August, 1987

« The Academy of Management EXECUTIVE, 1987, Vol. 1, No. 3,pp. 183-193

Behind the Mask:The Politics of Employee Appraisal

By Clinton 0. LongeneckerThe University of Toledo

Henry P. Sims, Jr.George Mason University and

The Pennsylvania State University

Dennis A. GioiaThe Pennsylvania State University

is really no getting around the factthat whenever I evaluate one of my peo-

ple, I stop and think about the impact—theramifications of my decisions on my relation-ship with the guy and his future here. I'd bestupid not to. Call it being politicallyminded, or using managerial discretion, orfine tuning the guy's ratings, but in the endI've got to live with him, and I'm not going torate a guy without thinking about the fall-out. There are a lot of games played in therating process and whether we [managers]admit it or not we are all guilty of playingthem at our discretion.

According to management hooks and manuals, em-ployee appraisal is an ohjective, rational and, we hope,accurate process. The idea that executives might delih-erately distort and manipulate appraisals for politicalpurposes seems unspeakable. Yet we found extensiveevidence to indicate that, behind a mask of objectivityand rationality, executives engage in such manipulationin an intentional and systematic manner. In perform-ance appraisal, it appears that some of the Machiavel-lian spirit still lives.

Our original goal was to conduct a scholarly inves-tigation of the cognitive processes executives typicallyuse in appraising subordinates. We held in-depth inter-views with 60 upper-level executives who had extensiveexperience in formally evaluating their subordinates ona periodic basis. During these interviews, we heardmany frank admissions of deliberate manipulation offormal appraisals for political purposes. In this articlewe'll discuss the "why and the how" of such politicallymotivated manipulation.

On the Appraisal ProcessAlmost every executive has dreaded performance

appraisals at some time or other. They hate to givethem and they hate to receive them. Yet, like them ornot, every executive recognizes that appraisals are afact of organizational life. In terms of time, a formalappraisal of a subordinate takes perhaps three or fourhours out of the working year; in terms of impact onthe lives of executives and their employees, appraisalshave significance that reaches far beyond the few hoursit takes to conduct them.

Because of the important role appraisals play inindividual careers and corporate performance, a greatdeal of attention has been given to trying to under-stand the process. Special attention has been directedtoward the issue of accuracy in appraisals.' Academi-cians in particular have expended (some might saywasted) substantial energy trying to design the perfectinstrument that would yield an accurate appraisal.That effort now appears to be a hopeless, even impossi-ble, task.

More recently, a flurry of activity has centered onthe arcane mental processes of the manager who givesthe appraisal. It is an intruiging approach because itinvolves a kind of vicarious attempt to climb inside anexecutive's head to see how he or she works. Predict-ably, however, this approach has confirmed the elusive-ness of deciphering managerial thought processes.Moreover, it has not yet resulted in appraisals that areany more accurate than existing appraisals.^

Even more recently, some efl'ort has been directedtoward demonstrating that appraisal is, in addition toeverything else, a highly emotional process as well.When emotional variability gets dragged into the pro-cess, any hope of obtaining objectivity and accuracy inappraisal waltzes right out the ofifice door.^

Taken together, all these approaches apparentlylead to the depressing conclusion that accuracy in ap-praisals might be an unattainable objective.' More re-alistically, perhaps accuracy is simply a wrong goal topursue. Even if we have a perfect understanding of in-struments and mental and emotional processes, wouldthat result in accurate appraisals? Our research indi-cates that it would not. All of these avenues to under-standing appraisal tend to ignore an important point:

183

Academy of Management EXECUTIVE

Method

Our research approach involved in-depth, semis-tructured interviews with 60 executives. The partici-pants in the study came from seven large organizationsand represented 11 functional areas. As a group, theyaveraged more than 20 years of work experience andmore than 13 years of managerial experience. Collec-tively, they had performance appraisal experience in197 organizations. Conclusions reported here, then, arederived from a diversity of executives.

Each tape-recorded interview was designed to tapthe executive's perception of his or her own perform-ance appraisal processes. The interviews averaged morethan one and one-half hours in length. Although theinterview used some a priori "probes," the interviewingstrategy mainly encouraged the subject to respondfreely and subjectively.

The data collection yielded more than 100 hoursof tape-recorded verbal data. All data from each inter-view were transcribed onto five-by-eight cards thatmainly consisted of executives' directly quoted state-ments, with each card containing one statement,thought, or observation by an executive on a given

topic. The transcription process yielded 1,400 cards,which were then classified according to various politicalissues that emerged during the interviews.

