DOT/FAA/AM-92/17 Organizational Politics,

Office of Aviation Medicine Participation in Decision-Washington, D.C. 20591 M aking, and Job Satisfaction

AD-A251 878 D TI CL.A. Witt JNi17?99

Civil Aeromedical Institute CFederal Aviation AdministrationOklahoma City, Oklahoma 73125

April 1992

Final Report

This document is available to the public

Nthrough the National Technical Information

Service, Springfield, Virginia 22161.

0U.S. Departmentof TransportationFederal AviationAdministration

92 6 1 i

NOTICE

This document is disseminated under the sponsorship ofthe U.S. Department of Transportation in the interest of

information exchange. The United States Governmentassumes no liability for the contents or use thereof.

Technical Report Documentation Page1. Report No. 2. Government Accession No. 3. Recipient s Catalog No.

DOT/FAA/AM-92/17

4. Title and Subtitle 5. Report Date April 1992ORGANIZATIONAL POLITICS, PARTICIPATION IN DECISION-

MAKING, AND JOB SATISFACTION 6. Performing Organization Code

7. Auhrs) 8. Performing Organization Reort No.

L. A. Witt, Ph.D.9. Performing Organization Name and Address 10. Work Unit No. (TRAIS)

FAA Civil Aeromedical Institute

P. 0. Box 25082Oklahoma City, OK 7312513. Type of Report and Period Covered

12. SpSsnsoring Agency Name and AddressOffice of Aviation MedicineFederal Aviation Administration800 Independence Avenue, S.W. 14. Sponsorng Agency CodeWashington, D.C. 20591

15. Supplementary Notes

This work was performed under task AM-C-92-HRR-122.

16. Abstract

The study tested two hypotheses: (a) that organizational politics as measured by the Kacmar and Ferris (1991)Perceptions of Organizational Politics Scale would be negatively related to feelings of job satisfaction; and (b)that participation in decision-making (PDM) would moderate that relationship. In line with concerns fordispositional affect as a contributor to method variance and the possibility that biodata may explain some of theeffects of organizational politics and PDM on job satisfaction, dispositional affect and biodata variables wereincluded in the analyses. Hierarchical moderated multiple regression analyses conducted on data collected from1,083 Federal employees confirmed the hypotheses.

17. Key Words 18. Distribution Statement

Organizational Politics Document is available to theParticipation in Decision-making public through the NationalJob Satisfaction Technical Information Service,

Springfield, Virginia 22161

19. Security Classif. (of this report) 20. Security Clossil. f~i this page) 21. No. of Pages 22. Price

Undla iicd Unclassified 15

Form DOT F 1700.7 (8-72) Reproduction of completed page outhorized

ORGANIZATIONAL POLITICS, PARTICIPATION INDECISION-MAKING, AND JOB SATISFACTION

Experiential reality suggests that political quences of organizational politics. For example,behavior is a pervasive force in organizational Frost and Hayes (1977, p. 8) wrote that politicallife. As noted by several researchers (Farrell & behaviors are:Peterson, 1982; Mintzberg, 1985; Narayanan &Fahey, 1982; Porter, Allen, & Angle, 1983; "... the activities of organization mem-Vredenburgh & Maurer, 1984), most organiza- bers... when they use resources to enhance ortions possess political norms as part of their protect their share of an exchange... in waysinformal structure. Discussed under the concep- which would be resisted, or ways in whichtual rubric of "organizational politics," political the impact would be resisted, if recognizedbehavior in organizations has received increasing by the other party(ies) to the exchange."attention (e.g., Cavanaugh, Moberg, &Velasquez, 1981; Macher, 1986) since Gandz Similarly, Vredenburgh and Maurer (1984, p.and Murray (1980) reported finding very little 50) suggested that organizational politics:empirical research in a literature review ofworkplace politics. As noted by Kacmar and (a) is undertaken by individuals or interestFerris (1991), organizational political behavior groups to influence directly or indirectlyhas been described as both omnipresent and target individuals, roles, or groups toward theelusive. They suggested that its elusiveness is actor's personal goals, generally in oppositionreflective of an inadequate conceptual understand- to others' goals, (b) consists of goals oring of organizational politics. Indeed, while we means either not positively sanctioned by anmay "know it when we see it," organizational organization's formal design or positivelypolitics has not been well defined, sanctioned by unofficial norms, and (c) is

objective and subjective in nature, involvingGandz and Murray (1980) presented two real organization events as well as perceptual

