essay questions based on articles

GROUP & ORGANIZATION MANAGEMENTWanous et al. / CYNICISM

Cynicism About Organizational Change

MEASUREMENT, ANTECEDENTS, AND CORRELATES

JOHN P. WANOUS

ARNON E. REICHERS Fisher College of Business The Ohio State University

JAMES T. AUSTIN The Ohio State University

A new construct called Cynicism About Organizational Change (CAOC) was proposed and dis- tinguished from related concepts. The measure of CAOC was supported by confirmatory factor analysis and has acceptable internal consistency reliability. Potential antecedents (measured 21 months before the measurement of CAOC) were examined. Little support was found for CAOC having dispositional roots in one’s general negative affectivity. More support was found for CAOC being learned as a result of little previous change, ineffective leadership practices, and lack of participation in decisions. CAOC was negatively correlated with a concurrent measure of organizational change and with the motivation to keep on trying to support change efforts. In addition, CAOC was negatively correlated with factors outside the realm of change: organiza- tional commitment and the number of labor grievances. Finally, CAOC weakened the instru- mentality perception of the relationship between performing well and earning more money, while holding the actual pay system constant.

The newspaper comic strip Dilbert and the current best-selling book by the same author, The Dilbert Principle (Adams, 1996), as well as other media, are filled with examples of employee cynicism in organizations that are led by uncaring and/or incompetent leaders. The Dilbert Factor was a sympo- sium title at the 1997 American Psychological Association’s annual meeting (Shapiro, 1997).

Cynicism has been studied by academic researchers in a variety of ways (see recent reviews by Andersson, 1996; Andersson & Bateman, 1997). In brief, cynicism research ranges from studies of general cynicism to the cyni-

Group & Organization Management, Vol. 25 No. 2, June 2000 132-153 © 2000 Sage Publications, Inc.

132

cism associated with specific occupations, organizations, or leaders. As an example of the more general approach, those who developed the Minnesota Multiphasic Personality Inventory (MMPI) treated cynicism as an aspect of personality, defining it as a basic distrust of other people because their motives are self-serving and open to suspicion (Pope, Butcher, & Seelen, 1993). Kanter and Mirvis (1989) also measured cynicism as a general, global view of other people and then related their measure to both societal and organiza- tional problems.

Others, however, have treated cynicism as a more specific construct. For example, there are at least 30 studies of cynicism among police officers, most of which were inspired by the seminal work of Niederhoffer (1967). Cyni- cism directed at one’s work organization has been of more recent origin. Although Kanter and Mirvis (1989) treated cynicism as a general orientation, they also suggested that it can be applied to one’s work organization. A study by Andersson and Bateman (1997) is a good example of studying cynicism as both a general and specific construct. They measured cynicism in three ways: (a) toward one’s specific organization, (b) toward business executives, and (c) toward human nature in general.

Most of the work on cynicism conducted to date is typical of the first stage of scientific research development. (See Reichers & Schneider, 1990, who developed a stage model. Although their model specifically concerned the development of organizational climate and culture as a research topic, we think that their model also applies to cynicism.) The first stage is called intro- duction and elaboration because concepts are introduced into a particular field and their presence is justified in a variety of ways. In this first stage a variety of operational measures are introduced and the concept is often treated as both an independent and a dependent variable.

The present research on cynicism complements other recent efforts by focusing on a specific form of cynicism within an organizational context, Cynicism About Organizational Change (CAOC) (Reichers, Wanous, & Austin, 1997; Wanous, Reichers, & Austin, 1994). As is typical of first stage research, we will examine both its antecedents (using longitudinal data) and its consequences (using concurrent data). CAOC is defined here as a pessi- mistic viewpoint about change efforts being successful because those responsible for making change are blamed for being unmotivated, incompe- tent, or both.

The present research on cynicism about organizational change seems timely given the widespread change-oriented turmoil in which many Ameri- can organizations currently find themselves. Current popular books on man- agement, business-oriented magazines, and newspapers are filled with advice

Wanous et al. / CYNICISM 133

and stories about rapid change, how to avoid failure when attempting change, how individuals should cope with change, and so forth. Aktouf (1992) has criticized some of the writers on organizational change because they fail to account for the reactions of individual employees to change attempts. For example, change efforts that use approaches such as team spirit (e.g., Tjosvold & Tjosvold, 1991), empowerment (e.g., Peters & Waterman, 1982), transfor- mation (e.g., Peters & Austin, 1985), culture change (e.g., Schein, 1985), reengineering (e.g., Hammer & Champy, 1993), and commitment to quality (e.g., Juran & Gryna, 1980) all fail to consider the employees of the organiza- tion on whom the new leadership or new culture is supposed to work.

If there is widespread cynicism about change in a particular organization, it seems likely that even the most sincere and skillful attempts at organiza- tional change will be impeded by the prevailing cynicism. It is our assump- tion that high levels of CAOC tend to become their own self-fulfilling proph- ecy. Failure to address this type of cynicism is another example of what Aktouf (1992) referred to as a blindness among current management gurus.

However, another body of work on organizational change has paid con- siderable attention to the motivation of employees to accept change. For example, Likert (1967) and others from the University of Michigan (e.g., Mann, 1957) expressly acknowledge the critical role of individual employ- ees in successful change. Their approach to gaining employee support for change efforts involves the gathering of extensive survey data followed by data feedback and discussion to proceed in a participative way. This has come to be known as the action-research approach to change (see Susman & Evered, 1978, for the advantages of this method). An underlying assumption of this approach is that individual employees are viewed as active, important elements in change and as collaborators in research.