For a classification group to qualify as a potential"finding," a minimum of 72% of the respondents hadto have brought up that issue. A research assistantthen read each group of cards and assigned a label thatcaptured the "essence" of the executives' views on aparticular aspect of the appraisal process. The out-comes from this process were the designated findings ofthe study. To further enhance the reliability and valid-ity of the research, two research assistants then inde-pendently developed frequency counts for each finding.They tallied the number of cards in each classificationgroup that supported the finding that had been identi-fied in the second step of the analysis. The frequenciestabulated by each judge ranged from a low of 43 re-sponses (72%) to a high of 57 (95%). A correlationanalysis of the frequencies revealed an r = .94 as ameasure of inter-rater reliability in identifying thefindings.

Appraisals take place in an organizational environmentthat is anything but completely rational, straightfor-ward, or dispassionate. In this environment, accuracydoes not seem to matter to managers quite so much asdiscretion, effectiveness or, more importantly, survival.Earlier research has either missed or glossed over thefact that executives giving appraisals have ulterior mo-tives and purposes that supercede the mundane con-cern with rating accuracy.

On Politics in PerformanceAppraisal

Any realistic discussion of performance appraisalmust recognize that organizations are political entitiesand that few, if any, important decisions are madewithout key parties acting to protect their own inter-ests." As such, executives are political actors in an or-ganization, and they often attempt to control theirdestinies and gain influence through internal politicalactions.

Thus, it is likely that political considerations in-fluence executives when they appraise subordinates.'Politics in this sense refers to deliberate attempts byindividuals to enhance or protect their self-interestswhen conflicting courses of action are possible. Politicalaction therefore represents a source of bias or inaccu-racy in employee appraisal. To understand the ap-

praisal process thoroughly, thus, we must recognizeand account for the political aspects of the process.

Politics in Appraisal: Findingsfrom the Study

The political perspective emerged as a surpris-ingly important and pervasive issue afl'ecting the wayexecutives appraise their employees. Conclusions de-rived from our interviews are summarized in Exhibits 1through 4. Because a strong attempt was made to allowexecutives to speak for themselves in describing thepolitics of performance appraisals, direct quotationsfrom the interviews have been included in our analysis,where appropriate. Our findings are discussed below.

Politics as a Reality of OrganizationalLife

The most fundamental survey finding was anopen recognition and admission that politics were a re-ality in the appraisal process. In fact, executives admit-ted that political considerations nearly always werepart of their evaluation process. One vice-presidentsummarized the view these executives shared regardingthe politics of appraisal:

184

August, 1987

As a manager, I will use the review process todo what is best for my people and the divi-sion. . . . I've got a lot of leeway—call it dis-cretion—to use this process in that manner.. . . I've used it to get my people betterraises in lean years, to kick a guy in thepants if he really needed it, to pick up a guywhen he was down or even to tell him that hewas no longer welcome here. It is a tool thatthe manager should use to help him do whatit takes to get the job done. I believe most ofus here at operate this way regard-ing appraisals. . . . Accurately describing anemployee's performance is really not as im-portant as generating ratings that keepthings cooking.

Executives suggested several reasons why politicswere so pervasive and why accuracy was not their pri-mary concern. First, executives realized that they mustlive with subordinates in a day-to-day relationship.Second, they were also very cognizant of the perma-nence of the written document:

The mere fact that you have to write outyour assessment and create a permanent rec-ord will cause people not to be as honest oras accurate as they should be. . . . We softenthe language because our ratings go in theguy's file downstairs fthe Personnel Depart-ment] and it will follow him around hiswhole career.

Perhaps the most widespread reason why execu-tives considered political action in the appraisal pro-cess was that the the formal appraisal was linked tocompensation, career, and advancement in the organi-zation. The issue of money was continually cited as amajor cause of intentional distortions in ratings.

/ know that it sounds funny, but the factthat the process is ultimately tied to moneyinfiuences the ratings a person receives. . . .Whenever a decision involves money thingscan get very emotional and ticklish.

Although the logic of tying pay to the outcome ofperformance ratings is sound, pay linkages increase thelikelihood that ratings will be manipulated. Both man-agers and the organization as a whole are guilty of us-

ing the rating process as an opportunity to reach salaryobjectives regarding employee compensation that havelittle, if any, relationship to pay for performance. A di-rector of research and development very candidly de-scribed the predicament from the rater's perspective:

Since the pay raise my people get is tied tothe ratings I give them, there is a strong in-centive to inflate ratings at times to maxi-mize their pay increases to help keep themhappy and motivated, especially in leanyears when the merit ceiling is low. . . . Con-versely, you can also send a very strong mes-sage to a nonperformer that low ratings willhit him in the wallet. . . . There is no doubtthat a lot of us manipulate ratings at timesto deal with the money issue.

