Theoretical Implications of Measurement Inconsistencies in Organizational Decline Research

Kanak Gautam, David A. Whetten, and Kim Cameron

Figures

Since the first studies that focused on institutions of higher education (e.g. Cameron, Kim, & Whetten, 1987; Cameron, Whetten, & Kim, 1987), a lack of consensus has characterized the measurement of organizational decline. Researchers have used a variety of indicators, both objective and perceptual, in identifying declining organizations. Lack of uniformity in measures can make it difficult to compare results and cumulate findings in new fields of research (e.g., Cameron, Kim, & Whetten, 1987). Consequently, some writers regard lack of consensus about research measures as impeding advancement of knowledge about organizational decline (Whetten, 1987).

In this paper, we argue that, in some instances, consensus regarding measurement may be neither possible nor desirable. This is the case when specific measures, such as different types of objective and perceptual indicators, capture separate facets of the studied phenomenon. As a number of writers on research methods have argued, objective and perceptual indicators sometimes address different realities. Therefore, it may be incorrect to always regard them as substitutes for each other (Campbell & Fiske, 1959; Jick, [End Page 181] 1979). If divergent measures may be capturing different aspects of the research problem, that very divergence should be systematically examined in the hope of revealing new perspectives and reconfiguring existing theories (Jick, 1979).

We agree with this general approach to studying divergent measures and feel that it will provide much-needed research attention on alternate and less studied aspects of decline. For example, studying the divergence between objective and perceptual measures could reveal new aspects of the decline phenomenon--situations where decline is present under one set of definitions and measurement criteria but not under another. Such inquiry can highlight the unique nature of different conceptualizations of decline and their relation to specific types of measures.

To this end, we used both objective and perceptual measures of decline and examine their divergence in a sample of 332 institutions of higher education over an eight-year period. We expected objective and perceptual measures to be related to different conceptualizations of decline and expected their divergence to highlight these alternate conceptualizations. Specifically, we suggested that the existence of objectively measured decline in the absence of perceived decline, and perceived decline in the absence of objectively measured decline, characterize theoretical positions described in the organizational literature as "decline-as-crisis" and "decline-as-stagnation," respectively (Whetten, 1987). We tested these propositions by classifying institutions as experiencing "crisis" or "stagnation" based on divergence between objective and perceptual measures alone. Thereafter, we examined the internal processes in institutions classified as experiencing "crisis" or "stagnation," analyzing whether the internal processes of these institutions were actually consistent with descriptions of "crisis" and "stagnation." We found that institutions suffering from "objective decline" without "perceptual decline" were characterized by processes of "decline-as-crisis." In contrast, institutions with "perceived decline" only were characterized by processes of "decline-as-stagnation."

In this paper, we first review types of measures in the decline literature. Second, we describe the advantages of exploring the divergence between objective and perceptual measures and propose that seemingly contradictory combinations of objective and perceptual measures represent two theoretically distinct and previously unmeasured forms of decline: "decline-as-crisis" and "decline-as-stagnation." Finally, we test our propositions and discuss our findings.

Traditional Measurement of Decline

Past empirical studies on decline used decrease in various objective measures such as revenues, profits, demand, and workforce size, to identify declining [End Page 182] organizations (Greenhalgh, 1983). We preferred such "objective" measures for a number of reasons: for creating uniformity in measurement (Cameron, Kim, & Whetten, 1987); for being relevant to organizations' economic survival; and for being salient signals for decision-makers (Krueger & Willard, 1991). On the other hand, perceptual indicators were advocated for other reasons: for better identifying the early warning signs of decline (Weitzel & Jonsson, 1989), for studying cognitive and emotional reactions to decline (Cummings, 1988), and for addressing its socially constructed aspects (Meyer, 1988).

Researchers' preferences for either objective or perceptual measures are not unusual. In fact, similar preferences have been expressed in other areas of organizational research such as organizational environment and organizational climate. However, some authors have expressed concern about the lack of consensus on the types of measures to use in studying declining organizations (Whetten, 1987; Cameron, Kim, & Whetten, 1987) and have argued for greater convergence among researchers on the type of measures used.

A broad consensus on measures is necessary to advance knowledge in any field of scientific endeavor. On the other hand, lack of consensus among researchers may sometimes reflect differences in conceptions of decline, and a ceaseless quest for measurement uniformity may ignore these alternative perspectives. Therefore, from time to time, instead of seeking uniformity among measures, it may be more productive for researchers to focus on their differences. Divergent measures may reveal new aspects of a phenomenon that went unresearched, due to the use of conventional measures alone.