However, even this second body of literature seems to assume that the action research method and/or employee participation programs will be suc- cessful regardless of the organization’s past history of change attempts. In doing so, it overlooks the possibility that past failures may limit or even doom efforts at organizational change because cynicism might become its own self-fulfilling prophecy.

Research on organizational development (OD) has been reviewed a num- ber of times over the last 25 years, beginning with Friedlander and Brown (1974) and followed in order by Porras and Berg (1978), Guzzo, Jette, and Katzell (1985), Blanck and Turner (1987), Neuman, Edwards, and Raju (1989), and Porras and Robertson (1992). The size of the OD literature reviewed depends greatly on how one defines the domain. Some reviews included laboratory studies of goal setting, which seems to us to be outside

134 GROUP & ORGANIZATION MANAGEMENT

the OD domain. Most reviews categorize OD interventions and then assess impact on a category-by-category basis. Others assess the overall strength of a study’s research design and assess results separately for groups that range from strong designs to weak ones. However, none of the reviews reported any study that accounted for employee cynicism. Furthermore, no individual difference variables that might affect the success of OD efforts have been assessed in these reviews. In a word, the assessment of OD has been exclu- sively concerned with its implementation process and specific types of interventions.

The present research is an initial study of three exploratory questions con- cerning cynicism about organizational change. The first research question concerns the meaning and measurement of this construct. Confirmatory fac- tor analysis is used to assess the conceptual definition of CAOC. The second research question concerns potential antecedents of cynicism and is exam- ined with data gathered 21 months before the measurement of CAOC. The third research question concerns some concurrent relationships between CAOC and several job attitudes and behaviors.

Exploratory Research Question 1: Defining cynicism about organizational change

Cynicism about organizational change is defined as a construct that has two elements: a pessimistic outlook for successful change and blame placed on “those responsible” for lacking the motivation and/or the ability to effect successful change. This approach to CAOC originated in verbal reports from the employees of the organization in which this research was conducted (i.e., both formal interviews that were transcribed and informal conversations with employees at work). The phrase “those responsible” was chosen to include both managerial and union leaders because the organization studied was unionized. Although CAOC resulted from the action-research nature of this study, it has strong similarities to elements from both expectancy theory (e.g., Vroom, 1964) and attribution theory (e.g., Jones & Davis, 1965).

Expectancy theory is relevant because beliefs about the likelihood of suc- cessful organizational change are similar to beliefs that good performance will result from personal efforts (i.e., the expectancy component in expec- tancy theory). The frame of reference for cynicism is the likelihood of suc- cessful organizational change resulting from the efforts of those responsible. Thus, it is a type of expectancy.

Attribution theory is relevant because it is concerned with how individu- als assess the causes of another person’s behavior. Cynicism is believed to

Wanous et al. / CYNICISM 135

result from a dispositional attribution (i.e., blaming others) for the failure of change. This blame can be based on a lack of motivation, a lack of ability, or both. In contrast, cynicism is unlikely to result when a situational attribution for the failure of change is made by employees (e.g., unforeseen events or forces beyond the control of management) because there is no one to blame.

Exploratory Research Question 2: Antecedents of cynicism about organizational change

The second exploratory research question concerns some possible ante- cedents of CAOC. The gathering of antecedent data (21 months prior to the measurement of cynicism) was completed before the present conceptualiza- tion of CAOC was formulated. As a result, the analysis of antecedents is kept relatively simple due to the post hoc nature of this particular question. The rationales for selecting the antecedents follow next.

Negative affectivity (NA) (Watson, Clark, & Tellegen, 1988) is a person- ality variable that concerns an individual’s general outlook on life. To the extent that people are generally negative, they might be predisposed to be cynical as opposed to learning to become cynical from experiences at work. A second reason for choosing NA is related to the grounded aspect of this research. That is, numerous managers voiced a favorite hypothesis about why some employees were cynical about organizational change. They believed that these individuals were negative about everything, not just change. In fact, they were referred to as bad attitude or bad apple employees. By examining the strength of relationship between NA and CAOC we address the pet hypothesis of the managers. Furthermore, we can also address the basic issue of how a personality factor like NA might predispose some employees to be cynical.

The second type of antecedent concerns those factors that might account for the emergence of CAOC as a result of one’s experience in an organiza- tion. Although there are probably many possibilities, three were selected here for the following reasons. The first organizational factor measured was the amount of change previously experienced because one aspect of cynicism concerns pessimism about change efforts being successful. The second orga- nizational factor is the role effectiveness of one’s own supervisor, for exam- ple, providing information, listening effectively, being accessible, showing concern, and so forth. If supervisors are perceived as generally ineffective, it is more likely that they will be blamed for failed change as opposed to attrib- uting the failure to factors beyond their control. Third, supervisors control the

136 GROUP & ORGANIZATION MANAGEMENT

amount of employee participation, and participation tends to reduce blaming leaders because employees have more say in decisions and greater under- standing of them. Thus, measures of participation in decision making (PDM) were included as the final type of organizational factor that might lead to the development of cynicism.

Exploratory Research Question 3: Concurrent relationships with cynicism about organizational change

Because the type of cynicism considered here concerns organizational change, those factors that are directly related to the change process are con- sidered first. Two such factors are the amount of recent organizational change and the motivation to keep on trying to make change. The first of these two factors was examined because cynicism might have a reciprocal relationship with the amount of change. Employees might interpret present organizational change efforts through the lens of preexisting cynicism. Thus, highly cynical employees might perceive far less change than those who are less cynical.

The second factor considered was the motivation of individual employees to try to support organizational change through their own efforts. If highly cynical employees do not support—or actively try to inhibit—change efforts, then cynicism becomes a self-fulfilling prophecy.