At times, an organization uses the appraisal pro-cess as an instrument to control merit increase expend-itures. The manipulative process can be summarized asfollows:

This thing fthe appraisal process] can reallyturn into an interesting game when the HRfHuman Resources] people come out with ablanket statement like, "Money for raises istight this year and since superior performersget 7% to 10% raises there will be no supe-rior performers this year." Talk about mak-ing things rough for us fraters]!. . . They tryand force you to make the ratings fit themerit allowances instead of vice versa.

Exhibit 1

Politics as a Reality of Organizational Life

• Political considerations were nearly always partof executive evaluative processes.

• Politics played a role in the evaluation processbecause:—executives took into consideration the daily

interpersonal dynamics between them andtheir subordinates;

—the formal appraisal process results in a per-manent written document;

—the formal appraisal can have considerableimpact on the subordinate's career andadvancement.

185

Academy of Management EXECUTIVE

Influences on Political Culture

Executives made it clear that if an organizationwas political, the appraisal process would reflect thesepolitics:

Some organizations are more aggressive andpolitical than others, so it just makes sensethat those things carry over into the ratingprocess as well. . . . The organization's cli-mate will determine, to a great extent, howsuccessful any rating system will be, and itfollows that if any organization is very politi-cal, the rating system will be political. . . .

Several factors were identified hy the executivesas having a strong influence on the political culture inwhich the performance appraisal process operates. Per-haps the strongest was the extent to which the formalappraisal process was "taken seriously" hy the organi-zation. A plant manager in this study describes what itmeans for an organization to "take the processseriously":

At some places the PA fperformance ap-praisal] process is a joke—just a bureaucrac-tic things that the manager does to keep theIR [industrial relations] people off his back.At the last couple of places I've worked, theformal review process is taken really seri-ously; they train you how to conduct a goodinterview, how to handle problems, how tocoach and counsel. . . . You see the things[appraisals] reviewed by your boss, and he'sserious about reviewing your performance ina thorough manner. . . . I guess the biggestthing is that people are led to believe that itis a management tool that works; it's got tostart at the top!

This quote suggests another important factor thatturns the appraisal process into a political process: theextent to which higher level executives in the samecompany use political factors in rating subordinates. A"modeling" effect seems to take place, with managerstelling themselves, "If it's okay for the guys upstairs todo it, then we can do it, too."

According to one executive we interviewed,

I've learned how not to conduct the reviewfrom the bosses . . . but you do learn fromyour boss how much slack or what you canget away with in rating your people. . . . Itseems that if the manager's boss takes it [theappraisal] seriously, the subordinate [man-ager] is more likely to follow. If the bossplays games with the review, it seems likethe subordinate [manager] is more likely todo so.

The economic health and growth potential of theorganization appeared as important factors influencingthe organization's culture and, consequently, the ap-praisal event. Similarly, the executive's own personalbelief system—his or her perception of the value of theappraisal process—also seemed to have an impact.Generally, executives who honestly believed the pro-cess contributed to the motivation of their subordi-nates were less likely to allow political factors to afl'ectthe appraisal. Conversely, executives who saw the ap-praisal as a useless bureaucratic exercise were morelikely to manipulate the appraisal.

Moreover, if executives believed the appraisalswould he seriously scrutinized, reviewed, and evaluatedby their superiors, then the influence of political fac-tors was likely to be reduced.

/ / somebody is carefully reviewing the marksyou give your people, then the game playingis reduced. . . [but] as you rise in the organi-zation, your boss has less direct knowledge ofyour people and is less likely to questionyour judgment, so the door is open for morediscretion.

The degree of open communication and trust be-tween executives and subordinates seemed to havesome influence on the impact of political factors. Themore open the communication, the less likely thatpolitics would play a role:

/ / the manager and employee have a trustingand open relationship and shoot straightwith each other, then the manager is lesslikely to play games with ratings.

186

August, 1987

Last, but not least, the appraiser's level in the or-ganization's hierarchy also seemed to have an influ-ence. Executives generally believed the appraisal pro-cess became more political and subjective as onemoved up the organizational ladder:

The higher you rise in this organization themore weird things get with regard to howthey evaluate you. . . . The process becomesmore political and less objective and it seemslike the rating process focuses on who youare as opposed to what you've actually ac-complished. . . . As the stakes get higher,things get more and more political.

