Motivation 1.pdf

7/27/2019 Motivation 1.pdf

1/22

Employee Positive Emotion and Favorable Outcomes at the Workplace

Author(s): Barry M. Staw, Robert I. Sutton and Lisa H. PelledSource: Organization Science, Vol. 5, No. 1 (Feb., 1994), pp. 51-71Published by: INFORMSStable URL: http://www.jstor.org/stable/2635070 .

Accessed: 13/09/2013 12:07

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of

content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

INFORMSis collaborating with JSTOR to digitize, preserve and extend access to Organization Science.

http://www.jstor.org
http://www.jstor.org/action/showPublisher?publisherCode=informshttp://www.jstor.org/stable/2635070?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/2635070?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=informs


2/22

Employee Positive Emot ion a n d FavorableOutcomes a t t h e Workplace

Barry M. Staw * Robert I. Sutton * Lisa H. PelledHaas School of Business, Universityof Californiaat Berkeley,Berkeley, CA 94720Departmentof IndustrialEngineeringand EngineeringManagement,Stanford University, Stanford, CA 94305Departmentof IndustrialEngineeringand EngineeringManagement,Stanford University,Stanford, CA 94305

T his paper contributesto the small but important iteratureon emotions in the workplace.Moving beyond rational and economic views of employees, the authors integrate diverseliteraturesand use archivaldatato showthatpositiveaffecthas favorableoutcomesat work n termsof supervisor valuationandcoworker upport.The impactof positiveand negativeemotions s a richarea of researchabout organizations. RichardL. DaftAbstractThis paper draws on writings in psychology, ociology andorganizational ehaviorto develop a conceptualframeworkthat specifieshow positive emotionhelps employees obtainfavorable outcomes at work. We propose that feeling andexpressingpositiveemotionson thejob have favorable onse-quenceson:(1) employees ndependentof their relationshipswith others (e.g., greaterpersistence), 2) reactionsof othersto employees (e.g., "halo," or overgeneralizationo otherdesirable traits), and (3) reactionsof employees to others(e.g., helping others). These three sets of interveningpro-cesses are proposed, in turn, to lead to work achievement,job enrichmentand a higher qualitysocial context.A partialtest of this frameworks made in an 18-month tudy of 272employees.Results indicate that positiveemotion on the jobat time 1 is associated with evidence of work achievement(morefavorable upervisor valuationsandhigherpay)andasupportive ocialcontext(moresupport romsupervisors ndcoworkers)at time 2. But positive emotion at time 1 is notsignificantlyssociatedwithjob enrichmentat time 2.(Positive Emotion; Employee Success)

Many basic elements of research on job attitudes havechanged in recent years. Instead of simply assessingemployee responses to various aspects of the work role,attention has begun to shift toward how social contextsshape attitudes and needs (Salancik and Pfeffer 1978),

dispositional determinants of affect (e.g., Staw et al.1986), and expressed emotions in organizations (Rafaeliand Sutton 1989). This broader formulation of jobattitudes has opened organizational research to a widerarray of concepts such as positive and negative affect,optimism, depression, liking and happiness. This refor-mulation of job attitude research has also stimulated anew search for the consequences of emotion in theworkplace. In lieu of the usual process of correlatingjob satisfaction with absenteeism, turnover, and perfor-mance, it is now more evident that research can prof-itably examine how emotion influences a wider set ofpersonal and organizational outcomes.The present study is among the first attempts toempirically examine the broader consequences of atti-tudes in the workplace. Our general hypothesis is quitesimple. We propose that employees who feel and dis-play positive emotion on the job will experience posi-tive outcomes in their work roles. This overarchingtheme is used to weave together diverse literature inpsychology, sociology and organizational behavior per-tinent to emotion at the workplace. We focus on ex-plaining outcomes provided by the workplace that arefavorable from the employee's perspective, not theorganization's perspective.Several specific relationships are posited to underliethe general hypothesis that positive emotion yieldsfavorable work outcomes; these are outlined in Figure

1047-7039/94/0501/0051/$01.25Copyright ? 1994. The Institute of Management Sciences ORGANIZATION SCIENCE/VO1. 5, No. 1, February 1994 51

This content downloaded from 86.31.51.223 on Fri, 13 Sep 2013 12:07:20 PMAll use subject to JSTOR Terms and Conditions
http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsp


3/22

BARRYM. STAW,ROBERT I. SUTTON AND LISA H. PELLED EmployeePositive Emotion and Favorable Outcomes

Figure 1 The Link Between Employee Positive Emotion and Favorable Outcomes at the WorkplaceIntervening Processes

Effectson employeePositive Emotion * Greater askactivityand Favorable Employeeat Work persistence Outcomes at Work. Enhanced ognitivefunctioning

|Feltpositive l Others' esponses o employee WorkAchievementemotion l Greater nterpersonalattraction

i . "Halo", vergeneralizationto otherdesirable raits r JobEnrichment. Moreproneto respondfavorably o employee'sExpressed social influenceattempts Supportivepositiveemotion Social ContextEmployee'sresponses o others. Moreproneto help others