Finally, relationships between CAOC and employee attitudes, percep- tions, and behaviors outside the realm of organizational change were exam- ined to assess the extent to which there was spillover into two other areas: organizational commitment and the number of labor grievances filed.

Then, the effect of CAOC on the instrumentality perception that good per- formance (P) leads to earning more money ($), that is P → $, was examined in a unique way. This particular facet was selected because it was the only sur- vey measure that also has a clear macro (and objective) element to it. That is, employees in this organization were in one of three types of compensation system: (a) hourly, (b) hourly plus incentive pay, and (c) salary. Because of the considerable differences in payment method, we expect to find differ- ences among the three groups in their P → $ instrumentality perception. More important, however, is the possibility that CAOC might also affect this instrumentality perception. Combining perceptual data from individuals (CAOC and the instrumentality perception) with a true organizational vari- able (pay system) is what some (House, Rousseau, & Thomas-Hunt, 1995)

Wanous et al. / CYNICISM 137

have called meso-analysis because different organizational levels are com- bined into a single analysis.

METHOD

RESEARCH SITE AND DATA COLLECTION

The data reported here were gathered in a large unionized (United Auto Workers) Midwestern manufacturing plant that produces component parts for the automotive industry. Initial entry into the organization was requested by plant management and union leaders. During the course of this 3-year research project, two surveys were administered 21 months apart. Prior to the two surveys, a representative sample of 84 employees was interviewed in groups of 2 to 4 persons for 1 hour by members of the research team.

Survey data were collected on site during the last hour of each work shift. Employees were paid by the company for the time spent completing the ques- tionnaires. If they chose not to participate, they were paid but were not allowed to leave early.

Surveys were collected from all employees. For the hourly employees, 1,164 of 1,773 (65.7%) responded to the first survey, and 870 of 1,648 (52.8%) responded to the second. For the salaried employees, 241 of 349 (69.1%) responded to the first survey, and 162 of 310 (52.2%) responded to the sec- ond. Although the response rates decreased at the second survey, a higher percentage of survey items were completed, making the effective sample sizes about equal when the data were analyzed. It should be noted that the illiteracy rate at this plant was estimated by both management and union offi- cials as 10% to 20%. Furthermore, there was an average 8% daily absentee- ism among the hourly employees. Thus, the response rates for the hourly employees understate the true response rate for that group.

The demographic characteristics of this sample are 80% male, 80% White, and 77% married; average tenure at the plant is 19 years. These char- acteristics are quite typical for those employed in the other plants of this cor- poration, as well as the American automotive industry.

When addressing the first research question (measurement of CAOC) and the third (correlates of CAOC), only the data from the second survey are used. When the second research question (antecedents of CAOC) is consid- ered, data from both surveys are used. This means that the analyses concern- ing the antecedents of CAOC are limited to those individuals who responded to both surveys and who provided their names so the two surveys could be matched.

138 GROUP & ORGANIZATION MANAGEMENT

MEASUREMENT OF CYNICISM ABOUT ORGANIZATIONAL CHANGE

Following the conceptual definition of CAOC developed earlier (Reichers et al., 1997; Wanous et al., 1994), eight items were written to measure its two components (four items per component, alpha = .86): pessimism about change being successful and a dispositional attribution about the likely fail- ure of change efforts (two motivational and two ability reasons). In addition, four other items were written to measure a Situational Attribution about the likely failure of change. All 12 items are presented in the appendix. The 12 items were scrambled rather than grouped, although all appeared on the same page of the questionnaire. All items were measured on a 5-point scale (1 = strongly disagree, 2 = agree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree). Although the situational attribution for the failure of change (alpha = .61) is not considered part of CAOC, it was included in these analy- ses as a check on discriminant validity. It was hypothesized that the situa- tional attributions would correlate with a latent variable separate from the combination of pessimism and the dispositional attributions.

MEASURES OF ANTECEDENTS OF CAOC

All antecedents of cynicism about organizational change were measured on the first survey except for NA. NA (Watson et al., 1988) was measured with a 10-item scale (alpha = .91), for example distressed, irritable, nervous, hostile, and so forth. A 5-point scale was used where 1 = very slightly, 2 = a little, 3 = moderately, 4 = quite a bit, and 5 = extremely. The instructions asked for the respondent to think of the average day at work.

The amount of previous change was measured by summing a scale of five facets ( job activities, others expectations, supervision, effort, and amount of skill needed) for the past 6 months. In addition, the amount of expected change in the same five facets for the upcoming 6 months was included for a total of 10 items (alpha = .90). The amount of participation in decision mak- ing (alpha = .84) was measured with a 5-item scale that resulted from sum- ming four specific facets of decisions (setting production goals, solving pro- duction problems, solving quality control problems, and setting own work procedures) and a global item. Finally, supervisory role effectiveness was measured (alpha = .88) with a 15-item scale that concerned how one’s super- visor treated employees: keeping people informed, personally caring about products being produced, trying to understand each employee’s point of view, keeping commitments, answering questions, caring about employees, acting rationally in a crisis, communicating the same message regardless of who is listening, having a positive attitude about change, not ignoring conflict,

Wanous et al. / CYNICISM 139

admitting mistakes, not blaming others for mistakes, not singling out people for preferential treatment, providing initial and ongoing training, and encour- aging participation.

