence of companies’ tendency to overlook the validity of the causal links between driver and outcome measures of the balanced scorecard (BSC), and to ignore the underlying strategically-linked causal business models. It is posited that this propensity leads to conflict between top management and divisional managers because of the failure of the former to evaluate and consider strategy effectiveness in performance evaluation. The present study hypothesizes that individuals in the top-manager role do not take into account strategy effectiveness unless they are explicitly required to do so. In contrast, individuals in the store-manager role automatically consider the quality of strategy without being prompted to do so. A study using 63 evening MBA students provides support for the hypotheses. The results have implications for the study of evaluation biases in BSC as well as in other performance measurement systems, and for devising means to mitigate them.

Performance evaluation is an essential function in any organization. Consequently, it is important to understand how performance measurement systems influence such appraisals. In a seminal study on the role of accounting data in performance evaluation, Hopwood (1972) highlighted problems with traditional accounting measures of performance. In particular, he noted the lack of comprehensiveness of the measures, the imprecision with which accounting systems measure performance, the limited focus on outcome measures, and the over-emphasis on short-term performance (pp. 157–158). Hopwood (1972) hypothesized and found that depending on the evaluation style, reliance on these accounting performance measures can result in dysfunctional consequences including disagreement and conflict between supervisors (raters) and subordinates (ratees). Subsequent studies (e.g., Hartmann, 2000 and Otley and Pollanen, 2000) have similarly provided evidence suggesting that, depending on various contextual factors, negative consequences (e.g., job-related tension and distrust in supervisor) may result from the reliance on conventional accounting measures when evaluating performance. In contrast to traditional performance measurement systems, the balanced scorecard (BSC) introduced by Kaplan and Norton (1992) is expected to address many of the concerns raised by Hopwood (1972). The BSC is a multi-dimensional performance measurement system that includes financial, outcome and short-term as well as non-financial, driver and long-term measures. A key feature of the BSC is its emphasis on linking the performance measures with business unit strategy (Otley, 1999, pp. 374–375). Because of its comprehensive and strategy-linked measures, the BSC can be expected to reduce the likelihood of previously observed disagreement and conflict between raters and ratees by promoting a more holistic approach to performance evaluation. For example, when evaluating poor performance, the inclusion of strategy-linked outcome and driver measures in the BSC may direct raters to attend to strategy quality as an explanation, which they are unlikely to consider when using a system strictly based on outcome measures. Consequently, raters using the BSC may ascribe inferior performance less to ratees and more to strategy-related causes. This would presumably lead to more favorable performance appraisal of ratees and reduce the likelihood of disagreement and conflict between raters and ratees. Recent studies (Ittner and Larcker, 2003 and Ittner et al., 2003b), however, suggest that the causal links between driver and outcome measures are often overlooked. For example, Ittner et al.’s (2003b, p. 725) study of financial services firms found that of those claiming to use a balanced scorecard, 76.9% place little or no reliance on their strategically linked causal business models. In another field study of manufacturing and service companies, Ittner and Larcker (2003, p. 90) note more specifically that among those that create causal models, only 21% go on to validate the causal links between driver and outcome measures. They also observe “businesses often fail to establish such links partly out of laziness or thoughtlessness” (p. 89). Thus, the evidence from these studies indicates that although information about strategy effectiveness is available in the BSC, it is not used as much as would have been expected presumably because of cognitive limitations. The present study posits that the above-noted tendency to overlook the validity of the causal links between driver and outcome measures of the BSC is a potential source of conflict between top management and divisional managers. Specifically, it is proposed that discrepancies in performance ratings between raters and ratees may result from top management’s failure to consider the quality of its chosen strategy when evaluating divisional managers’ performance. Based on research (Feldman, 1994, Ilgen et al., 1993 and Landy and Farr, 1980) that views performance appraisal primarily from a cognitive process perspective, the present study further posits that top management’s failure to consider strategy effectiveness is due to its selective attention bias. Thus, it is hypothesized that increasing top management’s awareness of the impact of strategy effectiveness on performance, may reduce conflict resulting from differences in evaluation between top management and divisional managers. The purpose of this research is to test the foregoing propositions. To establish the existence of a bias, the present research first assesses the extent to which, when using the BSC, divisional performance evaluation differs between individuals who adopt the perspective of top management (the rater) and those who assume the role of divisional managers (the ratees). The study then examines the effectiveness of a mechanism for reducing top management’s bias, and thus, the disagreement in performance appraisal between the two perspectives. The specific procedure involves increasing the rater’s awareness of the quality of top management’s strategy in influencing divisional performance. This is achieved by requiring an explicit assessment