Exhibit 2

Factors Influencing the Political Cultureof the Organization

• The economic health and growth potential ofthe organization

• The extent to which top management supportedand, more importantly, did or did not practicepolitical tactics when appraising their ownsubordinates

• The extent to which executives sincerely be-lieved that appraisal was a necessary and worth-while management practice or just a bureau-cratic exercise

• The extent to which executives believed thattheir written assessment of their subordinateswould be evaluated and scrutinized by theirsuperiors

• The extent to which an organization was willingto train and coach its managers to use andmaintain the performance appraisal system

• The degree to which the appraisal process wasopenly discussed among both executives andsubordinates

• The extent to which executives believed the ap-praisal process became more political at higherlevels of the organizational hierarchy

Inflating the Appraisal

Although academicians have been preoccupiedwith the goal of accuracy in appraisal, executives re-ported that accuracy was not their primary concern.Rather, they were much more interested in whethertheir ratings would be efl"ective in maintaining or in-creasing the subordinate's future level of performance.In fact, many reported they would deliberately mis-state the reported performance level if they felt per-formance could be improved as a result:

When I rate my people it doesn't take placein a vacuum . . . so you have to ask yourselfwhat the purpose of the process is. . . . I usethis thing to my advantage and I know mypeople and what it takes to keep them goingand that is what this is all about.

Overall, executives reported that deliberate dis-tortions of the appraisal tended to be biased in thesubordinate's favor:

Let's just say that there are a lot of factorsthat tug at you and play on your mind thatcause you to tend to soften the ratings yougive. It may not have a great impact all thetime but when you know a "5" will piss aman off and "6" will make him happy. . . .You tell me which one you'd choose. . . .Plus, you don't want to be the bad guy, thebearer of gloom. It seems like ratings are al-most always a little inflated at a minimumbecause of people aspects in the evaluationprocess.

Typically, executives tended to inflate the overallrating rather than the individual appraisal items. In-terestingly, although the overall rating was generallythe last item on the appraisal form, this overall ratingwas determined first; then the executive went back andcompleted the individual items.

Most of US try to be fairly accurate in assess-ing the individual's performance in differentcategories. . . . If you are going to pump upa person's ratings, for whatever reason, it'sdone on the subordinate's overall evaluationcategory. That's all they really care about,anyway. . . . The problem is these thingshave to match up, so if you know what theguy's overall rating is in the first place it willprobably color the rest of the appraisal.

Of course, this backward procedure is usually con-trary to the recommended procedure and is also incon-sistent with the typical assumptions about how deci-sions are supposed to be made "objectively."Executives articulated several reasons as justificationfor consciously infiating subordinate ratings. The mostfrequently given reason was to maximize the merit in-creases that a subordinate would be eligible to receive.This reason was more likely to be given by executivesin organizations that closely linked the numerical score

187

Academy of Management EXECUTIVE

on the formal appraisal and the subsequent merit raise.Sometimes executives wanted to protect or en-

courage a subordinate whose performance was tempo-rarily suffering because of personal problems. In a sim-ilar vein, executives would sometimes inflate a ratingsimply because they felt sorry for a subordinate. Theywanted to avoid short-term "punishment" in the hopethat the subordinate would recover and perform onceagain at an acceptable level.

It may sound kind of funny to say this, butsometimes there is a tendency to give subor-dinates ratings a little higher than they de-serve because you feel sorry for them. . . . Ijust had a guy go through a divorce and I'mnot going to kick him when he's down, even ifhis performance drops off. . . . If anything,you might use the review to help pick him upand get him back on his feet.

If the appraisal was reviewed by people outsidethe department, executives sometimes inflated ratingsto avoid "hanging dirty laundry out in public." Clearly,many executives preferred to keep knowledge ofproblems contained within the department.

There are two reviews at times, the writtenone and the spoken one. The spoken reviewis the real one, especially if there are thingsof a sensitive nature. . . . I generally don'tput those things down on paper in the reviewfor the whole world to read because it is gen-erally none of their damn business. . . . Icould make all of us look bad or worse thanwe really are.

Executives also admitted to inflating a rating toavoid a confrontation with a subordinate with whomthe executive had recently had difficulties. They tookthis action mainly to avert an unpleasant incident orsometimes to avoid a confrontation that they believedwould not lead to an effective outcome.

On occasion, an executive might inflate the ratingbecause the subordinate's performance had improvedduring the latter part of the performance period, eventhough the overall performance did not merit such arating. Again, the motivation for this higher-than-de-served rating was a desire to encourage the subordinatetoward better performance in the next period:

Many of us have trouble rating for the entireyear. If one of my people has a stellar threemonths prior to the review. . . . you don'twant to do anything that impedes that per-son's momentum and progress.