MEASURES OF FACTORS RELATED TO CAOC

The amount of previous change (alpha = .82) was measured in the same way as on the first survey; however, the time frame was much more recent because of the 21 months that had elapsed between surveys. One’s motiva- tion to keep on trying (alpha = .68) was measured with four items: I person- ally support attempts to make things better around here, I believe in trying to do everything I can to solve problems around here, I would be willing to serve on a task force to help solve problems around here, and I am willing to take on extra duties in order to make improvements around here. Organizational commitment (alpha = .87) was measured using the nine-item short form of a scale developed by Mowday, Porter, and Steers (1982).

The number of grievances was measured by an open-ended question that asked “How many grievances have you filed in the previous 2 years?” The instrumentality perception between performing well on the job and earning more money was also measured with a single item, as is typical (Ilgen, Nebeker & Pritchard, 1981).

CONFIRMATORY FACTOR ANALYSIS FOR CAOC

The appendix shows the 12 items used to measure these three facets: pes- simism, the dispositional attribution, and the situational attribution. Confir- matory factor analysis (CFA) was conducted with the program RAMONA (Browne, Mels, & Cowan, 1994). The advantages of RAMONA are econ- omy of parameter matrices (one matrix contains path coefficients, the other covariances) and simple, direct specification of equations from a path dia- gram. Another benefit is that the estimated standard errors are correct for cor- relation input (cf. Cudeck, 1989). A two-factor structure was expected with items 1 through 8 loading on a CAOC latent variable and items 9 through 12 loading on a situational attribution latent variable. The Root Mean Square Error of Approximation (RMSEA) (Steiger, 1990; Steiger & Lind, 1980), which represents the average error of approximation of the estimation proce- dure, was used to assess the fit of the models. RMSEA would be zero for a model with perfect fit. RMSEA served as the major criterion for model evalu- ation because of the sensitivity of the chi-square test statistic to sample size. Guidelines proposed for this index (Browne, Mels, & Cowan, 1994) are good fit (.00 to .05), moderate fit (.05 to .08), mediocre fit (.08 to .10), and poor fit (.10). A power analysis conducted for all models using the program of

140 GROUP & ORGANIZATION MANAGEMENT

MacCallum, Browne, and Suguwara (1996) indicated power levels above 0.95. This finding suggests that the sample size was appropriate for minimiz- ing Type II errors.

Results of the CFA are presented in Figure 1. The goodness of fit index decreased from an RMSEA of .118 to .085 to .076 as the number of latent variables increased from 1 to 2 to 3. The middle value represents the fit of the hypothesized two-factor model, which is the one shown in Figure 1. Although it is in the mediocre range, it is essentially on the border between

Wanous et al. / CYNICISM 141

Figure 1: RAMONA Program Confirmatory Factor Analysis

moderate and mediocre fit. Given the small width of the confidence interval around this estimate (.076 to .095) and the early phase of research on this con- struct, we decided that the proposed model fits well enough to maintain its plausibility for the present.

Another output of a CFA is estimated interfactor correlations (for the two and three factor models). For the hypothesized two-factor model, the correla- tion between the latent variables (CAOC and situational) was estimated as 0.20, which is low and supports the two-factor model. For the three-factor model, the correlation of 0.83 between the pessimism and dispositional latent variables further supports the two-factor model.

ANTECEDENTS OF CAOC

At this stage of the research process no formal model has been proposed that would suggest a covariance structure model. Thus, a less complex approach is taken that is appropriate for an initial study of possible anteced- ents (Brannick, 1995; Kelloway, 1995). Hierarchical regression analysis is used to assess effects of two basic types of factors, the person and the situa- tion. NA, representing a person factor, is entered at the first step. At the sec- ond step a set of three factors representing one’s experience in this organiza- tion is entered.

Table 1 shows all of the means, SDs, and correlations among the variables used for all of the variables measured for this study. In addition to the ante- cedents and the correlates of CAOC, the Situational Attribution scale is included as a form of discriminant validity. As can be seen, the Situational Attribution is significantly correlated with three of the other nine variables used in this study: CAOC, NA, and organizational commitment. All three of the correlations are less than r = .20.

Table 2 shows the results of the two-step hierarchical regression. The set of three organizational factors was entered after NA because the authors believed cynicism most likely resulted from one’s experiences working in this organization. (Placing this set last in the hierarchy is a conservative approach.)

As can be seen in Table 2, NA accounts for a very small amount of the variance in CAOC (1.8%), whereas the set of three organizational factors adds an additional 12.4% ( p < .01) of explained variance. It is also clear that the majority of the variance in CAOC remains to be explained because the overall adjusted R2 is .142.

142 GROUP & ORGANIZATION MANAGEMENT

CONCURRENT RELATIONSHIPS WITH CAOC

In keeping with the focus on change, two correlates of cynicism are exam- ined. First, CAOC is significantly correlated with employee motivation to keep on trying to make change (r = –.40, p < .01). Second, CAOC is signifi- cantly correlated with the amount of recent previous change (r = –.16, p < .01). Next, CAOC was related to attitudes and behaviors not directly con- cerned with organizational change, but which have been considered impor- tant indicators in organizational research for many years, that is, organiza- tional commitment (r = –.46, p < .01) and grievance filing (r = .13, p < .01).

To demonstrate more clearly the effects of CAOC, the hourly employees were split into high versus low CAOC scores as follows. Those whose scores averaged 3.5 to 5.0 were classified as high (N = 183); those whose scores averaged 1.0 to 2.5 were classified as low (N = 147). Those with scores in the middle of the scale (2.51 to 3.49) were dropped from these analyses (N = 310). It should be noted that those in the middle range represented 48% of those who responded with sufficient data to calculate a CAOC score.