categories of conceptual work on organizational attributions.politics. In one, writers have defined organiza-tional politics as an inherent process associated Drory and Romm (1990) suggested that thewith the use of power revolving around most domain of organizational politics should includepolicy decisions (Pfeffer & Salancik, 1974) or any organizationally-controlled event desirablespecific policy decisions, such as resource alloca- for the individual. Such an approach would maketion (Wildavsky, 1964), or executive compensa- difficult operationalization of organizationaltion (Ungson & Steers, 1984). In the second politics. In contrast, Kacmar and Ferris (1991)category, writers have discussed organizational argued that organizational politics may best bepolitics in terms of self-serving behaviors often conceived as having three elements, namelyassociated with organizational ineffectiveness general political behavior, going along to get(Ashforth & Lee, 1990; Cobb, 1984; Gray & ahead, and pay and promotion.Ariss, 1985; Martin & Sims, 1974; Mayes &Allen, 1977; Pettigrew, 1973; Robbins, 1976, Following Ferris, Russ, and Fandt's (1989)1983; Schein, 1977). With respect to the latter call for work to assess both the antecedents andcategory, Batten and Schwab (1965, p. 13) wrote consequences of political behavior in organiza-that individuals "engage in company politics tions, the present study examined the relationship ,__J -because they believe that they can best achieve between perceptions of organizational politics andwhat they want in a devious, indirect, and under- job satisfaction. Specifically, it was hypothesizedhanded way." Jones (1987) entered a third that individuals working amidst high levels ofcategory arguing that definitions of organizational organizational politics would feel less satisfied :j, £ tpolitics should include mutual caring, trust, with their jobs (i.e., that organizational politics -

tolerance, and sensitivity to other. However, and jo!" satisfaution would be negatively related).most definitions focus on the negative conse- This hypothesis is consistent with arguments

Avnllablil

Dist

*[

(Batten & Schwab, 1965; Drory & Romm, 1990; differing definitions of PDM have caused pro-DuBrin, 1988) and empirical findings (Luthans, blems for PDM researchers. Thibaut and WalkerHodgetts, & Rosenkrantz, 1988) suggesting that (1975) conceptualized two forms of participation:political behavior may be dysfunctional for (a) choice or decision control, where the par-organizations. Indeed, it may be difficult to be ticipant has some control over the outcome; andhappy about job situations in which some in- (b) voice or process control, where the par-dividuals are behaving in an "underhanded" or ticipant articulates his/her interest to the decision-"back-door" fashion to meet their objectives at maker. Voice may include influence over defin-others' expense. This rather straightforward ing the problem, gathering information bearingrelationship may be intuitively obvious but on the decision, and identifying alternatives, butsimultaneously impervious to intervention. In not making the decision. Cohen (1985) arguedother words, there may be limitations on the that to the extent subordinates can express opin-extent to which any individual employee and ions to the supervisor, they have a "voice."his/her supervisor can directly influence andreduce organizational politics that may be or- It is suggested here that both voice and choiceganization-wide or traditionally inherent in the reflect PDM. When subordinates make decisionsculture of an organization or organizational without discussions with the supervisor, they aresubsystem. acting autonomously and essentially have choice,

but they are making decisions without the super-The present study also examined participation visor's voice. This reflects a low level of PDM.

in decision-making as a possible moderator of the A low level of PDM is also reflected in situa-organizational politics-job satisfaction relationship tions where decisions are made by the supervisor-- a moderator in the practical sense through without discussion, as the subordinate has neitherwhich individual employees and supervisors may choice nor voice. Two moderate levels of PDMbe able to reduce the negative effects of or- occur when: (a) the supervisor usually makes theganizational politics on job satisfaction, decisions following discussion, as both have

voice but the subordinate usually has littleParticipation in Decision-Making choice; and (b) the subordinate generally makes

the decision after discussion, as both have choiceThe outcomes of increased worker participa- and voice (the supervisor has choice, because

tion in decision-making (PDM) have been of he/she permitted the subordinate to make thetheoretical and practical interest for several decision). However, it may be likely that theyears. Argyris (1964) argued that workers will supervisor did not engage the issue if the subor-manifest responsible adult behaviors only when dinate usually makes the decision, even if thetheir managers realize that they want to be supervisor has globally assigned permission toinvolved in making decisions. Indeed, "psycholo- the subordinate to make decisions. When super-gical folklore" (Greenberg & Folger, 1983, p. visors and subordinates discuss issues and reach235) suggests that PDM will generally have a decision based on consensus (i.e., a collabora-positive outcomes. Although empirical investiga- tion to find resolution), both have optimal voicetions of the effects of PDM have yielded mixed and choice, which results in a high level ofresults (cf. Cotton, Vollrath, Froggatt, Lengnick- PDM.Hall, & Jennings, 1988), a number of positiveoutcomes have been identified, including more Based on this conceptual definition of low,favorable job attitudes (Argyris, 1964), improved moderate, and high levels of PDM, it was hypo-employee health (Jackson, 1983), increased thesized that PDM would moderate the relation-organization information-processing capabilities ship between organizational politics and job(Castrogiovanni & Macy, 1990), and a better satisfaction. In other words, employees whoundersiJing of work tasks (Niehoff, Enz, & actively engage in discussions about importantGrover, 1990). issues with their supervisors and who, as a

result, develop a consensus to resolve the issuesDachler (1978) suggested that inadequate and might be able to reduce the negative conse-

