Contrasting Views of Results-Based Management Tools from Different Organizational Levels

ABSTRACT

Do top managers view the implementation of results-based management tools differently than do lower-level workers?A total of 4,186 workers in eight state departments of revenue were surveyed about the deployment and the impacts of nine results-based management tools in their agency. The top hierarchical levels were consistently, and often dramatically, more optimistic than lower-levels. The largest differences were generally found in the assessments of worker participation, and in the assessments of the customer relations tools and cross-functional coordination tools. Follow-up interviews were used to suggest reasons for the consistent perceptual differences. Six overlapping causes were suggested by agency members: top isolation, bottom spin, different perspective ranges, inadequate communication, top selective perception, and lower level cynicism.

These findings suggest that management reformers must be prepared to bridge very large perceptual differences between levels, and a number of possible bridging approaches are discussed. The findings also suggest that survey-based public management research will often be misleading if based on surveys of just top managers.

Notes

*Chi-square = 105.600, df = 12, p < .000.

**Chi-square = 265.574, df = 12, p < .000.

***Chi-square = 124.226, df = 12, p < .000.

When one of the sampled revenue departments did have unusual duties, the employees involved in the atypical mission were excluded from this survey.

The preliminary survey was conducted by the first author in October 2002, with a follow-up in April 2003.

In measuring the three impacts, the survey asked respondents to agree or disagree about recent changes. (For example, “Over the last year, there have been measurable improvements in the way we deliver services.”).

Throughout this discussion, percentages are rounded to the nearest whole number.

As noted, when we break the results down by agency the two middle levels (middle managers and supervisors) are sometimes more or less favorable about the management tools than their level would predict. In most cases the deviations seem to reflect an organizational culture that is specific to that agency. Among some agencies, certain levels were consistently less positive about management practices and results than would be predicted by their level in the hierarchy. For example, managers at the state agencies labeled “2 Southwest” and “8 Mid-West” were frequently less approving than first level supervisors; first level supervisors at “4 Mid-Atlantic” were frequently less positive than non-supervisors; and senior executives at “6 Southwest” frequently had lower opinions than their direct reports.

For all items we can speak of different perspectives. However, for some survey items we cannot speak of “objectively correct” because these items ask each respondent about his or her own job environment. For example, consider two contrasting items within the “rewards and recognition” index. One item states, “Recognizing employees for their achievements are important activities for managers and supervisors.” Any differences of perceptions on this item reflect disagreement about an underlying reality. But another item states “I receive enough recognition for the work that I do.” Disagreement between top and bottom on this item reflects importantly different perceptions of their own individual job rewards under results-based management, and therefore is relevant to the concerns of this article. Nonetheless, the underlying reality—maybe both sides agree the top does get more recognition than the bottom—may not be in dispute.

This research also attempted to use objective measures, such as refund times and complaint rates, as an additional way to judge the accuracy of the responses. However, the results were inconclusive for two reasons. First, many agencies tracked very few measures. Second, measurement categories were inconsistent from agency to agency and also varied over time. These measurement problems recur throughout the public management literature and help explain why most public management research is based on surveys. As discussed in this paper's last section, the management field's continuing dependence on surveys makes this article's implications for survey response accuracy of continuing importance.

Most business studies provide the correlations between top responses and objective measures without explicitly discussing the direction of the miss. However, the few that do indicate it show the top to be too optimistic.

Telephone interviews were conducted at the agencies labeled “3 South,” “5 Northwest,” “6 Southwest,” “7 South,” and “8 Mid-West.” The initial interviewees were chosen randomly, with a few of the later interviewees selected based on suggestions of the initial interviewees. Those interviewed included three senior executives, four managers, five first-level supervisors, and ten non-supervisors. On average, those interviewed had been state government employees for 18 years.

Each of the causes suggested by agency members reflect one or more of the perspectives suggested by four different theoretic schools: organizational culture theories, communications theories, sociological theories of power and status, and psychological theories of selective perception. Explanations grounded in organizational culture theories (for example, Hofstede, Neuijen, Ohayv, and Sanders 1990) focus on organizational readiness to change. Organizational culture explanations suggest that agencies that lack a participative, human relations focus will find that their frontline workers do not accept or believe the organization's characterizations of new management tools. The agency members' suggested causes of top-down miscommunications, top isolation, and lower level pessimism all draw in part from this tradition. Explanations grounded in communications theories (for example, 2003, 309) suggest that disparate perceptions arise less from the preexisting organizational culture, and more from unclear and unproductive patterns of communication which in turn produce misunderstandings and anxiety. The agency members' suggested causes of spin from the bottom and top down miscommunications draw in large part from this tradition. (A single possible cause, such as top down miscommunications, can be supported by several different theoretic schools.) Explanations that draw from political and sociological theories of power and status focus on the ways that all levels will distort information intentionally and unintentionally to make themselves look more successful, and to suggest that the causes of any problems lie at another level. (2006) By imputing an internal power or status motive to the communication and isolation behaviors, all six of the members' explanations could be connected to this tradition. Finally, explanations grounded in individual psychology, especially selective perception explanations, suggest that upper-level operatives view the new management changes with a sense of ownership that will predispose them to unconsciously focus on successes and screen out failures. This clearly fits the members' self-delusion explanation. Such managerial selective perception was first empirically examined by Dearborn and Simon (1958), who posited that a manager's functional position in an organization will serve as a perceptual filter. (See also George and Chattopadhyay 2002.)

This is a shortened version of the definition commonly employed by Wanous et al. Some other definitions emphasize other aspects of the term. For example, Abraham (2000) defines organizational cynicism as “the belief an organization lacks integrity” (p. 269). There is also an ongoing battle in the literature over whether organizational cynicism can be distinguished from organizational skepticism, but that lies outside our focus here.

The continuing commitment promotes successful institutionalization, Kelman finds, for two major reasons: as time unfolds, there is more time for positive feedback mechanisms (including virtuous circles) to develop; and more time allows the development of the “mere exposure” effect, as workers come to embrace the (increasingly) familiar reform.

After discussing their study's finding of different responses between higher and lower UK local managers, Walker and Enticott (2004) state this argument more strongly: “The implication is that empirical reform evidence previously presented needs to be treated with caution—it does not necessarily paint an accurate picture, and if researchers persist in using elite cross-sectional surveys, they will deliver overly simplistic conclusions” (p. 432).

Additional information

Notes on contributors

M. Andrew Frazier

Andy Frazier is Deputy Commissioner for Standards and Inspections with the North Carolina Department of Labor. He received his doctorate in public administration from North Carolina State University.

James E. Swiss

James Swiss ([email protected]) is an associate professor of public administration, School of Public and International Affairs, North Carolina State University.