of the role of strategy quality in determining divisional performance prior to performance evaluation. Examining potential conflict and disagreement that may result from top management’s bias in performance evaluation within the context of the BSC is significant for at least three reasons. First, the study is important given that, as noted earlier, features specific to the BSC are expected to lessen the likelihood of bias and thus, conflict and disagreement in performance evaluation. However, contrary to expectations, recent field studies (e.g., Ittner et al., 2003a and Malina and Selto, 2001) document disagreement and conflict between top management and divisional managers when using the BSC to evaluate the latter’s performance. In order to obtain the intended benefits of the BSC, it is important to understand the nature of biases that may result in conflict, and investigate ways in which the biases can be mitigated. Second, the present study in the specific context of the BSC is important because top management’s biases may also influence the effectiveness of the BSC as a strategic management system. Kaplan and Norton, 1996 and Kaplan and Norton, 2001 specifically note how the BSC provides companies with the capacity for strategic learning, enabling them to modify strategies when necessary. In evaluating performance, the exclusive focus on divisional managers’ ability and effort to execute the company’s strategy, and the failure to consider the effectiveness of the strategy, may impede strategic learning. Specifically, because of the failure to see potential problems with the strategy, it may not be adjusted when it is beneficial and justified to do so. Third, Keeping and Levy (2000) and Cardy and Dobbins (1994) argue that one criterion to use in evaluating performance appraisal systems is the reaction of ratees (i.e., divisional managers, in the present context). Although the present study is conducted in the context of the BSC, the results may have implications for reducing conflict and tension between raters and ratees in any performance measurement system. The remainder of this paper is organized as follows. The next section provides the theoretical background, which leads to the development of the hypotheses. The research method is then described followed by the results, and a discussion of the implications as well as the limitations of the study.

Following research (e.g., Lipe and Salterio, 2000 and Lipe and Salterio, 2002) that examines biases in BSC evaluations, the present study explored the role of rater’s selective attention in contributing to conflict between raters and ratees. Specifically, based on research on biases and performance appraisal, the present study investigated (1) the extent to which performance evaluation using the BSC diverges between individuals who assume the role of top management (rater) and those who adopt the role of a store manager (ratee), and (2) the degree to which the difference in BSC performance evaluation is reduced as a result of requiring participants to assess the significance of strategy effectiveness in influencing performance, prior to rating divisional performance. As hypothesized, individuals who assumed the role of top management rated divisional performance lower than those who assumed the role of store managers. This is consistent with the evidence of disagreement in BSC evaluations between top management and divisional managers, gathered from field studies (Ittner et al., 2003a and Malina and Selto, 2001). While this finding is not unlike those observed in other performance evaluation contexts (Atwater et al., 1998, Atwater and Yammarino, 1997, Harris and Schaubroeck, 1988 and Viswesvaran et al., 1996), it is particularly important in the present situation since the comprehensive and strategically linked measures of the BSC are expected to lead to lower rater bias than traditional performance measurement systems that include only financial and outcome measures. Thus, the finding suggests that contrary to expectations, the unique features of the BSC do not significantly reduce selective attention bias that presumably leads to conflict between top management and divisional managers. The predicted effect of increasing top management’s awareness of the importance of strategy effectiveness was also obtained. Specifically, as postulated, individuals who were assigned to the role of top management rated store managers significantly better when they were required to assess the role of strategy quality in determining divisional performance than when they were not. Apparently, without explicit prompting, participants assigned to the role of top management do not automatically consider the strategy. In contrast, among participants who assumed the role of store managers, there was no significant difference in performance ratings between those prompted to assess the role of strategy effectiveness and those who were not. It appears that participants in the store manager role consider the strategy even when they are not explicitly required to do so. As a result of the required assessment, the difference in performance evaluation between participants in the top management and the store manager roles was no longer significant. This pattern of results is consistent with the present study’s conjecture that the observed performance evaluation bias may be due to top management’s failure to attend to the quality of the strategy when appraising divisional managers’ performance. Consistent with the findings of Banker et al. (2004), the Store main effect observed in the current study implies that participants were able to distinguish between strategically linked and non-linked measures, and assigned more weight to the former. Since performance bias is observed in the no assessment condition when participants are able to differentiate between strategically linked and non-linked measures, this evidence suggests that participants’ attention to strategically linked measures is not sufficient to reduce selective attention bias that presumably leads to conflict between top management and divisional managers.