1. The model suggests that positive emotion bringsabout favorable outcomes on the job through three setsof intervening processes. First, positive emotion hasdesirable effects independent of a person's relation-ships with others, including greater task activity, persis-tence, and enhanced cognitive functioning. Second,people with positive rather than negative emotion ben-efit from more favorable responses by others. Peoplewith positive emotion are more successful at influenc-ing others. They are also more likable, and a haloeffect may occur when warm or satisfied employees arerated favorably on other desirable attributes. Third,people with positive feelings react more favorably toothers, which,is reflected in greater altruism and coop-eration with others. We hypothesize that the combina-tion of these intervening processes leads to favorableoutcomes in the workplace, including achievement (e.g.,favorable supervisor evaluations and greater pay), jobenrichment (e.g., variety, autonomy, feedback andmeaning), and a more supportive social context (e.g.,support from coworkers and supervisors).'

Before explicating this framework, however, it isimportant to emphasize that we do not contend thattraveling through organizational life in a good mood isa panacea. We take care throughout this paper to pointout conditions under which pleasant emotions maylead to undesirable consequences for employees.Often, however, findings that positive emotion has un-desirable consequences from the organization's per-

spective also imply desirable consequences from theemployee's perspective. Research by industrial psychol-ogists, for example, indicates that positive affect by jobcandidates (Arvey and Campion 1982, Eder and Ferris1989) threatens the validity of selection interviews, andthat the extent to which an employee is likable (Cardyand Dobbins 1986) or satisfied (Smither et al. 1989)threatens the validity of performance evaluations. Froman employee's perspective, these findings suggest thatpositive affect will help him or her to get a job and toreceive favorable performance evaluations.

The Intertwining of Felt and ExpressedPositive EmotionFigure 1 does not distinguish between the conse-quences of felt and expressed positive emotion. We donot make this distinction, in part, because it is sodifficult to measure inner feelings (e.g., researcherstypically infer internal emotions from expressed emo-tions gleaned from questionnaries, interviews and ob-servations). We also do not make this distinctionbecause theory and research suggest that felt andexpressed emotion are closely intertwined. Researchon social desirability (Crowne and Marlow 1964), self-monitoring (Snyder 1974), and emotional labor(Hochschild 1983) do suggest that expressed and feltemotions are conceptually distinct. But a substantial

52 ORGANIZATION SCIENCE/Vo1. 5, No. 1, February 1994



4/22

BARRYM. STAW,ROBERT I. SUTTON AND LISA H. PELLED EmployEmotion and Favorable Outcomes

bodyof literature ndicates that there are strongrecip-rocaleffects betweenfelt an expressedemotions.Wheninvestigatorshave used manipulationspresumedto in-duce positive and negative feelings, they have foundthat such feelings are highlycorrelatedwith both self-reportsand facialexpressions see Ekman 1982,p. 75).There is also evidence showingthat, when organiza-tions try to require members to express an emotionthey do not feel, employeesoften come to internalizesuch requiredemotions Hochschild1983,Sutton 1991).And even when employeestry to feign requiredemo-tions that they do not feel, their inner feelings maynonetheless shape expressedemotions. For example,service employeestend to smile less as externalstressincreases Suttonand Rafaeli1988).Likewise,Ekman's(1985)researchon lyingindicatesthat when people tryto display alse emotions,theirinnerfeelings often leakout in unintendedways.

Felt and expressedemotionsmay be intertwined orboth cognitiveand physiologicalreasons. In terms ofcognition,the expressionof a givenemotion may leadindividualsto solidify their self-perceptionthat theyare feeling in a particularway; in a sense, becomingbehaviorallycommitted(Kiesler 1971, Salancik 1977)to a particular motion.This processmay explainwhyresearch indicates that expressed hostility often in-creases rather than decreases anger toward anotherparty(e.g., Ebbesonet al. 1975,Berkowitz1978),andwhy positiveinterpersonalnteractioncan increase forliking for others (Cialdini 1984). And some researchsuggeststhatexpressedemotions can triggerphysiolog-ical processesthat determine felt emotions.For exam-ple, Zajoncet al. (1989)found that utterancescausingsubjectsto form facial expressionssimilar to a smile(e.g., repeating he letter "e")were associatedwiththemost pleasantmood, and utterancesthat caused sub-jects to form facialexpressions imilar o a frown (e.g.,repeating he letter "u")were associatedwith the leastpleasantmood. Zajoncandhis colleaguespresent phys-iological evidence indicating these effects occur be-causeunpleasantfacialexpressionscause insufficientlycooled blood to enter the brain,whichpeople experi-ence as unpleasant,while pleasant facial expressionsfacilitatethe coolingof blood flowingto the brain. Byextension, these findings suggest that displayingun-pleasant emotions may cause employees to have un-pleasant feelings because they-literally-becomemore hot-headed.Taken together, the literaturesug-geststhat,although elt positiveemotionandexpressedpositive emotion can be distinguished n theory, thepresence of one often implies the presence of theother.