First, there is a very significant difference (t = 8.96, p < .01, omega squared = .21) in the mean scores between the high CAOC group (M = 3.36, SD = .69, N = 167) versus the low group (M = 4.02, SD = .58, N = 156) in terms of their motivation to keep on trying to make change successful. The

Wanous et al. / CYNICISM 143

TABLE 1

Study Variables and Correlations a

(N = 352)

Variable M SD 1 2 3 4 5 6 7 8 9 10

CAOC 2.91 .70 (.86) NA 1.42 .55 .14 (.91) PDM 3.10 1.05 –.23 .03 (.84) SE 3.14 .78 –.27 –.05 .34 (.88) Change 1 2.01 .77 –.17 .12 .06 .04 (.90) KOT 3.83 .62 –.40 .04 .17 .24 .26 (.68) Change 2 2.29 .81 –.16 .14 .01 .08 .36 .20 (.82) OC 3.62 .69 –.46 –.16 .13 .27 .13 .46 .13 (.87) Grievance 0.61 1.64 .13 .01 –.03 –.11 .08 –.04 .08 –.14 — Situation 3.12 .63 .15 .18 –.07 .01 –.02 –.03 .08 –.12 .02 (.61)

NOTE: CAOC = cynicism about organizational change, NA = negative affectivity, PDM = partic- ipation in decision making, SE = supervisory effectiveness, Change 1 = amount of change at Sur- vey 1, KOT = motivation to keep on trying to make change, Change 2 = amount of change at Sur- vey 2, OC = organizational commitment, Grievance = number of grievances filed, and Situation = situational attribution for failed change. a. For this sample size, p < .05 for r = .11, and p < .01 for r = .15.

omega squared effect size from this high versus low comparison (.21) is greater than an effect size based on the squared correlation (–.402 = .16) that was based on all employees. (A recent article by Austin, Boyle, & Lualhati, 1998, advocates omega squared as the preferred effect size estimate.)

The second concurrent relationship examined was with organizational commitment. Because organizations desire high commitment employees, this group was defined as those with OCQ scores averaging 3.5 to 5.0 (the same scale cut-off as used for high vs. low cynicism). Those who have high CAOC scores are much less likely to be high in organizational commitment (37.7%) compared with those with low CAOC scores (80.7%). The organiza- tional commitment mean for all of the high CAOC employees is 3.15 (SD = .84) versus a mean of 4.04 (SD = .60) for all of the low CAOC employees. This difference is significant (t = 10.23, p < .01, omega squared effect size = .27). Again, the omega squared effect size associated with a comparison of the extremes (.27) is greater than that based on the correlation between these two variables based on the total sample (.462 = .21)

The third relationship examined was with the number of grievances filed in the 2 years preceding the second survey. The high CAOC employees filed more grievances (M = 1.42, SD = 4.04) than the low CAOC employees (M = .62, SD = 2.48). This difference was also significant (t = 2.20, p < .05, omega squared effect size = .02). Another way to express this is that 38% of the high CAOC group filed at least one grievance versus 21% of those in the low CAOC group. Although the majority of employees did not file grievances, the rate of grievance filing is almost twice as high for the high versus low CAOC group. In this case, however, the omega squared effect size obtained from the high versus low split is identical to that obtained from the correla- tion based on the overall sample (.132 = .02).

144 GROUP & ORGANIZATION MANAGEMENT

TABLE 2

Antecedents of Cynicism About Organizational Change

Step Variable Adjusted R2 beta t p

1 NA .018 .144 2.85 .01 2 NA .142 .157 3.30 .01

Change 1 –.158 –3.34 .01 PDM –.176 –3.52 .01 SE –.211 –4.22 .01

NOTE: NA = negative affectivity, Change 1 = amount of change at Survey 1, PDM = participa- tion in decision making, and SE = supervisory effectiveness.

The last analysis of possible spillover effects of CAOC to areas beyond organizational change focused on employee perceptions of an important instrumentality perception related to work motivation. Namely, employees were asked to assess the strength of the relationship between performing well on the job and earning more money as a result.

The employees in this organization were paid in one of three ways: (a) hourly, (b) hourly plus incentive, and (c) salaried. By grouping employees (n = 720) into their respective pay categories, one is using an organizational- level variable that is also objective (i.e., not measured by self-report data). To conduct this analysis, the three groups of employees were identified and the perceived instrumentality of good performance (P) for earning more money ($) was obtained from a survey item; that is, P → $. Following this, employ- ees were separated into high, medium, and low groups based on CAOC scale scores. Splitting the sample restricts the range in how CAOC is measured, but it also makes its measurement comparable to that of the pay system, which was also categorical. When comparable measures are used, the result is a fair comparison (Cooper & Richardson, 1986) between measures of the pay sys- tem and CAOC because both are measured similarly (i.e., categorically). Then, a two-way ANOVA was conducted to assess the effects of pay system, level of CAOC, and the interaction between pay system and level of CAOC on the instrumentality perception of P → $. Figure 2 shows a plot of the cell means with their respective sample sizes.

The pay system variable was dummy coded as an a priori ordinal variable (1 = hourly, 2 = hourly plus incentive, 3 = salaried) representing the closeness of the connection between performance and monetary earnings. This coding was done for the following reasons. First, there is no question that the hourly versus hourly plus incentive should be positioned relative to each other as they are in Figure 2. Although the incentive pay was fairly easy to attain, there was a formula relating job performance to monetary incentives. Sec- ond, the salary group was represented as having the strongest P → $ instru- mentality because those in this group could receive merit raises, an annual bonus, and awards, as well as the possibility of promotion, which itself would mean a higher salary. In contrast, those in the other two groups had virtually no chance for awards or for promotion, and their yearly bonuses had been eliminated.