This view receives support from certain recommendations in the methodology literature. For examples, beginning with Campbell and Fiske (1959), methodologists have argued that the combination of quantitative and qualitative measures generates a fuller, more enriched explanation of research problems (cf., Denzin, 1978; Jick, 1979; Patton, 1980; Webb et al., 1966). This argument is based on the idea that quantitative measures may be capturing aspects of the researched phenomenon not amenable to qualitative measurement, and vice versa. Following this approach, the divergence between different categories of measures, such as qualitative and quantitative measures, is considered a potentially instructive area of inquiry. Divergence between measures may reveal new insights that would be obscured by convergent measurement (Patton, 1980).

A case in point is Jick's (1979) field study that used both archival and qualitative measures of employee satisfaction to examine employee-satisfaction in a post-merger situation. Jick arrived at new insights about employee morale when he followed up contradictory results arising from different types of measures and, based on his study, concluded that discrepant results "may help to uncover the deviant or off-quadrant dimension of [End Page 183] a phenomenon. Divergent measures uncover elements that do not fit conventional theories or models. Thus, old theories are refashioned or new theories developed" (p. 609). Consistent with this view, we propose as a promising area of research the likely divergence between objective and perceptual measures of decline. This approach is particularly appropriate for organizational decline research because it is consistent with complementary, but previously unstudied, theoretical perspectives on decline. In some of the early writing on this subject, Whetten (1987) discussed two seemingly contradictory aspects of decline termed as "decline-as-stagnation" and "decline-as-crisis." The first form of decline involves organizations that are slowly losing their competitive edge, such as a state university, the budget of which lags behind the rate of inflation. There is no precipitating crisis to galvanize faculty support for change; the university's financial strength is slowly dissipating even though the erosion is difficult to quantify. The long-term consequences of stagnation can lead members to report considerable dissatisfaction with organizational conditions. The paradox is that these reports of dissatisfaction would seem unjustified, based strictly on the objective revenue picture.

In the second case, an organization faces an emergency. Its revenues have dropped off substantially due to increased competition, shifting consumer preferences, etc. One would naturally expect to find high levels of member dissatisfaction under these difficult, stressful, and uncertain circumstances. However, these conditions are often the breeding ground for organizational renewal, in which new directions are charted and members' faith in the mission and leadership of the organization is renewed. When this renewal occurs, the associated optimism of organizational participants contrasts sharply with the gloomy picture that would logically be derived from recent income statements.

What is common about these twin paradoxical outcomes is the expected inconsistency between objective and perceptual measures of organizational decline. Our purpose, therefore, is to isolate inconsistent cases of decline measurement in a large sample of organizations and to examine plausible theoretical explanations for their occurrence, based on these two forms of decline. Specifically, we focus on cases where either (a) members perceive less decline than objective measures indicate ("Feels good, looks lousy"), or (b) members perceive a greater magnitude of decline than objective measures warrant ("Feels lousy, looks good"). As shown in Figure 1, we call the first form of inconsistency "performance-decline," implying that objective measures are predominantly negative, as compared to members' perceptions, and that the organization's decline is mostly a matter of historical record. We call the second form of inconsistency "perceived-decline," implying that members' perceptions are predominantly negative in contrast [End Page 184] to objective measures and that the organization's decline is mostly a matter of negative perceptions. We then test whether the two sets of inconsistencies are systematically related to organizational characteristics associated with "decline-as-stagnation" and "decline-as-crisis."

Performance-Decline, or Decline-as-Crisis

Performance-decline develops when members of an organization view its performance more optimistically than objective measures of its performance warrant. One explanation for this apparent inconsistency is that the organization is experiencing a period of "renewal." Strategy theorists studying organizational decline have primarily focused on the positive "turn-around" effects of organizational decline. Their view is that crises often shock organizational leaders into making adaptive changes (Bibeault, 1982; Hambrick & Schecter, 1983; Hofer, 1980). For example, based on their study of several business near-failures, Nystrom and Starbuck (1984) argue that performance decline and the attendant shock of failure in the marketplace often stimulate major changes in mature organizations. The rejuvenating effects of declining performance are, apparently, not restricted to business organizations. Manns and March (1978) and Koberg (1987) present evidence of resource scarcity associated with curriculum, process, and strategic changes in educational institutions. In addition, Meyer's (1982) study [End Page 185] of "environmental jolts" in health care organizations indicates that organizational discontinuities create opportunities to introduce adaptations in hospitals that would not have otherwise been possible.