Direct Influence of Positive Emotion on EmployeesFigure 1 proposesthat positiveemotioninfluencestwosets of performanceprocesses: (1) task activity andpersistenceand (2) cognitivefunctioning.These forcesmay lead to beneficial outcomes for the employeewhen actual work performanceis increased. Theseforces mayalso lead to favorableoutcomes when oth-ers, especiallypowerfulpeople who controlincentives,noticethat an employee haswork-related kills.Power-ful people can rewardpersistenceandenhancedcogni-tive functioningwith incentivessuch as pay, favorableperformance atings, nterestingwork,and a moresup-portivesocialcontext.TaskActivity and Persistence. There is evidence thatmost people tend to be optimistic n anticipating uc-cess (Weinstein1980)and also have, "illusionsof con-trol,"believing that they can influenceoutcomesthatare randomlydetermined(Langer1975).Experiencingpositive emotion seems to exacerbate these beliefs.People who are positive in disposition or mood aremore likelyto overestimate heircontroloverthe worldand the outcomes they will receive than those withnegative emotion (e.g., Alloy and Abramson 1979,Alloy et al. 1981,Taylorand Brown 1988). Similarly,people with positive moods or dispositionsare moresubject o self-servingbiases,(tendingto attributeposi-tive consequences to personal rather than externalcauses)than people with negativeemotion(see Taylorand Brown, 1988 and Fiske and Taylor, 1991 for re-views).One recent study (Dunning and Story 1991) hasshownthatpositivepeople actuallydo experiencemorepositive outcomes, disputingthe notion that positivebeliefs are necessarily llusory.Yet, no matterwhetherpeoples' beliefs about the future are accurateor not,one conclusion is relatively certain. Anticipation ofsuccess and thoughtsthat one can bring it about arelikely to facilitate task activityand persistence.Whenpeople believe that their actionswill lead to positiveresults, they are more likely to initiate difficultanduncertaintasks. And, when people believe they havesome degree of control over task success (e.g., self-efficacy), hey are more likelyto persistunder difficultor failing conditions (Bandura1982, 1991).Thus, be-cause positive emotion increases tendencies towardoptimism and perceived control, we would expectgreater task activity and perseverance. Indeed, asTaylorand Brown(1988) point out, the link betweenhappinessand activityhas been suggestedby observersof humanbehaviorgoingback to Aristotle.A studyby Seligmanand Schulman 1986)is illustra-tive of the predictedeffectsof optimismon task persis-

ORGANIZATION CIENCE/VOl. 5, No. 1, February 1994 53



5/22

BARRY M. STAW, ROBERT I. SUTTON AND LISA H. PELLED Employee Positive Emotion and Favorable Outcomes

tence. These researchers divided a sample of 103 newlife insurance agents on measures of dispositional opti-mism, using measures of how people typically attributethe causes of success versus failure rather than simpleassessments of expected outcomes. Seligman and hiscolleagues (e.g., Abramson et al. 1987) have arguedthat people with pessimistic coping styles will construebad events in their lives as resulting from internal ("itis my fault"), enduring ("it will go on forever"), andglobal causes ("it will ruin everything that I do"). Incontrast, people with optimistic coping styles believethat bad events are due to external, temporary, andlocal causes. Results from the life insurance studyshowed that optimists remained on their jobs at twicethe rate of pessimists and sold more insurance thanpessimists. These results are particularly interestingbecause insurance agents repeatedly encounter failure,rejection and indifference from clients-that is theymust work on a task in which persistence is necessaryfor success. Other research on optimism has yieldedsimilar results. For example, college freshmen withoptimistic coping styles were shown to have highergrade point averages than those with pessimistic styles(Peterson and Bennett 1987). Also, controlling for priorlevels of achievement, school children with optimisticstyles subsequently performed better on standardizedachievement tests than students who had pessimisticstyles (Nolen-Hoeksema et al. 1986).Enhanced Cognitive Functioning. Although peoplewho are happy may be more active and persistent ontasks, they may be less accurate in their cognitivefunctioning. Research has shown, for example, that themore depressed an individual is, the less susceptible heor she is to cognitive biases such as overconfidence,self-serving attributions, and illusions of control (e.g.,Sweeney et al. 1982, Kuiper 1978, Alloy and Abramson1979). Such "sadder but wiser" effects have also beensupported by attitude change experiments. Using vari-ous mood manipulations, persuasion studies have foundthat those in positive moods tend to be less sensitive tothe quality of arguments than those in negative affec-tive states. That is, when a person is happy a strongargument is not much more persuasive than a weakargument; but when a person is in a negative mood, heor she is much more likely to be persuaded by a strongthan a weak set of arguments (see Mackie and Worth,1991, Schwartz et al. 1991 for reviews).In opposition to these "sadder but wiser" findings isan extensive stream of research linking positive emo-tion and enhanced cognitive functioning. These "hap-pier and smarter" studies indicate that subjects in-duced to be in good moods are more likely, compared