2

quences of organizational politics on job satisfac- may share a single information processing andtion more so than employees who make the dispositional basis. Implicit in this argument isdecisions or whose supervisors make decisions the notion that individual disposition influencesfor them. Indeed, such a joint decision provides the relationships of interest. A problem in assess-both joint ownership and understanding, which ing method variance has involved identifying themay generate feelings of security and protection disposition that underlies job attitudes and thenin a political environment, measuring the operaticnalized disposition. Recent

advances in the affect literature provide a meansThis notion is consistent with findings in both of approaching a resolution to this problem.

the PDM and organizational politics literatures.Baumgartel (1957) reported that employee perfor- Factor analytic studies of self-reported moodmance, job satisfaction, and positive attitudes have indicated that negative affect (NA) andtoward the supervisor were higher among emp- positive affect (PA) are the two dominant andloyees whose supervisors engaged in participa- relatively independent dimensions (Diener &tion, were lowest among employees whose Emmons, 1985). Watson and Clark (1984)supervisors directed activities with no discussion, argued that the tendency to experience negativeand were intermediate among employees whose or positive affect reflects a stable, ongoingsupervisors who took a laissez faire (subordinate disposition. They stated (p. 483) that "high-NAdecides) approach. An argument in the organiza- individuals are more likely to report distress,tional politics literature (Cobb, 1986a, 1986b; discomfort, and dissatisfaction over time andMarch & Simon, 1958; Tushman, 1977) suggests regardless of the situation, even in the absence ofthat uncertainty is central in the development of any overt or objective source of stress." Thesepolitical behaviors. Empirically identified uncer- individuals focus on their disappointments,tainty-related antecedents of organizational poli- mistakes, and shortcomings, generally taking atics include: (a) a disparity between formal negative perspective of life experiences. Low-authority and actual influence (Dalton, 1959), (b) NA persons, on the other hand, tend to be moreincomplete knowledge about cause and effect self-secure, calm, and satisfied, adopting a more(Tushman, 1977), (c) the absence of objective resilient approach to setbacks and focusing lessperformance standards (Pfeffer, 1978), (d) re- on daily frustrations. Watson and Clark (1984)wards based on non-performance criteria stressed that negative affect represents subjective(Robbins, 1979), (e) decisions about important differences in temperament, mood, and cognitiveissues made in secrecy (Pfeffer, 1978), (f) over- orientation rather than an index of psychologicalall uncertainty (Madison, et al., 1980), (g) health. Individuals high in PA are characterizedconcerns for career success (Luthans, Hodgetts, by high energy, full concentration, and pleasura-& Rosenkrantz, 1988), (h) career stage, con- ble engagement, whereas low-PA individualsfidence, and maturity (DuBrin, 1978, 1988; typically experience sadness and lethargy. Pre-Schein, 1978), and (i) a discussed disagreement vious research (e.g., Levin & Stokes, 1989;over outcomes ('ushman, 1977). High levels of Witt, 1991) has demonstrated the utility of affectPDM are likely to reduce uncertainty between the as a predictor of job attitudes, suggesting thatsubordinate and supervisor, as the subordinate dispositional affect may confound relationshipswill have a consensually-developed decision and between self-reported job attitudes and othersome anticipation of the supervisor's responses to variables of interest.outcomes.

The Present StudyAffect and the Problem of Method Variance

Following the literature cited above, theThe notion of method variance (Podsakoff & primary purpose of the present study was to test

Organ, 1986) suggests that collecting data with a the hypotheses that perceptions of organizationalsingle type of measurement leads to an artificially politics would be negatively related to job satis-inflated relationship between the constructs faction and that participation in decision-makingmeasured, because the ratings for each individual would moderate that relationship. Given the often

3

assumed, and sometimes demonstrated, impor- levels 11 to 13 (43.3%), and grade levels 14+tance of biodata in the prediction of job satisfac- (16.1%).tion (cf. Brush, Moch, & Pooyan, 1987), biogra-phical variables were entered into the equation. MeasuresIn line with concerns for dispositional affect as acontributor to method variance, PA and NA Participation in decision-making (a = .90, Mscores were included in the analyses. By entering = 10.77, SD = 4.21) was examined by 6 itemsdispositional affect into the equation before asking respondents to indicate "what best de-organizational politics and PDM, the incremental scribes the way you and/or your immediatevariance contributed by affect beyond that al- supervisor make decisions about:" (a) "yourready contributed by biodata could be assessed. performance appraisal review," (b) "mostSimilarly, organizational politics, PDM, and their things," (c) "communicating information outsidecross-product term (to assess the interaction) your work unit," (d) "overall work activities ofwere added after the biodata and affect variables your work unit," (e) "what your job duties willin order to determine the amount of incremental be," and (f) "how or when you will get yourvariance over-and-above the variance contributed work done." These items were presented on a 5-by both biodata and dispositional affect. point Likert-type scale (1 = "we discuss things

a great deal and come to a decision based onMETHOD consensus regarding the issue at hand;" 2 = "we