Renewal in distressed organizations requires a significant alteration of prevailing organizational mindsets because the shared way of thinking and behaving among participants is often characterized by despondency and alienation. In particular, the organizational renewal literature argues that leaders must create an optimistic vision for the future by defining new missions and goals for the organization (Tichy & Ulrich, 1984; Tichy & Devanna, 1986). By helping members of troubled organizations identify with new strategic initiatives, leaders encourage organizational participants to identify with the organization's hopeful future, rather than its dismal past. The literature on turn-around management also indicates that organizational leaders need to respond to environmental changes with an aggressive proactive stance (Whetten, 1980). Studies show that firms which successfully respond to declining performance tend to develop new products and policies in an effort to reposition the organization (Bibeault, 1982; Hambrick, 1985). Such efforts are based on well-integrated, broadly discussed plans and strategies (Mintzberg, 1978). For example, D'Aveni and MacMillan (1990) report that survivors of major downturns in organizational performance pay greater attention to the external environment than firms that eventually declare bankruptcy, and Mckinley (1987) shows that sales improvements in manufacturing plants are correlated with new domain initiatives. On the other hand, firms that fail to improve performance tend to be operations-oriented, not environment-oriented. Such organizations seem content with ad-hoc "efficiency" adjustments and nonprioritized budget cuts (Miles & Snow, 1978).

This line of reasoning suggests a plausible theoretical explanation for the decline-as-crisis condition (low objective performance figures, higher-than-expected employee views of performance). Employees' seemingly unwarranted optimism may be a function of their current experiences and expectations for the future. If their organization is undergoing a process of renewal, then that experience may lead them to discount the significance of poor past performance in favor of an optimistic view of the future. This effect could be significant enough to color their views of the past. That is, when members of a low-performing organization that is undergoing a major revitalization are asked to rate their organization's previous performance, their current optimism may produce a more favorable retrospective evaluation than performance statistics would justify.

In summary, the decline-as-crisis view suggests that organizations with negative performance figures are likely to engage in some form of strategic realignment, which, in turn, should be reflected in subsequent improvements in the organization's performance score card. Furthermore, members [End Page 186] of turnaround organizations should be aware of positive initiatives before they show up on the balance sheet. A logical test of whether this theoretical perspective accounts for the discrepancy between the negative performance indicators and the unexpectedly positive views of members would consist of (a) an increase in organizational performance following the period when members' reports are collected, and (b) members' awareness of current organizational initiatives to improve performance. Therefore, as a test of this explanation for the decline-as-crisis/performance-decline condition, we propose:

P1: Members of these institutions will report a clearer understanding of organizational goals and policies than their counterparts in perceptual-decline organizations.

P2: Members will attribute a more proactive, strategic orientation to their leaders than members in perceptual-decline institutions will.

P3: Institutions experiencing performance decline will improve their future performances.

Perceived-Decline and Decline-as-Stagnation

The decline-as-stagnation concept is derived from several enduring themes in the sociology of organizations literature, related to the dysfunctional consequences of success and its attendant rapid growth. Starbuck and his associates (Starbuck & Hedberg, 1977; Nystrom & Starbuck, 1984) coined the term "success breeds failure syndrome" to describe the common phenomenon of highly successful organizations which fall victim to their accomplishments. They argue that leaders of very successful organizations often become overconfident of their ability to dominate their markets, resulting in reduced product development and declining product quality. Another consequence of growth is "creeping bureaucracy," including highly formalized procedures and centralized decision-making processes that make it difficult for managers to stay in touch with the rank-and-file members as well as with the needs of their customers (Lorange & Nelson, 1987; Miller & Friesen, 1980). The natural tendency for growing organizations to become structurally more complex and hierarchical is typically associated with members' reporting dissatisfaction with lack of participation in key decision-making activities, as well as a sense of alienation from the senior decision makers (Greiner, 1972).

An associated characteristic of highly successful organizations is internal resistance to change (Whetten, 1987). As formal roles and positions become associated with successful programs and policies, members become increasingly threatened by proposed reorientations that might erode their power base, including their expertise in using established routines to solve problems (Starbuck, Greve, & Hedberg, 1978; Tushman & Romanelli, 1985). [End Page 187] In addition, as organizations mature, they develop long-standing dependencies with such stakeholders as unions, suppliers, and customers. These stakeholders resist changes that threaten their interest in the organization, even at the expense of risking organizational failure (Meyer & Zucker, 1989). Studies of reform-oriented government organizations have observed that flexible, open structures turn rigid during the give-and-take of negotiating relationships with opposing interest groups.