to those in a bad mood or subjects in a control group,to use efficient and rapid problem-solving strategies(Isen and Means 1983). Other experiments suggest thatsubjects in good moods take greater risks in hypotheti-cal situations or when the chances of winning are high(Isen et al. 1982, Isen and Patrick 1983, Isen et al.1990). In contrast, in situations where the possibilityfor loss is large or salient, positive affect appears tolead to cautious behavior (Isen and Geva 1987, Isenet al. 1988). Thus, positive emotion may foster deci-sion-making designed to maintain one's positive state-taking risks when success appears likely, yet beingcautious when the odds appear slim.There has also been research on the connectionbetween mood and creative problem-solving. Positiveaffect appears to promote more connection and inte-gration of stimuli. Subjects induced to be in goodmoods generate a broader range of associations withcommon words, recall longer lists of words that arerelated to one another, and are more likely to solveproblems that require seeing a broader set of potentialrelationships among the elements composing an issue(Isen and Daubman 1984, Isen et al. 1985, Isen et al.1987). Furthermore, research on negotiation byCarnevale and Isen (1986) indicates that people ingood moods are more likely to reach integrative ratherthan compromise solutions. Because integrative solu-tions require more creativity, Carnevale and Isen assertthat this finding shows that positive affect promotescreativity during negotiations.It may be possible to reconcile the "sadder butwiser" and "happier and smarter" literatures by notingthat the consequences of positive and negative affectcan depend on the task involved. For example, if thetask requires cognitive processing that is rapid, basedon heuristics and broad integrative categories, thenindividuals with positive affect may have an advantage.In contrast, when a task requires tighter informationprocessing, with more careful attention paid to detailedarguments and data, then people with negative affectmay have an advantage (Schwartz and Bless 1991).Although such a contingency approach to affect makesconceptual sense, it has not yet been upheld in anydirect tests. Staw and Barsade (1992) used a three-hourmanagerial decision exercise to see whether affectivedisposition would predict performance in a detaileddecision-making task. They gathered data on whetherMBA students with negative emotion were more care-ful in their decisions, used more information, andrecognized the interrelationships of various decisionelements. None of the "sadder but wiser" hypotheseswere upheld. Those who were positive in emotion

54 ORGANIZATION SCIENCE/VO1. 5, No. 1, February 1994



6/22

BARRY M. STAW, ROBERT I. SUTTON AND LISA H. PELLED EmployEmotion and Favorable Outcomes

made better decisionsand displayedmore evidence ofaccurate nformationprocessing n the managerial im-ulations. Given these results,one might ogicallyexpectthat positive emotionwill, on balance, enhance cogni-tive functioning n organizational ettings.Responses of Others to EmployeesFigure 1 proposes that positive emotionhas three setsof effects on employees'relationshipswithothers. First,employees with positive emotion will be viewed asmoreinterpersonally ttractive.Second, employeeswhohave positive affectwill tend to be rated by others ashaving a wide range of desirable traits, even whenothers lack informationabout such traits. Third,morelikable people have more success at wielding socialinfluence over others. Taken together, these patternssuggestthat people with positiveaffect will be viewedby others as more deservingof incentives and supporton the job.InterpersonalAttraction. The most direct evidencethat people withpositiverather than negativeemotionare moreinterpersonally ttractive s foundin researchon depression.Coyne(1976) conducted an experimentin which 45 undergraduatestudents were each ran-domly paired with a strangerand engaged in a 20-minute unstructured elephone conversationwith thestranger.The set of strangerscomprised15 depressedpsychiatric outpatients, 15 nondepressed psychiatricoutpatientsand 15 controls. Comparedto subjectsinthe other two groups,those who talked to depressedoutpatientswere less willingto engagein futureinter-actionswith the strangers.Subjects n the experimentalgroup also reported that, after the interaction, theywere more depressed, anxious,and hostile than sub-jectswho talked to nondepressedpatientsand controls.The literature on social supportalso providessomeindirect evidence that positiveemotion leads to inter-personalattraction.Cross-sectionalieldstudies consis-tentlyhave found positive relationshipsbetween socialsupportand indicatorsof pleasantemotion such as lifesatisfaction, elf-esteem, ack of depression,andlackofanxiety(e.g., Caplanet al. 1980,House 1981,Gansteret al. 1986). These findings usually are portrayedasevidence that people who receive aid, affirmationandemotionalsupportfrom others use these resources toenhance their mental well-being.But these data mayalso reflect the opposite causal relationship:positiveemotion may lead others to provide social support.Those who are emotionallypositive may be viewed asmore interpersonallyattractive.Providing emotionaland tangible assistance to a happy person provokesgood feelings in the helper, who will then seek the