Executives also recognized efl"ort, even though theeffort might not pay off in actual performance:

/ / a man broke his back trying to do the bestjob humanly possible, his ratings will gener-ally refiect this if his boss understands peo-ple. Take two people with the same perform-ance, but one tried much harder—theirratings will show it in my department. Lowratings might trample that person's desire toput forth effort in the future.

Last, although not frequently reported, a few ex-ecutives admitted to giving a higher rating to a prob-lem employee to get the employee promoted "up andout" of the department. Although executives only occa-sionally admitted to t;his, the "up and out" rating pro-cess was almost universally discussed as somethingother managers actually do. One plant manager can-didly remarked:

I've seen it happen, especially when you geta young guy in here who thinks he's only go-ing to be here a short while before he getspromoted. People like that become a realpain in the ass. . . . If you want to get rid ofthem quick, a year and a half of good ratingsshould do it. . . . A lot of people inflate rat-ings of people they can't stand, or who thinkthey are God's gift to the department, just toget rid of them. Amen.

Of course, this practice helps an executive avoiddealing with performance problems and passes theproblem along to someone else. Mainly, this tactic wasemployed when an executive felt unable or unwilling todeal with a performance problem or, especially, whenthe source of the problem seemed to be based on "per-sonality" or "style" conflicts.

188

August, 1987

Exhibit 3

Inflating the Appraisal

Executives inflated the appraisal to provide rat-ings that would effectively maintain or increasethe subordinate's level of performance (the pri-mary concern was not the accuracy of theratings).Inflated ratings occur primarily on the overallperformance rating, as opposed to the individualappraisal itemsExecutive justification for inflating the appraisal:— to maximize the merit increases a subordinate

would be eligible to receive, especially whenthe merit ceiling was considered low;to protect or encourage a subordinate whoseperformance was suffering because of personalproblems (feeling sorry for a subordinate alsoresulted in an inflated appraisal);

— to avoid hanging dirty laundry out in public ifthe performance appraisal would be reviewedby people outside the organization;

—to avoid creating a written record of poor per-formance that would become a permanentpart of a subordinate's personnel file;

—to avoid a confrontation with a subordinatewith whom the manager had recently haddifficulties;

—to give a break to a subordinate who had im-proved during the latter part of the perform-ance period;

—to promote a subordinate "up and out" whenthe subordinate was performing poorly or didnot fit in the department.

Deflating the Appraisal

For the most part, executives indicated that theywere very hesitant to deflate a subordinate's rating be-cause such a tactic would lead to subsequent problems:

/ won't say I've never given a subordinatelower ratings then he or she deserves becausethere's time and place for that type of thing,but let's just say I hesitate to do that sort ofof thing unless I'm very sure of what the out-come will be and that it won't backfire.

Nevertheless, negative distortions did occur. Ex-ecutives gave several reasons for using this tactic. First,an overly negative rating was sometimes used to jolt asubordinate to rise to his or her expected performancelevel:

I've used the appraisal to shock an employee.. . . If you've tried to coach a guy to get himback on track and it doesn't work, a low rat-ing will more often than not slap him in theface and tell him you mean business. . . . I'vedropped a few ratings way down to accom-plish this because the alternative outcomecould be termination down the road, whichisn't pretty.

Also, a deliberately deflated rating was sometimes usedto teach a rebellious subordinate a lesson:

Occasionally an employee comes along whoneeds to be reminded who the boss is, andthe appraisal is a real tangible and appropri-ate place for such a reminder . . . .

Deflated ratings were also used as part of a terminationprocedure. First, a strongly negative rating could beused to send an indirect message to a subordinate thathe or she should consider quitting:

/ / a person has had a questionable period ofperformance, a strong written appraisal canreally send the message that they aren't wel-come any longer and should think aboutleaving. . . . The written review sends a clearmessage if the person has any doubt.

Second, once the decision has heen made that the situ-ation was unsalvageable, negative ratings could then beused to build a strongly documented case against themarginal or poor performer:

You'll find that once a manager has made uphis or her mind that an employee isn't goingto make it, the review [the written docu-ment] will take on an overly negative tone.. . . Managers are attempting to protectthemselves. . . . The appraisal process be-comes downwardly biased because they [themanagers] fear that discussing and docu-menting any positives of the employee's per-formance might be used against them at alater point in time.