This view of decline suggests an explanation for the decline-as-stagnation condition (steady performance figures coupled with lower-than-expected employee views of decline). Assuming that members' ratings of an organization's performance are grounded in their current experiences as well as their expectations for the future, we might expect that stagnant organizational processes, typified by bureaucratic procedures, slow decision-making processes, entrenched commitment to the status quo, and so forth, would dampen members' views of their organization's well-being. Specifically, members' immediate negative experiences and their dim prospects for change could lead them to discount the organization's relatively attractive past performance record. In this manner, members' perceptions serve as leading indicators that growing organizations are becoming the victims of their success.

A logical test of whether this theoretical perspective accounts for the difference between the positive performance indicators and the less-positive-than-expected performance perceptions of members would consist of (a) a decrease in performance following the period when members report stagnant organizational processes, and (b) members' concerns about ineffective organizational processes that dampen the prospects for change. This explanation for the decline-as-stagnation/perceived-decline condition suggests the following propositions:

P4: Members of these institutions will report higher levels of resistance to change in their organizations than members in the performance-decline condition.

P5: Members will also report higher levels of centralized decision-making than their counterparts in performance-decline institutions.

P6: Institutions in the perceived-performance condition will experience relatively lower performance during the subsequent time period.

Study Design and Measures

This study used survey data collected from 332 institutions of higher education in the United States. The selected institutions represented a stratified random sample of all four-year colleges and universities based on institutional size (200-20,000 FTE), institutional type (public-private) and degrees offered (bachelor's, master's, doctorate). Respondents were 3,460 [End Page 188] members of the campus administration (departmental heads, central administrators, and trustees). We obtained respondent names from each institution and mailed a questionnaire personally to each, promising anonymity for themselves and their organizations. We sampled administrators and trustees randomly within each institution but gave special attention to assuring heterogeneity in disciplines among department heads and locations and titles among trustees.

For this study, we concentrated on institutions that had actually experienced revenue declines during one of the four previous years (1979-1983), or whose members had perceived a decline in revenues during at least one of these years (N = 240), based on performance data from the Higher Education General Information Survey (HEGIS). We assigned respondents to "performance-decline" and "perceptual-decline" categories as follows. We made a comparison between survey questions asking respondents to check the years in which they believed revenues had decreased from the previous year, and actual revenue data for each institution obtained from the HEGIS data files. The questions referred to the four years between 1978-1979 and 1981-1982. For each organization, we calculated the discrepancy between the actual number of decline-years and the perceived number of decline-years for each respondent, averaging the total across respondents to produce an organizational score. The distribution of organizational scores had a mean of 0.37 and standard deviation of 1.09. The distribution was somewhat skewed toward the left (kurtosis = 0.16, max = 3.09, min = -2.8), implying that respondents were somewhat more likely to err in inferring that decline had occurred when it had not than that decline had not occurred when in fact it had.

Institutions were classified as perceived-decliners if their discrepancy scores were at least one standard deviation below the mean (objective decline measure substantially less than the perceptual decline measure), and as performance-decliners if their discrepancy scores were at least one standard deviation above the mean (objective decline measure substantially greater than the perceptual decline measure). We used a cut-off of one standard deviation as a conservative measure to allow for examination of only those organizations whose respondents had significantly over- or underestimated the existence of decline. Again, since the discrepancy scores measured differences between actual and perceived number of decline-years, a high score below the mean implied that organizational respondents had overestimated the occurrence of decline beyond its actual frequency (perceived-decline), while a high score above the mean implied that respondents had underestimated the frequency of decline relative to its occurrence (performance-decline).

Based on this criterion, 82 institutions had discrepancy scores beyond the cut-off limits. Of these, we classified 42 institutions as perceived-decliners [End Page 189] and 40 as performance-decliners. We used the two categories of decline as independent variables in our subsequent analysis to examine if they were associated with significant differences in organizational processes and performance.