experienceagain. In contrast,providing upport to anunhappyperson may provoke unpleasantfeelings inthe helper, who may then shun the unhappy person.Evidencefor this effect is providedby a recent longitu-dinal field study of 486 men by Vinokur et al. (1987).They reported that people who had a generalizednegativeoutlook on life (e.g., lacked self-esteem andresented others)at time 1 later reportedreceiving esssocial supportfrom significantothers, and that theirsignificantothers (wives and close friends) reportedgivingthem less subsequentsupportas well.Halo. The notionthat people withpositive emotion(and who are more likable) have an advantagewhenobservers evaluate their other traits is an old andpersistenttheme in the social sciences. Asch's (1946)classic researchon impression ormation uggested hatthe warm-colddimensionwas a central trait that col-ored the perceptionsof numerousother traits, despitea complete absence of informationabout those traits.Two groupsof subjectswere each presenteda list of aperson'straits; the lists differedonly in that the word"warm" appeared in one list and the word "cold"appeared in the other list. Pronounced differencesbetween the "warm"and "cold"groupsincluded sub-jects ratingsof the hypotheticalperson as "generous"(91% versus 8%), "wise" (65% versus 25%), and"good-natured"94%versus17%).Asch'sfindingsareoften interpretedas an example of halo error (e.g.,Cooper 1981), a widely observed effect that occurswhen one salient feature of a personbeing evaluatedcolorsjudgmentsmadeaboutthatpersonacrossa widerangeof dimensions.The notion that employees' positiveemotion causeshalo error when others evaluate their performance sreflectedin a wide range of research.A correlationalstudyby Alexander and Wilkins(1982)found that theextent to which supervisors iked subordinateswas astronger predictor of supervisors'performance ap-praisals than objectiveindicatorsof subordinateper-formance.Similarly, everalexperimental tudiesfoundthat indicatorsof ratee positiveaffect includingsatis-faction (Smither et al. 1989), lack of nastiness andcoldness (Krzystofiaket al. 1988), as well as liking(Cardyand Dobbins1986),led othersto make inflatedevaluationsof ratee performance.The halo effect also is implied by literatureon selec-tion interviews.Typically,severalvideotapesof simu-lated job interviewsare presented to subjects, eachwith varying evels of displayedpositiveemotion suchas smiling,hand gesturingand eye contact.These ex-periments consistently reveal that interviewees whodisplaygreaterpositiveemotionthroughnonverbalbe-

ORGANIZATION SCIENCE/VO1. 5, No. 1, February 1994 55



7/22

BARRY M. STAW,ROBERT I. SUTTON AND LISA H. PELLED Employee Positive Emotion and Favorable Outcomes

havior are more likely to be rated as desirable employ-ees (e.g., Imada and Hakel 1977, McGovern and Tins-ley 1978, Rasmussen 1984). Forbes and Jackson (1980)report a similar finding in their field study of applicantsfor engineering apprenticeships.The halo effect often demonstrated in both the per-formance evaluation and selection literatures does notmean that positive emotion can always substitute forobjective qualifications or performance. No doubt, thedisplay of positive emotion combined with very poorcredentials can be interpreted as a sign of ingratiationby applicants. Likewise, positive emotion combinedwith performance that is viewed as deficient, couldmake the employee appear even weaker to the evalua-tor. These caveats notwithstanding, positive emotion isusually a beneficial contributor to performance andemployment evaluation. Logically, one would expecthalo effects to be most prevalent in cases where thereare not many objective differences in performanceamong candidates and where performance indicatorsare inherently ambiguous: two prevalent conditions inwork organizations. Although the applied psychologyliterature treats positive emotion as a source of distor-tion in performance evaluations, the focus here is onthe consequences of positive emotion for the individ-ual. From the employees' vantage point, any increasein evaluated performance due to positive emotion con-stitutes a favorable or successful outcome at the work-place.Social Influence. As we noted, the literature oninterpersonal attraction indicates that people who havepositive affect are more likable than those with nega-tive affect. Writings on social influence suggest, in turn,that people who are more well-liked are more success-ful at persuading others to comply with their requests(e.g., Drachman et al. 1978). In summarizing this re-search, Cialdini (1984) argues that the use of liking toproduce assent from others is so widely known thatsome sales jobs are designed so that occupants willonly sell products to people who already know and likethem (e.g., at Tupperware parties). Similarly, whenthey must sell products such as cars or encyclopedias tostrangers, Cialdini contends that the salespersons'"strategy is quite direct: They first get us to like them"(p. 165). As an extreme example, Cialdini describes JoeGirard, listed in the Guinness Book of World Recordsas the world's greatest car salesman, who regularly sentout 13,000 greeting cards each month to customers andpotential customers. The greeting on the front variedwith the season, but the message inside was alwayssimply: "I like you."Likeability may not be the only way that people with