Of course, this tactic has recently become morecommon because of lawsuits challenging the traditional"employment at will" concept. The courts have clearly

189

Academy of Management EXECUTIVE

stated that terminations must not be frivolous; theymust be justified by economic constraints or documen-tation of poor performance. In these cases managerswill use the process to protect themselves from litiga-tion associated with an unlawful termination lawsuit.'

Exhibit 4

Deflating the Appraisal

• Executives indicated that they were very hesi-tant consciously to deflate a subordinate's rat-ings because of potential problems associatedwith such a tactic.

• Nevertheless, they sometimes deflatedappraisals:—to shock a subordinate back on to a higher

performance track;—to teach a rebellious subordinate a lesson

about who is in charge;—to send a message to a subordinate that he or

she should consider leaving the organization;—to build a strongly documented record of poor

performance that could speed up the termina-tion process.

SummaryOur research clearly showed that executives be-

lieved there was usually a justifiable reason for gener-ating appraisal ratings that were less than accurate.Overall, they felt it was within their managerial discre-tion to do so. Thus our findings strongly suggest thatthe formal appraisal process is indeed a political pro-cess, and that few ratings are determined without somepolitical consideration. Although research on rater "er-ror" has traditionally suggested that raters can and doinflate ratings (leniency errors) and deflate ratings(stringency errors), researchers have typically not ac-counted for the realities of the appraisal context to ex-plain why these errors occur.

In the minds of the managers we interviewed,these thoughts and behaviors are not errors but, rather,discretionary actions that help them manage peoplemore effectively. Executives considered many factorsbeyond the subordinate's actual performance in theirratings. Thus, organizational politics was a major factorin the intentional manipulation of subordinate ratings.

Our findings provide support for the following po-litical realities of organizational life: (1) executives inlarge organizations are political actors who attempt toavoid unnecessary conflict; (2) they attempt to use theorganization's bureaucractic processes to their own ad-vantage; and (3) they try to minimize the extent towhich administrative responsibilities create barriersbetween them and their subordinates.

We also conclude that the organizational culturein which the appraisal event occurs significantly influ-enced the extent to which political activity would bothdevelop and operate. Of course, organizationwide pat-terns are also strongly influenced by the support andpractice of top management. Indeed, we know thatlower-level managers tend to emulate high-status exec-utives, and the way they use the appraisal process is noexception. Thus, if top managers prepare ratingspoorly or deliberately distort them, this behavior willtend to cascade down the organization.

Given these findings, what informative observa-tions or constructive recommendations might we maketo minimize, or at least manage, the detrimental effectsof politics in employee appraisal? In fact, we have sev-eral for both the individual manager and the organiza-tion as a whole.

The Individual Manager

1. Quite frankly, our data suggest there are timesin organizational life when political necessitysupercedes the usually desirable goals of accuracy andhonesty in appraisal. The executives interviewed sug-gested several compelling reasons for exercising mana-gerial discretion contrary to traditional appraisal re-search recommendations. Clearly, there are times whenindividual employees and the organization as a wholecan benefit as a consequence. The caveat, of course, isthat the occasions when politics and discretion necessa-rily intrude on the appraisal process should be chosenjudiciously. The overall effect on the organizationshould be given due consideration.

2. Performance appraisal is perhaps most usefullyviewed as a high-potential vehicle for motivating andrewarding employees, rather than as a mandatory, bu-reaucratic exercise used only for judgmental or manip-ulative purposes. Ideally, it should be treated as an op-portunity to communicate formally with employeesabout their performance, their strengths and weak-nesses, and their developmental possibilities.

3. Executives should bear in mind that appraisal-related actions, tike many other organizational activi-ties, serve as guides for subordinates. Employees whomust conduct appraisals often learn appraisal attitudesand behaviors from their bosses. Thus if appraisals areto be effective, high-ranking executives must treat theprocess as significant so that political manipulation isdiscouraged.

4. In addition, openness and trust between man-agers and subordinates seems to be associated with alower level of detrimental political activity. Cultivatingunderstanding seems to reduce the perceived need forresorting to interpersonal politics.

5. Finally, infiating or deflating appraisal ratingsfor political ends might serve temporarily to help exec-utives avoid a problem with certain employees or to ac-complish some specific purpose. However, such inten-

190

August, 1987

tional manipulation may eventually come back tohaunt the perpetrating executive and, ultimately, theorganization as a whole. This is especially likely if thecompany comes to accept political manipulation of ap-praisals as part of the norm.