Dependent variables were measures of organizational processes associated with both types of decline (proactiveness, goal-identification, resistance to change, and centralization). We operationalized these measures as the average members' ratings of the extent to which these attributes were present at each school. Since the dependent variables were based on scores averaged from the different respondents within an institution, we tested the amount of agreement among respondents within each institution through a repeated measures approach using one-way ANOVA. If more agreement existed in the ratings of the item among respondents within an institution than among respondents outside the institution, then the reliability of the aggregated "organizational" measures is high. The assumption is that the item is being measured multiple times (across multiple respondents) in a consistent fashion. A significant main effect for the institution resulted from each item (F's ranged from 1.32 to 1.79, d.f. = 329, 3070, p < .000), suggesting that each item was indeed reliable in terms of the repeated measures criteria.

We based evaluations of organizational proactiveness on responses to the following items regarding the extent of strategic activity within the organization: (a) the top administration team had developed multi-year strategies to achieve long term institutional objectives; (b) this institution had established new domains of activity; and (c) innovative activity in this organization was increasing. (Cronbach's Alpha = 0.63.)

We measured the extent of goal-identification through the following items: (a) people associated with this institution share a common definition of its mission; (b) there is a general sense that this institution has a distinctive purpose to fulfill; and (c) this institution has a special identity unlike any other in higher education. (Cronbach's Alpha = 0.85.)

We measured resistance to change as the summed response to three items that assessed the level of opposition to change in the organization: (a) there is a lot of resistance to change in this organization; (b) special interest groups within the institution are becoming more vocal; and (c) conflict is increasing within this institution. (Cronbach's Alpha = 0.79.)

Three items measured the extent to which organizational decisions were centralized: (a) major decisions are very centralized; (b) resource allocation is decided autocratically; and (c) resource allocation is decided by one person. (Cronbach's Alpha = 0.79.)

In addition to these dependent variables, we constructed measures of past and future performance using the HEGIS data. Past performance was [End Page 190] measured as the rate of reported growth or decline in revenues over the four year period (1979-1982) prior to the survey. Revenues in year t +1 were divided by those in year t. The resultant rate of change for each year was summed over the period of 1979-1982 with future performance measured as the rate of reported growth or decline in revenues for the four-year period (1983-1986) after the survey. Again, we divided revenues in year t +1 by revenues in year t and summed the rate of change over the period of 1983-1986.

Two organizational characteristics were also included as controls in the analysis. We included institutional size (0 = enrollment < 5000, 1 = enrollment > 5000), and institutional type (0 = public, 1 = private) to test for structural and cultural differences that might account for variations in performance and internal processes among universities.

Analysis and Results

We first analyzed the data using Multivariate Analysis of Variance (MANOVA), using decline categories (perceived-decline versus performance-decline) as independent variables, and internal process variables as dependent variables. The results were significant at the .01 level with an F value of 6.51 (d.f. = 6), suggesting that there are significant differences between the two categories based on the multiple dependent variables. We conducted further analyses using the control variables as covariates. Our results remain significant, despite the effects of institutional size and control. To determine more precisely and parsimoniously the meaning of these results, we performed univariate analyses of variance, using each of the eight different measures as dependent variables.

The results shown in Table 1 support all eight propositions. To begin with, the performance-decline institutions exhibit the characteristics typical of decline-as-crisis organizations: negative past performance (-6%), higher proactiveness, and higher awareness of organizational goals. Moreover, the fact that these institutions show improved future performance during the four-year period following the study (+14%) provided further evidence that they were undergoing renewal during the time of our survey. This renewal suggests that members' reports of positive organizational processes were not mere rationalizations but were, instead, harbingers of improved performance.

Table 1 also indicates that the perceived-decline institutions exhibit the characteristics associated with decline-as-stagnation. These were: positive past performance (+11%), higher levels of centralization, and resistance to change. In addition, these institutions displayed deterioration in future performance following the survey (5%), suggesting again that members' views foreshadowed impending changes in organizational performance. [End Page 191]

Discussion

It is important to acknowledge certain limitations of this study. First, using previously collected survey data forced us to use categorical indicators of decline, causing loss of variance in our measures. Second, though our study emphasizes the use of multiple measures, we were limited to using objective and perceptual indicators related only to organizational revenues. Third, the differences we report are statistically significant but relatively small, given our categorical independent variables. Thus, our conclusions must be regarded as tentative, pending further explorations involving more robust measures.