positive emotion wield social influence. As we havelearned from research on social information processing(Salancik and Pfeffer 1978, Zalesny and Ford 1990),positive and negative emotions are infectious, spread-ing from one person to another. And when a person isin a positive emotional state, he or she may be highlysusceptible to social influence. As noted above, experi-ments have shown that, when a person is in a goodmood, he or she may not pay close attention to thequality of arguments in a message, with the resultbeing that weak arguments can be just as persuasive asstrong arguments for those in positive as opposed tonegative emotional states (see Schwartz et al. 1991).Thus, people may comply with more requests frompeople with positive (rather than negative) emotion forseveral reasons. First, people who display positive emo-tion may put others in a good mood, making themsusceptible to weak as well as strong arguments inpersuasion attempts. Second, as Fiske and Taylor (1991)note, people who are in good moods take steps tomaintain their pleasurable state. Coworkers and supe-riors may therefore say "yes" to positive employees inorder to encourage further pleasant interactions withsuch employees. Finally, the norm of reciprocity(Gouldner 1960) provides a third explanation for thissocial influence. Positive emotion may, in itself, beviewed as something of value that an organizationalmember gives to others. As a result, others may feelobligated by the rules of exchange to reciprocate bysaying "yes" to their requests. Regardless of the the-oretical mechanism, we can infer from the literature onsocial influence that-all other things beingequal-employees who have more positive emotionmay be more successful when they make requests forhigher pay, more interesting work, and other desirableoutcomes available at the workplace.Employees'Reactions to OthersHelping Behavior. Experimental research on altru-ism has consistently shown that people who are in-duced to be in positive moods are more likely to behelpful to others. Subjects who experience success attasks are more likely to help others (e.g., Berkowitzand Conner 1966, Isen 1970, Isen et al. 1973), as aresubjects who find a dime in a telephone booth (Isenand Levin 1972), or are given free stationery (Isen et al.1976). In summarizing this research, Isen (1984) con-cludes that positive affect consistently brings aboutgreater sociability and benevolence.There is not yet firm agreement about the explana-tion for this "feel good, do good" phenomenon. Oneexplanation is that being in a good mood is reinforcing,




8/22

BARRY M. STAW, ROBERT I. SUTTON AND LISA H. PELLED EmployEmotion and Favorable Outcomes

andhelpingothers is a form of self-reward hatenablesa personto maintainthis pleasurablestate (Fiske andTaylor 1991).Another explanation s that people whoare in good moods are generally more attracted toothers(e.g., Gouaux1971,Mehrabian nd Russell 1975,Bell 1978). As the literature on social influencesug-gests, people are more willing to help people they likecompared o those they dislike.The implication s that,comparedto employees who are usually grouchy,em-ployees in good moods are more helpful to othersbecausetheyfindthemselves ikinga largerproportionof people encounteredat the workplace.2The hypothesisthat positive emotion leads to morealtruism also is suggested by the emerging body oftheory and research on organizational itizenshipde-veloped by Organand his colleagues (see Organ1988).Organizational itizenship reflects contributionsmadeby employees at the workplacethat go beyond formalrole expectations.Organ summarizesa series of cross-sectional studies indicating that job satisfaction isamong the most robust predictorsof organizationalcitizenshipbehaviors, even when self-reports of jobsatisfactionare correlatedwith independentreportsofcitizenship. George and Brief (1993) make a similarargumentthat a positive mood at work is likely tocontribute to organizationalspontaneity. They notethat positive emotion can be linked with helping co-workers, protecting the organizationand spreadinggoodwill.Such contributionsby the employeemay bereciprocatedby others in the organization hrough heallocation of more favorableperformanceevaluations,higher pay, desirable tasks and both supervisorandpeer support.

HypothesesOur review of the literature,summarized n Figure 1,shows that one can expect positiveemotion to have arangeof favorableresults.As we noted, much of theexistingliteratureshows that positive affect has bene-ficialconsequences n terms of cognition, nterpersonalattraction,and helpingbehavior.Yet this prior litera-ture has been relativelysilent on consequencesat theworkplace from the individual'spoint of view. Weknow relatively ittle about whether positive emotiontranslates nto greaterachievement,better job assign-ments, and a richersocial environment or individualsworking n organizational ettings.Such questionswillbe the mainfocus of the presentresearcheffort.One might expectthatcognitive mprovements suchas greater persistenceand creativity)will be reflectedin greater achievementby employees. For example,if

positive emotion improves cognitive functioning,onemight also expect parallel improvements n perfor-mance, if the job entails cognitive requirements.Ofcourse,performance n the organizational ontext candepend as much on interpersonal s cognitive unction-ing (e.g., the ability o get alongwith others may be asimportant as doing good individual work). In suchcases, positive emotion might contribute to achieve-ment via interpersonal unctioning.Finally, regardlessof actualperformance, ne can expect that the displayof emotion will influence the assessment of achieve-ment and allocationof rewards by others. Prior re-search,for example,has shown that positiveaffect is asystematicsource of halo in performanceevaluations(Cardyand Dobbins 1986, Smitheret al. 1989). Thereis also experimentalevidence that subordinateswhoengage in ingratiationby acting friendlyand offeringcomplimentsare morelikelyto receive payraisesfromsupervisors han subordinateswho do not engage iningratiation Kipnis and Vanderveer 1971). Althoughsuch effects are often considered a source of bias orerror,especially n the literatureon performance val-uations,we consider hese improvementsn ratingsandrewards as achievements n their own right, at leastfrom the individual'spoint of view. Thus, due to thesubjectiveas well as objective influences of positiveemotion,we propose:

HYPOTHESIS 1. Employees with greaterpositive emo-tion will receive more favorable evaluations of theirperformance and higher pay.In addition to rated performanceand pay, anothervaluedoutcomeis the work itself. Job designresearchhas shown that most people desire an enriched jobwith characteristics uch as variety,significance, den-tity, feedback and autonomy (Hackmanand Oldham1980). In addition, Oldham and his colleagues haveshown(Oldhamet al. 1982, Oldhamet al. 1986)thatjob characteristicsan operate like othermoretangiblerewardsn creating atisfaction r dissatisfaction mongemployees. Workers apparentlycompare themselveswith others on job dimensions and can experiencedeprivationwhenothers have more enriched obs. Like

money and other readily recognized resources (Foa1971), an enriched job may have both personal andsocialmeaning.It maybe a symbolthat one's positionin the organization valued by others (Salancik andPfeffer1978).Positive emotion can be expected to influence jobenrichmentthrough three principal routes. First, ifpositive emotion leads to an actual improvement n

ORGANIZATION SCIENCE/VOl. 5, No. 1, February 1994 57



9/22

BARRYM. STAW, ROBERT I. SUTTON AND LISA H. PELLED Employee Positive Emotion and Favorable Outcomes

work performance, then a more difficult or complex jobassignment might logically follow. Second, the samelogic applies to any positive halo or bias resulting froman employee's positive emotionality, because it is theperception (not the reality) of an employee's abilitythat determines a supervisor's view of whether theperson can handle a job assignment. Third, the litera-ture leads us to expect that positive employees aremore successful in persuading supervisors and othersin the organization to give them more interesting work.Because they are well liked, supervisors may offerpositive individuals better assignments and yield morereadily to their requests for more enriched jobs. Thus,we propose:

HYPOTHESIS 2. Employees with greater positive emo-tion will receive more enrichedjobs.Our third hypothesis concerns the social context ofwork. Employees with positive emotion may be morelikely to receive social support, since their supervisorsand coworkers use psychological and technical assis-tance as rewards for good performance. For example,if those with positive emotion are likely to be betterperformers in terms of persistence or decision making,then both superiors and coworkers may offer moresupport to such people. Likewise, if those with positiveemotion are more helpful in work situations, thencoworkers and supervisors may "repay" such citizen-ship with supportive actions. Finally, positive employ-ees may receive more social support simply because

interaction with them is more reinforcing than withnegative employees. Thus, we contend that others aresupportive of positive individuals because of the re-warding qualities of positive emotion itself, and be-cause support serves as compensation for valued ac-tions contributed by those in a positive state.We treat social support as a favorable outcome forthe employee for several reasons. People who receivemore tangible and emotional assistance from their su-pervisors are, by definition, engaging in more frequentinteraction with higher ranking members of the organi-zation, suggesting that they are valued by those inpower. Social support from coworkers is also a valuedwork outcome since employees can use it to protectthemselves from occupational stress and health prob-lems (Cobb & Kasl, 1977; House, 1981). Finally, be-cause most people value social interaction on the job(Locke, 1976), social support is by itself a positiveconsequence of working, a reward that may often rankalongside money and the task as important personaloutcomes from employment. Thus, we propose:

HYPOTHESIS 3. Employees with greaterpositive emo-tion will receivemore social support rom supervisorsandcoworkers.An ExploratoryTestAs we have noted, positive emotion may have a num-ber of direct and indirect consequences. Most of theprior literature has outlined relationships betweenemotion and what we have labeled intervening vari-ables in Figure 1. These proximal or intermediaterelationships (perhaps with the exception of some as-pects of cognitive functioning) have been validated byenough prior research as to be relatively uncontrover-sial. In contrast, there has been little research linkingaffect to the more distal outcomes of employees. Thus,the present study will address the three major hypothe-ses concerning the relationship between positive emo-tion and employee outcomes.Much theory and research suggests that job perfor-mance may cause job satisfaction (Petty et al. 1984).Similarly, laboratory researchers have found that suc-cess can be used to induce positive mood among sub-jects (Isen and Shalker 1982, Weiner and Graham1984). As a result, this test of the framework presentedhere uses a longitudinal design in which positive emo-tions were measured well before indications of favor-able outcomes suggesting success at the workplace.In testing the three hypotheses outlined above, anideal research design might not only examine prior andsubsequent levels of the consequences of emotion, butalso the intervening variables underlying the relation-ships between emotion and employee outcomes. Thestudy that follows is more limited and exploratory. Weuse a data set that contains good measures of bothwork outcomes and emotionality, but unfortunatelydoes not allow a test of the intervening processes. Theanalyses that follow should thus be viewed as testingthe plausibility of the model in Figure 1 rather than itsaccuracy. If relationships between positive emotion andemployee outcomes can be established, then the pro-cesses underlying such relationships become plausible,if not proven. The contribution of this preliminary testis that it underscores the value of subsequent research.If overall relationships are discovered between emotionand employee outcomes, it then becomes important tostudy the relative contribution of each of the processesunderlying these effects.MethodsSampleThe present study uses data collected by The Univer-sity of Michigan's Survey Research Center for a study