discuss things a great deal, and his/her decisionSubjects and Procedure is usually adopted;" 3 = "we discuss things a

great deal, and my deci-sion is usually adopted;"Questionnaires were mailed to the 2,103 4 = "we don't discuss things very much, and

employees of the Mike Monroney Aeronautical his/her decision is usually adopted;" and 5 =Center. One thousand eighty-three (53.8%) "we don't discuss things very much, and I makeemployees (61.5% males and 38.5% females) most of the decisions"). Items 4 and 5 werereturned completed questionnaires by mail. recoded as a 1, responses 2 and 3 were recodedIncluded were biodata items assessing racial as a 2, and response 1 was recoded as a 3. Itemgroup, age, tenure, supervisory status, education, means and standard deviations are presented inand pay grade. About 20% of the employees Table 1. A sum of the responses was computedclassified themselves as members of a minority to provide a total PDM score. High scores reflectracial group. Responses to the age item were as perceptions of greater PDM.follows: ages 17 to 29 (7.9%), 30 to 39(23.0%), 40 to 49 (39.0%), 50 to 59 (26.6%) Job satisfaction was measured by the vali-and 60+ years (3.5%). Responses to the item dated (McNichols, Stahl, & Manley, 1978) 4-assessing years in the organization at the current item Hoppock (1935) job satisfaction scale (a =site were: less than one year (12.1%), 1 to 3 .79, M = 14.22, SD = 2.71). High scoresyears (25.9%), 4 to 10 years (26.0%), 11 to 15 reflect feelings of high job satisfaction.years (14.6%), and 16 or more years of service(21.4%). Supervisory status reported by the Organizational politics was measured by arespondents was as follows: non-supervisor revised version of the 12-item Kacmar and Ferris(86.6%), supervisor (9.5%), and manager (191) Perceptions of Organizational Politics(3.9%). The sample was relatively well educated: Scale (POPS; t = .93, M = 38.25, SD =12 years or less of formal education (12.9%), 13 11.51). Response options were presented on a 5-to 15 years (51.7%), 16 years (24.5%), and 17 point, Likert-type scale (1 = strongly disagree;or more years (10.9%). Salary within the Federal 2 = inclined to disagree; 3 = neither disagreeGovernment is based upon grade level. Re- nor agree; 4 = inclined to agree; and 5 =sponses to the grade levels item were: grade strongly agree). For this study, the wording waslevels 1 to 4 (12.0%), grade levels 5 to 7 changed slightly for two of the items -- from "no(19.2%), grade levels 8 to 10 (9.4%), grade place for yes men" to "it is safer to agree with

4

managers than to say what you think is right," trait (cf. Watson, & Clark, 1984; Watson et al.,and from "pay and promotion policies are not 1988). The PANAS consists of a 10-item Positivepolitically applied" to "pay and promotion deci- Affect Scale (a = .89, M = 36.04, SD = 7.12)sions are consistent with policies." This scale and a 10-item Negative Affect Scale (a = .86,was selected, because it includes items assessing M = 16.22, SD = 5.7), respectively. It mea-both proactive promotion of self-interests (e.g., sures to what extent the respondent generally"Generally, people around here build themselves perceives him/herself to experience certainup by tearing others down") as well as defensive feelings. Results reported by Watson et al.behaviors (e.g., "It is safer to agree with mana- (1988) suggest that these measures assess disposi-gers than to say what you think is right") as tional affect that may be independent of situation-discussed by Ashforth and Lee (1990). More- al influences in which the measure is taken.over, it was validated on employee samples Response options were presented on a 5-point,rather than scales developed by other researchers Likert-type scale (1 = very slightly or not at all;using university students holding jobs (Drory & 2 = a little; 3 = moderately; 4 = quite a bit;Romm, 1988). Although Kacmar and Ferris and 5 = extremely). High scores indicate high(1991) presented a three-dimension model of PA or NA, respectively.organizational politics, work by Nye and Witt(1992) suggested a unidimensional model of the TABLE 2. Organizational Politics Items,POPS, and this one-factor approach was adopted Means, and Standard Deviationshere. Item means and standard deviations arepresented in Table 2. High scores reflect percep-tions of greater levels of organizational politics in I. It is safer to agree with managers than to say what

the organization. you think is right. (M = 3.27, SD = 1.29)

2. There is one group of people (i.e., in an office or

TABLE 1. PDM Items, Means, and work unit) that usually gets their way here. (M

Standard Deviations = 3.63, SD = 1.24)

3. There is one very influential group of workers herethat people know not to cross or get in their way.(Al = 3.39, SD -- 1.30)

I. To what extent do you and/or your immediate super-

visor make decisions about your performance apprai- 4. Overall, policy changes are designed to help only asal review? (M = 1.63, SD = .81) few people here. (M = 3.95, SD = 1.23)

2. To what extent do you and/or your immediate super- 5. Generally, people around here build themselves upvisor make decisions about most things? (M = 1.94, by tearing others down. (M = 3.87, SD = 1.27)SD = .90)