The Organization as a Whole

1. The appraisal process should operate in a sup-portive organizational culture. Effective appraisal sys-tems are characterized by the support of top managers(who conduct appraisals themselves), training, opendiscussions of the appraisal process on an annual basis(perhaps a quality circle approach to appraisals), andrewarding the efforts of managers who do top-notchappraisals.

2. Systematic, regular, and formal appraisalsshould start at the top of the organization. We foundthat top executives want formal appraisals and rarelyget them. If appraisals are not done at the top, themessage sent to the rest of the organization is, "Theyaren't very important and thus shouldn't be taken seri-ously." As a result, the door to more political activity isopened wider.

3. Further, although training on how to do effec-tive appraisals is important, managers also need to betrained on why they need to be done. Understandingthe rationale for appraisals is important in building theperception that the appraisal process is an effectivemanagerial tool and not merely a required bureaucrac-tic procedure.

4. Open discussion of the political aspects of theappraisal process (and their legal ramifications) shouldbe included in appraisal training programs. Althoughmanagers made it clear that political manipulation ofratings is commonplace, political issues were neueropenly discussed in either training programs or in man-agement development efforts.

5. When money is tied to the rating process, polit-ically oriented ratings tend to increase. This creates adilemma: A "pay for performance" management phi-losophy depends on the "objective" measurement ofperformance. Yet the realities of politics in the mea-surement process often mean that measurement willnot be objective. Should we therefore divorce appraisalratings from salary decisions? We think not. Pay forperformance is still a good concept in our view, even inlight of our findings. Attention to the recommendationswe present in this section should minimize the impactof manipulative politics in appraisal ratings.

6. In addition, the number of people who have ac-

cess to the written appraisal should be minimized. Themore people who have access to the appraisal, thegreater the temptation for the rater to "impressionmanage" it. Remember, the fact that the appraisal iswritten down often means that it is less than com-pletely accurate, simply because it is publicly available.

7. The findings of this study have legal implica-tions as well. Organizations are more susceptible to liti-gation involving charges of unlawful discharge or dis-crimination than ever before. Accurate, valid appraisalscan help an organization defend itself; inaccurate, inva-lid appraisals can put the organization at risk. Ofcourse, the .relatively recent practice of extensive docu-mentation of poor performance has been in part a re-sponse to the modern legal climate. Paradoxically, thatclimate has arguably increased the role of politics informal appraisal, as organizations try to maintain legalgrounds for termination decisions. Still, the often polit-ically motivated practice of building a case for dismis-sal via documentation of poor performance has comeunder closer scrutiny as trends in employee appraisalare given closer examination. The best advice here is tostress honesty in appraisal as a "default option" policy.Credible and consistent appraisal practices are the bestdefense against ligitation. T'hus some counseling in thelegal ramifications of appraisal should become part ofexecutive training.

ConclusionPerhaps the most interesting finding from our

study (because it debunks a popular mythology) is thataccuracy is not the primary concern of the practicingexecutive in appraising subordinates. The main con-cern is how best to use the appraisal process to moti-vate and reward subordinates. Hence, managerial dis-cretion and effectiveness, not accuracy, are the realwatchwords. Managers made it clear that they wouldnot allow excessively accurate ratings to causeproblems for themselves, and that they attempted touse the appraisal process to their own advantage.

The astute manager recognizes that politics inemployee appraisal will never be entirely squelched.More candidly, most of us also recognize that there issome place for politics in the appraisal process to facili-tate necessary executive discretion. The goal, then, isnot to arbitrarily and ruthlessly try to eliminate polit-ics but, instead, to effectively manage the role politicsplays in employee appraisal. •

191

Academy of Management EXECUTIVE

Clinton O. Longenecker is an assistant pro-fessor of management in the College of BusinessAdministration at the University of Toledo. Hereceived his B.B.A. in industrial marketing andM.B.A. in management from the University ofToledo, and holds a Ph.D. in organizational be-havior from The Pennsylvania State Univer-sity. Dr. Longenecker has extensive work expe-rience in the construction industry and iscurrently an active trainer and managementconsultant to a variety of different organiza-tions and several government agencies special-izing in team building, organization develop-ment, and performance management systems.He is a staff consultant with the ManagementCenter at the University of Toledo, and hasalso worked in Haiti, managing several mission-ary building projects including school and hos-pital construction. Dr. Longenecker's currentresearch and writing interests focus on manage-ment effectiveness, increasing human perform-ance in the workplace, organization develop-ment, and understanding politics and cognitiveprocesses in the context of the real world. Hehas received three outstanding teacher awardsfrom two universities.