Within these limitations, our results suggest the following conclusions regarding the definition and measurement of organizational decline, their implications for organizations in general, and their implications for institutions of higher education in particular. First, and foremost, how we measure decline significantly affects the conclusions we draw from our research about the causes, processes, and consequences of decline. While this proposition sounds terribly obvious in the abstract, the history of research on organizational decline indicates how often it is overlooked in practice. As a case in point, Figure 2 demonstrates the erroneous conclusions that can be drawn when only one form of measurement is used. This figure portrays the indexed revenue performance of both types of institutions in the study based on objective indicators alone. A study done in 1983 using objective measures alone would classify institutions with the decreasing revenue curve as declining and those with the rising curve as growing. Yet researchers looking at perceptual measures alone would have reached quite the opposite [End Page 192] conclusion. This result also suggests that relying on any single type may be misleading from a practical viewpoint. For example, revenue and enrollment statistics in a university may have little relation to current plans and processes that affect its future performance. However, external stakeholders such as state regulators who lack access to internal processes may be disproportionately influenced by current performance figures.

Second, our study suggests that how we measure decline has a significant impact on how we conceptualize decline. This observation helps reconcile seemingly incompatible definitions of decline in the literature. It is apparent from our study that objective and perceptual measures of decline tap different aspects or phases of organizational decline. The dominant view in empirical literature is that organizations enter the stage of decline when consistently negative financial indicators appear as a result of "a substantial, absolute decrease in an organization's resource base" (Cameron, Kim, & Whetten, 1987, p. 209) or "loss of financial resources" (Sutton & D'Aunno, 1989, p. 210). An alternate view in the conceptual literature is that organizations [End Page 193] experience decline prior to the time when adverse performance numbers become visible, "when they fail to anticipate, recognize, avoid, neutralize, or adapt to external and internal pressures" (Weitzel & Jonsson, 1989, p. 209), or "when the organization fails to maintain the adaptiveness of its responses to a stable environment" (Greenhalgh, 1983, p. 232). Combining these perspectives suggests the usefulness of a conceptualization that uses multiple stages or phases of organizational decline, in which a decline in organizational adaptiveness is followed by a decline in actual performance. This view suggests that, rather than being alternative forms of decline, decline-as-stagnation and decline-as-crisis may actually be sequential, progressive stages of decline. Perhaps it would be useful to analyze organizational failures (for example, the closure of a college), examining such proxy measures for maladaptation as quality of administrators, administrative turnover, changes in programs and curricula, etc., to determine the lag between these measures and eventual financial distress and closure. Such information would be valuable in mapping the relation between the phases of "stagnation" and "crisis."

Third, the conceptualization of decline as a series of stages requires researchers to match measurement form with decline form. For example, the empirical identification of decline with negative accounting figures favors studies of organizational decline "after the fact" but discourages inquiry into its early stages when accounting indicators are not yet affected. Cameron, Kim, & Whetten (1987) found that negative attributes generally associated with decline were found in stable organizations with no prior record of decreased performance. However, there has been little follow-up work on processes of the decline-as-stagnation condition (Whetten, 1987), because such deteriorating adaptive processes remain a matter of perception rather than accounting statistics. Longitudinal studies using employee surveys and other qualitative information could be used to track early signs of organization stagnation as a university increases in size and begins losing its culture of growth and innovation.

This line of thought highlights the need for valid, operational measures for "prospective" studies of organizational decline. Because it is difficult, using traditional performance measures, to determine whether an organization is about to decline (Hambrick & D'Aveni, 1988), alternate "anticipatory" measures need to be explored. Various stakeholders' assessments of organizational vitality may be good prospective measures, especially in institutions of higher education where organizational effectiveness often depends on satisfying many different internal and external stakeholders (Cameron, 1978).

Fourth, the conception of decline as "anticipated phenomenon" serves as a cautionary note in interpreting past findings on the aftermath of decline. Cameron, Kim, and Whetten's (1987) study on the dysfunctional effects [End Page 194] of decline demonstrated that subjects' perceptions of low morale, conflict, and turnover were positively correlated with conditions of objective decline. However, these perceptions could also be interpreted as prospective rather than retrospective assessments. In other words, perceptions of negative processes may be anticipations of deteriorating future performance rather than perceived effects of past decline. Such reasoning is consistent with empirical studies of organizations experiencing downward spirals (Hambrick & D'Aveni, 1988), in which early signs of organizational impairment tend to be amplified in later stages. It follows logically that confusion regarding whether perceived organizational dysfunctions are cause or effect clearly influences the manner in which they will be managed. These considerations further underscore the need for time-series data to examine whether dysfunctions follow from deteriorating performance or are prognostications of future events (Hambrick & D'Aveni, 1988).