6. Favoritism (not merit) gets people ahead here.3. To what extent do you and/or your immediate super- (M = 3.35, SD = 1.37)

visor make decisions about communicating informa-tion outside your work unit? (M = 1.84, SD = .87) 7. Around here, it is not smart to speak up, because

there is retaliation against people who do. (M =4. To what extent do you and/or your immediate super- 3.97, SD = 1.32)

visor make decisions about overall work activities ofyour work unit? (M = 1.85, SD = .87) 8. Promotions go to top performers. (M = 3.19, SD

= 1.25) IReverse-scoredi5. To what extent do you and/or your immediate super-

visor make decisions about what your job duties will 9. Rewards come to hard workers. (M = 3.05, SD =be? (M = 1.80, SD = .86) 1.27) [Reverse-scored]

6. To what extent do you and/or your immediate super- 10. Pay and promotion decisions are based solely onvisor make decisions about how or when you will get merit. (- = 3.49, SD = 1.19) [Reverse-acored)your work done? (M = 1.70, SD = .85)

i1. Pay and promotion decisions are consistent with

Positive and negative affect were assessed by policies. (M = 3.04, SD = 1.20) Reverse-scored)

the Watson, Clark, and Tellegen (1988) PANAS 12. People here are encouraged to speak out. (M =

measure based upon their earlier research indicat- 3.00, SD = 1.22) Reverse-scoredl

ing that the experience of positive or negativeaffect represents a stable, ongoing dispositional

5

RESULTS tional politics would be less salient to the deve-lopment of job satisfaction among individuals

The intercorrelation matrix is presented in perceiving higher levels of PDM, regressionTable 3. As shown there, the biodata variables analyses were run without the biodata and affect(sex, minority vs. non-minority status, super- scores. By removing them from the equation, itvisory status, education, pay grade levels, age, was possible to citanly plot the interaction.and time in organization at current site) were Following the approach advocated by Stone andweakly related to perceptions of PDM, percep- Hollenbeck (1989), a graph was derived bytions of organizational politics, and job satisfac- plotting three slopes, one for a low PDM scoretion. However, both PDM (r = .37, p < .01) (one standard deviation below the mean: f =and organizational politics scores (r = -.62, p < 6.59), one for an average PDM score (at the.01) were significantly related to job satisfaction mean: f = 10.81), and one for a high PDMscores. PDM and organizational politics scores score (one standard deviation above the mean: fwere significantly related (r = -.49, p < .01). 15.03).As shown in Table 3, PA and NA scores werecorrelated with PDM, organizational politics, and As shown in Figure 1, among employeesjob satisfaction scores. perceiving higher levels of organizational poli-

tics, job satisfaction was highest among thoseResults of the regression analyses are pre- engaged in higher levels of PDM. In contrast,

sented in Table 4. Biodata variables were entered among employees perceiving little organizationalas a block into a regression equation predicting politics, the moderating effect of PDM wasjob satisfaction scores. As shown in Table 4, minimal, as most of these employees expressedthey accounted for a modest amount of variance high levels of job satisfaction.(R' = .04, f = 5.1, p < .01). Positive andnegative affect scores were then entered as one DISCUSSIONblock into the equation. They added significantvariance over-and-above that contributed by Several caveats should be emphasized beforebiodata (A R2 = .27, f = 179.6, p < .01). The discussing the results. First, the subjects mostbiodata and dispositional variables accounted for likely responded to the survey in one sitting;a modest amount of the variance in job satisfac- thus, these data may be subject to commontion scores (total adjusted R = .30). PDM and method variance. The inclusion of PA and NAorganizational politics scores were then put into scores in the equations permitted some assess-the equation, and they added significant in- ment of the contributions of PDM and organiza-cremental variance (A )e = .17, f = 313.2, p tional politics above the variance contributed by< .01). dispositional affect, but other components of

method variance (e.g., cognitive orientation)Hierarchical moderated multiple regression were not assessed. Second, the civilian govern-

analyses (Cohen & Cohen, 1983) were used to ment employees sampled in the present studydetermine whether or not PDM scores moder- may not be representative, and replication in theated the relationship between perceptions of private sector and military organizations isorganizational politics and job satisfaction. As needed. Third, these data were cross-sectional,shown in Table 4, the cross-product term added and a longitodinal design might permit an assess-a significant albeit small amount of variance to ment of causal linkages between job satisfaction,the equation (A R' = .01, f = 18.0, p < .01). PDM, organizational politics, and dispositionalOverall, the predictor variables accounted for a affect. Fourth, as noted by Madison, et al.moderate amount of the variance in perceived (1980), reliance on perceptions of organizationaljob satisfaction scores (total adjusted R2 = .48, politics may be subject to attributes of self-f = 74.5, p < .01). serving biases and self-defensive interpretations

of events. Fifth, the immoral vs. moral aspect ofIn order to identify the form of the interac- organizational politics (cf. Drory & Romm,

tion and thus test the hypothesis that organiza- 1988) was not examined. Despite these limita-