As can be graphically seen in Figure 2, there are two clear main effects for compensation system (F = 51.82, p < .001) and for CAOC level (F = 20.70, p < .001), and for their interaction (F = 2.37, p < .05). Because the sample sizes are fairly large, it made sense to calculate a measure of the effect sizes corresponding to these significance levels. The omega squared effect size

Wanous et al. / CYNICISM 145

estimates for these three are .11 for compensation system, .05 for CAOC, and .01 for the interaction.

DISCUSSION

SUMMARY OF RESULTS AND IMPLICATIONS

The first step in this exploratory research was to develop a reliable mea- sure of cynicism about organizational change, and this was successfully accomplished. Following this, confirmatory factor analysis provided some support for the proposed combination of two components of CAOC:

146 GROUP & ORGANIZATION MANAGEMENT

Figure 2: Effect of Cynicism About Organizational Change and Compensation System on the Pay-for-Performance Instrumentality Perception

pessimism about future change being successful and blaming those responsi- ble for one’s pessimism.

The second research question was a post hoc analysis of some likely ante- cedents of cynicism, and was examined using hierarchical regression with longitudinal data. The organizational factors category accounted for consid- erably more predictable variance than did NA. This lends support to our view that cynicism is more appropriately treated as a learned response, rather than a personality-based predisposition.

One practical implication of these results is that the organizational factors that result in cynicism among employees can be influenced by top manage- ment. In contrast, personality factors cannot be influenced by management, except when they are used as criteria for personnel selection. Specifically, the more employees believe that change has occurred, the more they have partic- ipated in decision making, and the more effective are their supervisors, the less is their cynicism about future organizational change.

The third research question was an examination of some concurrent rela- tionships with cynicism. Cynicism was negatively correlated with the amount of change that had previously occurred in this organization and with the motivation to keep on trying to make change. The relationship of CAOC to these two variables suggests that cynicism may be somewhat self-fulfill- ing. This is why OD practitioners and executives should take existing levels of cynicism into consideration when attempting new change efforts. The higher the preexisting level of cynicism about organizational change, the more change agents need to confront and discuss previous failures before moving ahead.

Going beyond the issue of organizational change, however, CAOC was significantly related to decreased organizational commitment and increased grievance filing. Reviews of research concerning organizational commit- ment have already detailed its effects on such costly employee behaviors as turnover, performance, and attendance at work (Cohen, 1993; Mathieu & Zajac, 1990; Tett & Meyer, 1993). With respect to grievances, more employ- ees did not file grievances than did. However, the almost two-to-one differ- ence in grievance filing rates between employees from a high level of CAOC compared to a low level suggests CAOC is another employee attitude vari- able that is related to employee behavior at work.

Finally, the spillover of CAOC’s effects into pay-for-performance per- ceptions shows possible damage to employee motivation. When employees are grouped according to the type of compensation system, this means that those within a group are subject to essentially the same set of factors related to their performance. However, high levels of cynicism appear to weaken the perceived pay-for-performance link even within a compensation group where

Wanous et al. / CYNICISM 147

the objective conditions should be similar. In addition, the stronger the P → $ link, the greater is the effect of CAOC on it. Thus, the salary-paid group is most affected, followed by the hourly-plus-incentive-paid group. For the straight hourly paid group, who had the weakest link of all, there was rela- tively little effect.

STRENGTHS AND WEAKNESSES

The primary weakness of this research is that cynicism was measured only on the second questionnaire as a consequence of the action-research method. This made the analysis of antecedents less comprehensive than if the CAOC scale had been in place at the time of the first survey. Specifically, measure- ment of the organizational antecedents approximates what we think are the most important causes of cynicism (i.e., failed change that is attributed to unmotivated and/or incompetent leadership). A clear historical documenta- tion of such events (with data from multiple sources) would complement sur- vey data.

This topic emerged from our experience with this organization; it is grounded research. Thus, it does not fit the textbook ideal for initiating research with an already existing theory in need of empirical examination. Rather, it was interviews and conversations with employees, as well as our 3-year relationship that led to the development of a specific form of cynicism that concerned organizational change. This approach to research is consid- ered valid and useful, particularly at the initial development stage of a new research area, as is the case here (Blanck & Turner, 1987; Susman & Evered, 1978).

Some of the strengths of this exploratory study include the fact that some of the data are longitudinal with a long interval between surveys. This makes for a very conservative test of antecedent factors because most relationships studied in organizational behavior tend to weaken over time (Hulin, Henry, & Noon, 1990).

Second, the focus on the individual employee as an important component of the organizational change process is a strength because too little attention has been paid to employees in the past.

DIRECTIONS FOR FUTURE RESEARCH

Future research should continue to examine both the conceptual and oper- ational definitions of cynicism. As a relatively new construct, its measure- ment may need refinement. For example, these data tend to support the combination of pessimism with the dispositional attribution rather than keep- ing them separate; however, this should continue to be examined in future

148 GROUP & ORGANIZATION MANAGEMENT

research. Furthermore, the stability of CAOC over time should also be exam- ined, and data from different types of organizations might prove useful. For example, if several organizations known to have differing levels of success with organizational change (and the likely cynicism associated with those that failed) were studied, the sensitivity of our CAOC measure to detect these interorganizational differences could be assessed.

Future research into other personality factors, such as the Big Five of per- sonality, also seems warranted. Although NA explained very little of the variance in CAOC, it is possible that other factors might. It is interesting to note that NA was measured concurrently with CAOC, whereas the three organizational factors were measured 21 months earlier. From a strictly methods viewpoint, this should have provided an advantage to NA as an explanation, but it did not. In some ways this increases the confidence in our conclusion that CAOC is more likely learned on the job rather than being the result of a personal predisposition.