Fifth, this treatment of decline as a continuous process, rather than as a discrete event experienced as a crisis, raises questions regarding the relevance of existing decline research for administrators (Weitzel & Jonsson, 1989). By the time an organization's characteristics match traditional definitions of decline, the organization has been seriously impaired (Weitzel & Jonsson, 1989). Not surprisingly, while studies have examined such dysfunctional consequences of decline as structural inertia, enhanced conflict, resistance to change and insolvency (e.g. Mckinley, 1987; Cameron, Whetten, & Kim, 1987; D'Aveni, 1989), there have been few investigations of interventions to prevent or manage decline. As D'Aveni and MacMillan (1990) suggest, helping practitioners deal preemptively with decline requires researchers to examine how inadequate scanning behaviors, short-term planning and crisis-denial contribute to deteriorating performance. For example, administrators in higher education need to monitor their reputation in the market irrespective of their financial performance, to anticipate possible budgetary problems in the future. Again, an important requirement is to develop qualitative indicators that can help administrators predict the early stages of decline, e.g., from employee surveys, expert assessments, and customer feedback. Further field research could be done to produce benchmarks regarding change in such measures of effectiveness across different stages of decline.

Finally, our study suggests that combining forms of measurement and examining their inconsistencies for clues to important substantive insights is a fruitful research strategy for organizational decline in general, as well as for research on institutions of higher education in particular. As Poole & Van de Ven (1989) suggest, organizational scholars tenaciously pursuing consistency in their explanations should focus on inconsistencies instead, as a way of exploring novel explanations for seemingly contradictory results. One approach they suggest for resolving inconsistent results is to locate [End Page 195] them in different time frames. This suggestion appears consistent with our use of the constructs of decline-as-crisis and decline-as-stagnation to explain our discrepant results, inasmuch as they occur at different temporal junctures of the organizational decline process. Other prescriptions, such as locating inconsistent elements at different analytic levels, may help resolve other inconsistencies, such as contradictory findings regarding increased versus decreased centralization in the wake of decline in institutions of higher education (Cameron, Whetten, & Kim, 1987). For example, in universities, centralization measures may be lower at the board level, parallel to the increased autonomy of divisions in recent corporate restructurings (Zajac & Kraatz, 1993), but higher at the departmental levels to facilitate coordinated and timely responses to the fast-changing market. Testing substantive explanations of measurement inconsistencies promises to provide further insights into complex process-outcome relationships in organizations in general and universities and colleges in particular. This study argues for more of this type of inquiry for building and testing theoretical propositions.

Kanak Gautam is Assistant Professor in the Department of Health Administration, Saint Louis University, St. Louis, Missouri. David A. Whetten and Kim Cameron are both Professors in the Department of Organizational Behavior, Marriott School of Management, Brigham Young University, Provo, Utah.

References

Bibeault, D. (1982). Corporate turnaround. New York: McGraw-Hill.

Cameron, K. (1978). Measuring organizational effectiveness in institutions of higher education. Administrative Science Quarterly, 23, 365-380.

Cameron, K., Kim, M., & Whetten, D. (1987, March). Organizational effects of decline and turbulence. Administrative Science Quarterly, 27, 1-16.

Cameron, K., Sutton, R., & Whetten, D. A. (Eds.). (1988). Readings in organizational decline: Frameworks, research, and prescriptions. Boston: Ballinger.

Cameron, K., Whetten, D., & Kim, M. (1987, March). Organizational dysfunctions of decline. Academy of Management Journal, 30, 126-138.

Campbell, D. T., & Fiske, D. (1959, March). Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulletin, 56, 81-105.

Cummings, L. L. (1988). Organizational decline from the individual perspective. In Cameron, Sutton, & Whetten (pp. 417-424)

D'Aveni, Richard. (1989, September). The aftermath of organizational decline: A longitudinal study of the managerial and strategic characteristics of declining firms. Academy of Management Journal, 32, 577-605.

D'Aveni, R., & MacMillan, I. (1990). Crisis and content of managerial communications: A study of the focus of intention of top managers in surviving and failing firms. Administrative Science Quarterly, 35, 634-657.

Denzin, N. (1978). The research act: A theoretical introduction to sociological methods. New York: McGraw-Hill.

Greenhalgh, L. (1983). Organizational decline. In S. Bacharach (Ed.), Research in the sociology of organizations, 2, 231-236. Greenwich, CT: JAI Press Inc.