6

tions, the results support the hypotheses that additional opportunities for decision-making forperceptions of organizational politics are nega- personnel who are involved in fewer than desiredtively related to job satisfaction and that PDM decisions is extremely important from an or-moderates that relationship. ganizational perspective. They stressed (p. 157)

that allowing "individuals to move toward aWhile some individuals may prefer highly decisional equilibrium state must be a high

political work environments because they are priority for the manager." Indeed, high levels ofpredisposed to thrive in political climates, most PDM are neither always appropriate nor alwaysindividuals may be less satisfied with their jobs wanted.in such situations. Indeed, it may be difficult tobe job-satisfied when promotions, awards, and/or Additional research is needed to examine: (a)pay raises are based on political, rather than individual differences in preferences for PDMmerit considerations, or when there are cliques levels, (b) voice versus choice aspects of PDMclosely tied to organization leaders that typically and their comparative effects on perceived PDMget their way even at the cost of productivity, and related outcomes, and (c) the extent to which

different stages of development or career statusThe importance of organizational politics as may affect the extent to which PDM reduces the

a target problem for organizational assessment negative effects of organizational politics on joband intervention has been increasingly identified satisfaction.(Bateman, 1980; Cobb, 1986a, 1986b; Cobb &Margulies, 1981; Margulies & Raia, 1984). REFERENCESDirect attempts by individual supervisors andmanagers to reduce political activity may be Argyris, C. (1964). Integrating the Individuallimited in potential for success, because norms of and the Organization. New York: Wiley.organizational politics may be beyond the in-fluences of a single individual. However, a Ashforth, B. E., & Lee, R. T. (1990). Defensivesupervisor or manager may be able to reduce the behavior in organizations: A preliminarynegative effects of organizational politics on job model. Human Relations, 43, 621-648.satisfaction by providing more PDM oppor-tunities for subordinates. Bateman, T. (1980). Organizational change and

the politics of success. Group and Organiza-As noted by Witt and Myers (1991), the tional Studies, 5, 198-209.

importance of PDM as an antecedent of organiza-tional outcomes has been empirically identified, Batten, J. D., & Schwab, J. L. (1965). How toyet some managers avoid PDM and sharing crack down on company politics. Personnel,information. Some of these managers may do so 42, 8-20.because they do not know how to collaboratewith their workers, while others may explicitly Baumgartel, H. (1957). Leadership style as adecide to maintain or exercise power and autho- variable in research administration. Adminis-rity to keep their people unaware of goings-on, trative Sciences Quarterly, 2, 344-360.and others may simply have not thought aboutalternative management styles. Some managers Brush, D. H., Moch, M. K., & Pooyan, A.may also believe that they have more information (1987). Individual demographic differencesthan subordinates and thus do not engage in what on job satisfaction. Journal of Occupationalthey may perceive as needless PDM. Whatever Behaviour, 8, 139-156.the reason, the present data and the growingPDM literature provide sufficient evidence to Castrogiovanni, G. J., & Macy, B. A. (1990).indicate that efforts to increase PDM may lead to Organizational information-processingcapabi-favorable outcomes, although affective predispos- lities and degree of employee participation.ition may affect the extent to which it does. Group and Organization Studies, 15, 313-Nogradi and Koch (1981) argued that providing 336.

7

Cavanaugh, G. F., Moberg, D. J., & Velasquez, Drory, A., & Romm, T. (1988). Politics inM. (1981). The ethics of organizational organization and its perception within thepolitics. Academy of Management Review, 6, organization. Organization Studies, 9, 165-363-374. 179.

Cobb, A. T. (1984). An episodic model of Drory, A., & Romm, T. (1990). The definitionpower: Toward an integration of theory and of organizational politics: A review. Humanresearch. Academy of Management Review, Relations, 43, 1133-1154.9, 482-493.

DuBrin, A. J. (1978). Winning at Office Politics.Cobb, A. T. (1986a). Informal influence in the New York: Van Nostrand Reinhold.

formal organization. Group and Organiza-tional Studies, 11, 229-253. DuBrin, A. 3. (1988). Career maturity, organiza-

tional rank, and political behavioral tenden-Cobb, A. T. (1986b). Political diagnosis: Ap- cies: A correlational analysis of organization-

plications in organizational development. Aca- al politics and career experience. Psychologi-demy of Management Review, 11, 482-496. cal Reports, 63, 531-537.

Farrell, D & Peterson, J. C. (1982). PatternsCobb, A. T., & Margulies, N. (1981). Organi7 .- of political behavior in organizations. Aca-

tion development: A political perspective. demy of Management Review, 7, 403-412.Academy of Management Review, 6, 49-59.

Ferris, G. R., Russ, G. S., & Fandt, P. M.Cohen, J., & Cohen, P. (1983). Applied Multiple (1989). Politics in organizations. In R. A.