Finally, the consequences of cynicism must be studied with longitudinal data. Delineating the full range of effects and their respective strengths is one way to measure the practical significance of cynicism about organizational change. Imbedded in the analysis of consequences must be a careful exami- nation of possible self-fulfilling effects when cynicism is high. Continued investigation into its spillover into areas unrelated to change also seems desirable.

GENERAL IMPLICATIONS FOR MANAGEMENT

Management can influence the level of cynicism by addressing its two components, pessimism and dispositional attributions. With respect to pessi- mism, one strategy is to make sure that all successful changes are clearly pub- licized. No matter how small the change, if it is in the direction intended by management, then it should be communicated. Karl Weick (1984) wrote about the value of small wins in surmounting social problems that often seem enormous and intractable. We think that organizational change, particularly in large hidebound organizations such as the one studied here, falls in the same category. Accordingly, management should not be embarrassed that they can only hit singles instead of home runs, which is the small wins notion Weick advocated.

Managing the attributions made about the lack of change, or its slow pace, is the second way to manage levels of cynicism. The key here is again infor- mation so that employees will make situational rather than dispositional attri- butions. This is the tougher of the two avenues because years of research on attribution theory show that the tendency is to blame others rather than the

Wanous et al. / CYNICISM 149

situation. Nonetheless, there are some actions that can be taken. First, the more employees are involved directly in the change process itself, the less they will make dispositional attributions because there is no they to blame. Second, the less employees are surprised by management actions and the more they understand the reasons for the actions, the more they will see things from the management perspective. This, too, will reduce the tendency to make dispositional attributions like blaming selfish or incompetent man- agement. Third, past failures need to be fully explained rather than ignored. It may be true that the past cannot be undone, but it can have a strong influence on the present and future through the attributions made by employees. If past failure was attributable to management mistakes, they should be acknowl- edged to preserve (or restore) credibility.

Finally, it is important that management make serious efforts to see the sit- uation from the lower level employees’ perspective to understand why they tend to blame management. A good example of this was observed during our study. Top management decided to close the executive dining room to encourage all managers to spend more time with the hourly employees. How- ever, managers tended to sit together in the cafeteria, which caused the hourly employees to attribute an attitude of superiority on the part of managers. Managers were completely unaware of this. They thought that they were just sitting with the people they knew best (other managers) and that their mere presence in the cafeteria was the correct signal to be sending to the rank-and-file.

APPENDIX

Pessimism

1. Most of the programs that are supposed to solve problems around here will not do much good.

2. Attempts to make things better around here will not produce good results. 3. Suggestions on how to solve problems will not produce much real change. 4. Plans for future improvement will not amount to much.

Dispositional Attribution

5. The people responsible for solving problems around here do not try hard enough to solve them.

6. The people responsible for making things better around here do not care enough about their jobs.

7. The people responsible for making improvements do not know enough about what they are doing.

150 GROUP & ORGANIZATION MANAGEMENT

8. The people responsible for making changes around here do not have the skills needed to do their jobs.

Situational Attribution

9. The people responsible for fixing problems around here cannot really be blamed if things do not improve.

10. The people responsible for solving problems around here are overloaded with too many job responsibilities.

11. The people responsible for fixing problems around here do not have the resources they need to get the job done.

12. The people responsible for making changes around here do not get the coop- eration they need from others.

REFERENCES

Adams, S. (1996). The Dilbert principle. New York: Harper Business. Aktouf, O. (1992). Management and theories of organizations in the 1990s: Toward an initial

radical humanism? Academy of Management Review, 17, 407-431. Andersson, L. M. (1996). Employee cynicism: An examination using a contract violation frame-

work. Human Relations, 49, 1395-1418. Andersson, L. M., & Bateman, T. S. (1997). Cynicism in the workplace: Some causes and

effects. Journal of Organizational Behavior, 18, 449-469. Austin, J. T., Boyle, K. A., & Lualhati, J. C. (1998). Statistical conclusion validity for organiza-

tional science researchers: A review. Organizational Research Methods, 1, 164-208. Blanck, P. D., & Turner, A. N. (1987). Gestalt research: Clinical-field-research approaches to

studying organizations. In J.W. Lorsch (Ed.), Handbook of organizational behavior (pp. 109- 125). Englewood Cliffs, NJ: Prentice-Hall.

Brannick, M. T. (1995). Critical comments on applying covariance structure modeling. Journal of Organizational Behavior, 16, 201-213.

Browne, M. W., Mels, G., & Cowan, M. (1994). Path analysis: RAMONA. SYSTAT for DOS: Advanced applications (Ver. 6; pp. 167-224). Evanston, IL: SYSTAT.

Cohen, A. (1993). Organizational commitment and turnover: A meta-analysis. Academy of Management Journal, 36, 1140-1157.

Cooper, W. H., & Richardson, A. J. (1986). Unfair comparisons. Journal of Applied Psychology, 71, 179-184.

Cudeck, R. W. (1989). Analysis of correlation matrices in covariance structure analysis. Psycho- logical Bulletin, 105, 317-327.

Friedlander, F., & Brown, L. D. (1974). Organizational development. Annual Review of Psy- chology, 25, 313-341.

Guzzo, R. A., Jette, R. D., & Katzell, R. A. (1985). The effects of psychologically based inter- vention programs on productivity: A meta-analysis. Personnel Psychology, 38, 275-291.

Hammer, M., & Champy, J. (1993). Re-engineering the corporation: A manifesto for business revolution. New York: Harper.