Greiner, L. (1972, July-August). Evolution and revolution as organizations grow. Harvard Business Review, 4, 37-46.

Hambrick, D. (1985). Turnaround strategies. In W. Guth (Ed.), Handbook of business strategy, pp. 10-1 to 10-31. Boston: Warren, Gorham and Lamont.

Hambrick, D., & D'Aveni, R. (1988, March). Large corporate failures as corporate spirals. Administrative Science Quarterly, 33, 1-23.

Hambrick, D., & Schecter, S. (1983, June). Turnaround strategies in mature industrial-product business units. Academy of Management Journal, 26, 231-248.

Hofer, C. (1980, Summer). Turnaround strategies. Journal of Business Strategy, 1 (Summer 1980): 19-31.

Jick, T. (1979, December). Mixing qualitative and quantitative methods: Triangulation in action. Administrative Science Quarterly, 24, 602-609.

Koberg, C. (1987). Resource scarcity, eenvirontal uncertainty, and adapative organizational behavior. Academy of Management Journal, 30, 798-807.

Krueger, D., & Willard, G. (1991). Turnarounds: A process, not an event. Academy of Management Best Paper Proceedings, 1991.

Lorange, P., & Nelson, R. (1987, Spring). How to recognize and avoid organizational decline. Sloan Management Review, 28, 41-48.

Manns, C., & March, J. (1978). Financial adversity, internal competition, and curriculum change in a university. Administrative Science Quarterly, 23, 541-552.

Mckinley, W. (1987, March). Complexity and administrative intensity: The case of declining organizations. Administrative Science Quarterly, 32, 87-105.

Meyer, A. (1982). Adapting to environmental jolts. Administrative Science Quarterly, 27, 515-537.

Meyer, A. (1988). Organizational decline from the organizational perspective. In Cameron, Sutton, & Whetten (pp. 411-416).

Meyer, M., & Zucker, L. (1989). Permanently failing organizations. Newbury Park, CA: Sage Publications.

Miles, R., & Snow, C. (1978). Organizational strategy, structure and process. New York: McGraw-Hill.

Miller, D., & Friesen, P. (1980). Archetypes of organizational transitions. Administrative Science Quarterly, 25, 268-299.

Mintzberg, H. (1978, May). Patterns in strategy formation. Management Science, 24, 934-948.

Nystrom, P., & Starbuck, W. (1984, Spring). To avoid organizational crises, unlearn. Organizational Dynamics, 12, 53-65.

Patton, M. (1980). Qualitative evaluation methods. Beverly Hills, CA: Sage Publications.

Poole, M., & Van de Ven, A. (1989, October). Using paradox to build management and organization theories. Academy of Management Review, 14, 562-578.

Starbuck, W., & Hedberg, B. (1977). Saving an organization from a stagnating environment. In H. Thorelli (Ed.), Strategy + Structure = Performance. Bloomington: Indiana University Press.

Starbuck, W., Greve, A., & Hedberg, B. (1978, Spring). Responding to crisis. Journal of Business Administration, 8, 111-118.

Sutton, R., & D'Aunno, T. (1989). Decreasing organizational size: Untangling the effects of money and people. Academy of Management Review, 14, 194-212.

Tichy, N., & Devanna, M. (1986). The transformational leader. Training and Development Journal, 40, 27-32.

Tichy, N., & Ulrich, D. (1984, Fall). The leadership challenge. Sloan Management Review, 59-68.

Tushman, M., & Romanelli, E. (1985). Convergence and upheaval: Managing the unsteady pace of organizational evolution. California Management Review, 29, 29-44.

Webb, E. J., Campbell, D., Schwartz, R., & Sechrest, L. (1966). Unobtrusive measures: Non-reactive research in the social sciences. Chicago: Rand McNally.

Weitzel, W., & Jonsson, E. (1989, March). Decline in organizations: A literature integration and extension. Administrative Science Quarterly, 34, 91-109.

Whetten, D. (1980, October). Organizational decline: A neglected topic in the organizational sciences. Academy of Management Review, 5, 577-588.

Whetten, D. (1987). Growth and decline processes in organizations. In Annual Review of Sociology, 13, 335-358.

Zajac, E., & Kraatz, M. (1993). A diametric forces model of strategic change: Assessing the antecedents and consequences of restructuring in the higher education industry. Strategic Management Journal, 14, 83-102.

[an error occurred while processing this directive].