Regression/Correlation Analysis for the Be- Giacalone and P. Rosenfeld (Eds.), Impres-havioral Sciences. Hillsdale, NJ: Lawrence sion Management in the Organization (pp.Erlbaum Associates. 143-170). Hillsdale, NJ: Lawrence Earlbaum.

Cohen, R. L. (1985). Procedural justice and Frost, P. I., & Hayes, D. C. (1977, September).participation. Human Relations, 38, 643-663. An Exploration in To Cultures of Political

Behaviors in Organizations. Paper presentedCotton, J. L., Vollrath, D. A., Froggatt, K. L., at the Conference on Cross Cultural Studies

Lengnick-Hal, M. L., & Jennings, K. R. in Organizational Functioning, Hawaii.(1988). Employee participation: Diverseforms and different outcomes. Academy of Gandz, J., & Murray, V. V. (1980). The ex-Management Review, 13, 8-22. perience of workplace politics. Academy of

Management Journal, 23, 237-251.Dachler, H. P. (1978). The problem nature of

participation in organizations: A conceptual Gray, B., & Ariss, S. S. (1985). Politics andevaluation. In B. King, S. Sweufert, and F. strategic change across organizational lifeE. Fiedler (Eds.), Managerial Control and cycles. Academy of Management Review, 10,Organization Democracy (pp. 17-29). New 707-723.York: John Wiley and Sons.

Greenberg, J., & Folger, R. (1983). ProceduralDalton, M. (1959). Men Who Manage. New justice, participation. ;ind the fair process

York: Wiley. effect in groups and organizations. In P.Paulus (Ed.), Basic Group Processes (pp.

Diener. E., & Emmons, R. A. (1985). The 235-256). New York: Springer-Verlag.independence of positive and negative affect.Journal of Personality and Social Psycho-logy, 47, 1105-1117.

8

Hoppock, R. (1935). Job Satisfaction. New Mayes, B. T., & Allen, R. W. (1977). TowardYork: Harper & Row. a definition of organizational politics. Aca-

demy of Management Review, 2, 672-678.Jackson, S. E. (1983). Participation in decision

making as a strategy for reducing job-related McNichols, C. W., Stahl, M. J., & Manley, T.strain. Jo,,rnal of Applied Psychology, 68, R. (1978). A validation of Hoppock's job3-19. satisfaction measure. Academy of Mnage-

mentJournal, 21, 737-742.Jones, S. (1987). Organizational politics: Only

the darker side? Management Education and Mintzberg, H. (1985). The organization as aDevelopment, 18, 116-128. political arena. Journal of Management Studi-

es, 22, 134-154.Kacmar, K. M., & Ferris, G. R. (1991). Per-

ceptions of Organizational Politics Scale Narayanan, V. K., & Fahey, L (1982). The(POPS): Development and construct valida- micro-politics of strategy formulation. Aca-tion. Educational and Psychological Mea- demy of Management Review, 7, 25-34.surement, 51, 193-205.

Niehoff, B. P., Enz, C. A., & Grover, R. A.Levin, I., & Stokes, J. P. (1989). Dispositional (1990). The impact of top-management

approach to job satisfaction: Role of negative actions on employee attitudes and percep-affectivity. Journal of Applied Psychology, tions. Group and Organization Studies, 15,74, 752-758. 337-352.

Luthans, F., Hodgetts, R. M., & Rosenkrantz, Nogradi, G. S., & Koch, S. A. (1981). TheS. A. (1988). Real Managers. Cambridge, relationship between decisional participationMA: Ballinger. and commitment to the organization, com-

munity, and profession among municipalMacher, K. (1986). The politics of people. recreation administrators. Leisure Sciences,

Personnel Journal, 65, 50-53. 4, 143-159.

Madison, D. L., Allen, R. W., Porter, L. W., Nye, L. G., & Witt, L. A. (1992). Dimen-Renwick, P. A., & Mayes, B. T. (1980). Or- sionality and Construct Validity of the Per-ganizational politics: An exploration of mana- ceptions of Organizational Politics Scalegers' perceptions. Human Relations, 33, 79- (POPS). Technical Report # DOT/FAA/AM-100. 92-10, Washingtun, D.C.: Office of Aviation

Medicine, Federal Aviation Administration.March, J. G., & Simon, H. A. (1958). Or-

ganizations. New York: Wiley. Pettigrew, A. M. (1973). The Politics of Or-ganizational Decision Making. London:

Margulies, N., & Raia, T. (1984). The politics Tavistock.of organization development. Training andDevelopment Journal, 38, 20-23. Pfeffer, J. & Salancik, G. R. (1974). Organiza-

tional decision making as a political process:Martin, J. H., & Sims, H. J. (1974). Power The case study of a university budget. Ad-

tactics. In D. A. Kolb, I. M. Rubin, and J. ministrative Sciences Quarterly, 19, 135-151.M. McIntyre (Eds.), Organizational Psycho-logy: A Book of Readings (pp. 177-178). Pfeffer, J. (1978). Organizational Design. Ar-Englewood Cliffs, NJ: Prentice-Hall. lington Heights, IL: AE:M Publishing.