Wanous et al. / CYNICISM 151

House, R. J., Rousseau, D. M., & Thomas-Hunt, M. (1995). The meso paradigm: A framework for the integration of micro and macro organizational behavior. In L. L. Cummings and B. M. Staw (Eds.), Research in organizational behavior (Vol. 17, pp. 71-114). Greenwich, CT: JAI.

Hulin, C. L., Henry, R., & Noon, S. (1990). Adding a dimension: Time as a factor in the generalizability of predictive relations. Psychological Bulletin, 107, 328-340.

Ilgen, D. R., Nebeker, D. M., & Pritchard, R. D.(1981). Expectancy theory measures: An empiri- cal comparison in an experimental setting. Organizational Behavior and Human Perfor- mance, 28, 189-223.

Jones, E. E., & Davis, K. E. (1965). From acts to dispositions: The attribution process in person perception. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 2, pp. 219-226). New York: Academic Press.

Juran, J. M., & Gryna, F. M. (1980). Quality planning and analysis. New York: McGraw-Hill. Kanter, D. L., & Mirvis, P. H. (1989). The cynical Americans: Living and working in an age of

discontent and disillusion. San Francisco: Jossey-Bass. Kelloway, E. K. (1995). Structural equation modeling in perspective. Journal of Organizational

Behavior, 16, 215-224. Likert, R. (1967). New patterns of management. New York: McGraw-Hill. MacCallum, R. A., Browne, M. W., & Sugawara, H. (1996). Power analysis and determination

of sample size for covariance structure modeling. Psychological Methods, 1, 130-149. Mann, F. C. (1957). Studying and creating change: A means to understanding social organiza-

tion. In C. M. Arensberg (ed.), Research in industrial human relation (pp. 146-167). New York: Harper.

Mathieu, J. E., & Zajac, D. M. (1990). A review and meta-analysis of the antecedents, correlates, and consequences of organizational commitment. Psychological Bulletin, 108, 171-194.

Mowday, R. T., Porter, L. W., & Steers, R. M. (1982). Employee-organization linkages: The psychology of commitment, absenteeism, and turnover. New York: Academic Press.

Neuman, G. A., Edwards, J. E., & Raju, N. S. (1989). Organizational development interven- tions: A meta-analysis of their effects on satisfaction and other attitudes. Personnel Psychol- ogy, 42, 461-483.

Niederhoffer, A. (1967). Behind the shield: The police in urban society. Garden City, NJ: Anchor.

Peters, T., & Austin, N. (1985). A passion for excellence. New York: Harper & Row. Peters, T., & Waterman, R. (1982). In search of excellence. New York: Harper & Row. Pope, K. S., Butcher, J. N., & Seelen, J. (1993). The MMPI, MMPI-2 & MMPI-A in court. Wash-

ington, DC: American Psychological Association. Porras, J. I., & Berg, P. O. (1978). The impact of organizational development. Academy of Man-

agement Review, 3, 249-266. Porras, J. I., & Robertson, P. J. (1992). In M. D. Dunnette & L. M. Hough (Eds.), The handbook

of industrial and organizational psychology (Vol. 3, pp. 719-822). Palo Alto, CA: Con- sulting Psychologists Press.

Reichers, A. E., & Schneider, B. (1990). Climate and culture: An evolution of constructs. In B. Schneider (Ed.), Organizational climate and culture (pp. 5-39). San Francisco: Jossey- Bass.

Reichers, A. E., Wanous, J. P., & Austin, J. T. 1997. Understanding and managing cynicism about organizational change. Academy of Management Executive, 11, 48-59.

Schein, E. H. (1985). Organizational culture and leadership: A dynamic view. San Francisco: Jossey-Bass.

152 GROUP & ORGANIZATION MANAGEMENT

Shapiro, B. (1997, October). Dilbert hits close to home for some I/O psychologists. APA Moni- tor, 34.

Steiger, J. H. (1990). Structural model evaluation and modification: An interval estimation approach. Multivariate Behavioral Research, 25, 173-180.

Steiger, J. H., & Lind, J. M. (1980). Statistically-based tests for the number of common factors. Paper presented at the annual meeting of the Psychometric Society, Iowa City, IA.

Susman, G. I., & Evered, R. D. (1978). An assessment of the scientific merits of action research. Administrative Science Quarterly, 23, 582-603.

Tett, R. P., & Meyer, J. P. (1993). Job satisfaction, organizational commitment, turnover inten- tion, and turnover: Path analyses based on meta-analytic findings. Personnel Psychology, 46, 259-293.

Tjosvold, D. W., & Tjosvold, M. M. (1991). Leading the team organization. Lexington, MA: Lexington Books.

Vroom, V. H. (1964). Work and motivation. New York: Wiley. Wanous, J. P., Reichers, A. E., & Austin, J. T. (1994). Organizational cynicism: An initial study.

Academy of management best papers proceedings, 269-273. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of a brief measure

of positive and negative affect: The PANAS scales. Journal of Personality and Social Psy- chology, 54, 1063-1070.

Weick, K. E. (1984). Small wins: Redefining the scale of social problems. American Psycholo- gist, 39, 40-49.

John P. Wanous is Professor of Management and Human Resources in the Fisher Col- lege of Business and joint professor of psychology at The Ohio State University. He received his Ph.D. from Yale University in 1972.

Arnon E. Reichers is Associate Professor of Management and Human Resources in the Fisher College of Business at The Ohio State University. She received her Ph.D. from Michigan State University in 1983.

James T. Austin is a research specialist at the Center on Education and Training for Employment at The Ohio State University. He received his Ph.D. from Virginia Tech in 1987.

Wanous et al. / CYNICISM 153