Dennis A. Gioia is associate professor oforganizational behavior at the College of Busi-ness Administration at the Pennsylvania StateUniversity. He received his B.S. in engineeringscience and his M.B.A. and his Ph.D. in man-agement from Florida State University. Prior tohis academic career, he worked as an engineer-ing aide for Boeing Aerospace at the John F.Kennedy Space Center and as a vehicle recallcoordinator for Ford Motor Company in Dear-born, Michigan.

Dr. Gioia's research and writing have fo-cused primarily on the nature and uses of com-plex cognitive processes by organizational mem-bers and the ways in which these processesfacilitate sense making, communication, and in-fluence. His most recent interests have to dowith the less rational, more intuitive, emo-tional, and political aspects of organizationallife (as represented by this article). Recentworks of Dr. Gioia have appeared in the Acad-emy of Management Review, Business Horizons,Journal of Applied Social Psychology, and Orga-nizational Behavior and Human DecisionProcesses, among others. He is also co-editor,with Dr. Henry P. Sims, Jr., of The ThinkingOrganization, a book of original contributionconcerning the dynamics of cognitive processesin organizations, published in 1986 by Jossey-Bass.

Henry P. Sims, Jr., is currently profes-sor of management at George Mason Univer-sity, on leave from The Pennsylvania StateUniversity. Dr. Sims's special area of researchis managerial leadership. He has published over75 articles on this subject, including articles inthe Academy of Management Journal and theAcademy of Management Review. His book. TheThinking Organization, edited with Dr. DennisA. Gioia, was published by Jossey-Bass in 1986.Another book. The Superleader: DiscoveringHow to Lead Others to Lead Themselves,coauthored with Charles Manz, will be pub-lished by Prentice-Hall in 1988. Dr. Sims hasconsulted with many U.S. and multinationalfirms, including Ford Motor Company and theU.S. Agency for International Development.

192

August, 1987

ENDNOTES1. For an extensive discussion of this point, see F.J.

Landy and J.L. Farr's "Performance Rating," PsychologicalBulletin, 1980, 87, 72-107. This issue is further developed inLandy and Farr's book. The Measurement of Work Perform-ance, New York: Academic Press, 1983. It is clear that thepsychometric aspects of the appraisal process are only onepart of understanding and improving appraisals.

2. DeNisi, Cafferty, and Meglino have recently dis-cussed the key issues and complications associated with un-derstanding the psychology of managerial decision making inthe appraisal process in their recent article, "A CognitiveView of the Performance Appraisal Process: A Model andResearch Prospective," Organizational Behavior and HumanPerformance, 1984, 33, 360-396. For a discussion of furthercognitive complications in the appraisal process as a result ofunconscious information processing, refer to D. A. Gioia andP.P. Poole, "Scripts in Organizational Behavior," Academy ofManagement Review, 1984, 9, 449-459.

3. For an exploration of some of the emotional and af-fective factors that might bear on appraisal processes, see 0.S. Park, H. Sims, Jr., and S. J. Motowidlo's "Affect in Orga-nizations: How Feelings and Emotions Influence ManagerialJudgment," in H. P. Sims and D. A. Gioia and Associates(Eds.) The Thinking Organization.

4. Jack Feldman suggests in his article, "Beyond Attri-bution Theory: Cognitive Processes in Performance Evalua-

tion," Journal of Applied Psychology, 1981, 66, 127-148, thatraters have certain cognitive flaws in information processingthat make complete objectivity and validity in rating unob-tainable. Also see W. C. Borman's "Explaining the UpperLimits of Reliabiity and Validity in Performance Ratings,"Journal of Applied Psychology, 1978, 63, 135-144.

5. Jeffrey Pfeflfer, in his book Power in Organizations,Marshfield, MA: Pittman Publishing Co., 1981, makes astrong case that political gamesmanship and the use of powerin organizations surround almost every important decision inorganizational life. The implications of the appraisal process(e.g., pay raises, promotions, terminations) make the ap-praisal of performance an important decision-makingenterprise.

6. Bernardin and Beatty, in their book PerformanceAppraisal: Assessing Human Behavior at Work, Boston,MA: Kent, 1984, suggest that extraneous variables that arenot performance related have an effect on the rater's decisionprocesses and that this influence is in fact a primary sourceof bias and inaccuracy in performance ratings.

7. For an in-depth treatment of the legal issues con-cerning performance appraisal, see P. S. Greenlaw and J.PKohl's Personnel Management, New York: Harper & Row1986, 171-173. See also W. F. Cascio and H. J. Bernardin's"Implications of Performance Litigation for Personnel Deci-sions," Personnel Psychology, Summer 1981, 217.

193