9

Pfeffer, J. (1981). Power in Organizations. Ungson, G. R., & Steers, R. M. (1984). Motiva-Marshfield, MA: Pitman. tion and politics in executive compensation.

Academy of Management Review, 9, 313-Podsakoff, P. M., & Organ, D. W. (1986). Self- 323.

reports in organizational research: Problemsand prospects. Journal of Management, 12, Vredenburgh, D. J., & Maurer, J. G. (1984). A531-544. process framework of organizational politics.

Human Relations, 37, 47-66.Porter, L. W., Allen, R. W., & Angle, H. L.

(1983). The politics of upward influence in Watson, D., & Clark, L. A. (1984). Negativeorganizations. In R. W. Allen and L. W. affectivity: The disposition to experiencePorter (Eds.) Organizational Influence Pro- aversive emotional states. Psychologicalcesses (pp. 408-422). Glenview, IL: Scott, Bulletin, 96, 465-490.Foresman, & Co.

Watson, D., Clark, L. A., & Tellegen, A.Robbins, S. P. (1976). The Administrative Pro- (1988). Development and validation of brief

cess: Integrating Theory and Practice. measures of positive and negative affect: TheEnglewood Cliffs, NJ: Prentice-Hall. PANAS scales. Journal of Personality and

Social Psychology, 54, 1063-1070.Robbins, S. P. (1979). Organizational Behavior.

Englewood Cliffs, NJ: Prentice-Hall. Wildavsky, A. (1964). The Politics of theBudgeting Process. New York: Little Brown.

Robbins, S. P. (1983). Organizational Behavior.Englewood Cliffs, NJ: Prentice-Hall. Witt, L. A. (1991). Negative affect as a

moderator of role stressor-job attitude rela-Schein, E. H. (1978). Career dynamics: Match- tionships. Military Psychology, 3, 151-162.

ing Individual and Organizational Needs.Reading, MA: Addison-Wesley. Witt, L. A., & Myers, J. G. (1991). Perceived

environmental uncertainty and participation inSchein, V. E. (1977). Individual power and decision-making in the prediction of fairness

political behaviors in organizations: An inade- of personnel decisions. In L. A. Witt and J.quately explored reality. Academy of Mana- G. Myers (Eds.), Two Studies of Participa-gement Review, 2, 64-72. tion in Decision-making and Equity Among

FAA Personnel (pp. 9-14). Technical ReportStone, E. F., & Hollenbeck, J. R. (1989). # DOT/FAA/AM-91-10, Washington, D. C.:

Clarifying some controversial issues sur- Office of Aviation Medicine, Federal Avia-rounding statistical procedures for detecting tion Administration.moderating variables: Empirical evidence andrelated matters. Journal of Applied Psycho-logy, 74, 3-10.

Thibaut, J., & Walker, L. (1975). ProceduralJustice: A Psychological Analysis. Hillsdale,NJ: Lawrence Earlbaum.

Tushman, M. (1977). A political approach toorganizations: A review and rationale. Aca-demy of Management Review, 2, 206-216.

10

TABLE 3.

Intercorrelation Matrix

Variable 1 2 3 4 5 6 7 8 9 10 11

1. Sex2. Minority/Non-Minority .053. Supervisor Status -. 16 .044. Education -.23 .02 .205. Grade levels -.35 -.03 .26 .376. Age -. 19 -.01 .13 .00 .257. Time in current org. -.05 .05 .17 .02 .30 .488. PDM .07 .04 .14 -.02 -.03 -.03 -.019. Organizational Politics .01 .04 -.21 -.05 -.09 .01 .07 -.49

10. Job Satisfaction .10 .00 .11 -.07 -.05 .04 -.06 .37 -.6211. Positive Affect .04 .03 .15 -.01 -.02 .02 -. 10 .28 -.33 .4412. Negative Affect -.03 -.02 -.03 -.02 -.02 -.03 .06 -.23 .42 -.41 -.33

Note: Correlations > r = .06 are p < .05. Correlations > r = .09 are p < .01 (2-tailed testsof significance).

TABLE 4.Results of Hierarchical Regression Analyses

Predicting Job Satisfaction

Predictor Variable(s) AR 2 F p <

Biodata .04 5.1 .01Addition of PANAS scores .27 179.6 .01Addition of PDM and politics scores .17 313.2 .01Addition of interaction term .01 18.0 .01

11

Figure 1.Job Satisfaction Regressed On Perceptions of

Organizational Politics: Low, Average, and High PDM Scores

Moderating Effect of PDM

Job Satisfaction14.1

13.1

12.1

11.1

Level of PDM

Low -- Avg * High

10.125 52

Organizational Politics

Note: Y = (-.204 + .O06f)X + (-.179f + 21.51), wheref = one standard deviation below the meanof PDM (or the mean of PDM or one standard deviation above the mean of PDM), Y = the job satisfac-tion score, and X = the politics score.

*U.S.GPO: 1992-661-063/40078

12