58 ORGANIZATION SCIENCE/Vol. 5, No. 1, February 1994



10/22

BARRYM. STAW, ROBERT I. SUTTON AND LISA H. PELLED EmployEmotion and Favorable Outcomes

of "Effectiveness in Work Roles: Employee Responsesto Work Environments" (Quinn 1977). Data were col-lected at two times, separated by 18 to 20 months.Time 1 data were gathered in Winter 1972. Time 2data were gathered in Fall 1974. The panel sample forboth periods included 272 employees of three midwest-ern organizations: a hospital and two manufacturers ofautomobile accessories. The sample was composed of152 men and 120 women. The mean age was 37.58. Forfurther details about the sample, see writings by Glicket al. (1986), Gupta and Beehr (1982), Jenkins et al.(1975), and Quinn (1977).Data SourcesA primary aim of the Effectiveness in Work Rolesstudy was to use multiple methods to measure at-tributes of and responses to work. The present studyuses data gathered with four different methods. First,we used data collected through face-to-face interviewswith employees. Professional interviewers from theSurvey Research Center staff administered these inter-views, typically at respondents' homes. The interviewquestions and format were based on the interviewsused in three national studies of the quality of employ-ment (see Quinn and Staines 1979). The interviewincluded a variety of methods to gather self-report datafrom respondents including closed-ended and open-ended questions that required oral responses, and sev-eral brief questionnaires. The interview also includedcard sorts, a method in which cards with questions orstatements are sorted into piles representing differentresponse anchors. Second, at the end of each interview,the interviewer recorded his or her observations ofseveral characteristics of the employee, including gen-der and apparent intelligence.Third, structured field observations were made byobservers while each employee performed his or herjob. These observers had two days of intensive training.At time 1, each employee was watched at work by atleast two different observers on two different occasions.At time 2, only 147 of the 272 employees who partici-pated were watched because no time 2 observationswere made at one of the automotive plants. Of these147 employees, 100 were watched by a single observer;the remaining 47 were watched by three observers,once by a single observer and once by a pair of differ-ent observers who watched simultaneously. Employeeswere watched by different observers at time 1 and time2. Each observation period lasted between 60 and 90minutes. The observer recorded information about par-ticipating employees and their jobs in a structuredobservation booklet. In all, more than 1,500 hours of

structured observations were made during the study.Jenkins et al. (1975) report that these observationshave satisfactory inter-rater reliability. Further descrip-tions of this methodology can be found in Jenkins et al.(1975), Glick et al. (1986), and Quinn (1977).Fourth, and finally, a supervisor rating form wasdeveloped for use in the Effectiveness in Work Rolesstudy. Employees were evaluated on eight aspects oftheir work behavior at both time 1 and time 2. This onepage instrument was completed by the employee's im-mediate supervisor and then mailed directly to theSurvey Research Center.MeasuresThe predictor, dependent and control variables used inthis study are described below. The Appendix presentsa complete list of the items that compose each scale,along with information about the number of responseanchors, data source, and scoring (positive or negative)for each item.Predictor Variable. Our conceptual approach tomeasuring emotion follows the assertion of Zajoncet al. (1989) that "Despite disagreement about thetaxonomic boundaries of emotion labels, there is virtu-ally full agreement about one important fact-emo-tions can be discriminated from each other quitereliably according to their positive-negative hedonicpolarity. Many theorists consider this polarity to be afundamental feature of all emotions." (p. 401). Zajoncet al. note further that "A focus on the one dimensionabout which there is general consensus-hedonic po-larity-might well be most fruitful for research at thistime" (p. 412, 1989). In this vein, rather than focusingon subtle differences in taxonomic boundaries, we em-phasize similarities among the lessons researchers havereported under a wide range of labels for pleasant andunpleasant emotions. This emphasis on integrationrather than differentiation is most useful for develop-ing and testing general theory about the consequencesof positive emotion at work. Such integration frees usfrom taking constant digressions to explain largely mi-nor differences between concepts such as optimism,satisfaction, pleasantness, happiness, absence of work-related depression, high positive emotion and low neg-ative emotion. Thus, although some recent literaturehas drawn finer distinctions between aspects of positiveemotion, especially between positive and negative af-fect (Watson and Tellegen 1985), we treat positiveemotion as a single dimension to predict employeeoutcomes.Our major predictor variable, positive emotion atwork, was operationalized by a composite scale mea-




11/22

BARRY M. STAW,ROBERT I. SUTTON AND LISA H. PELLED Employee Positive Emotion and Favorable Outcomes

suring the extent to which employees felt and ex-pressed positive emotion on the job at time 1. Toconstruct this scale we first used a modified and re-verse-scored version of Quinn and Shepard's (1974) 10item work-related depression scale. The ten items inQuinn and Shepard's depression scale were drawnfrom the 20 item Zung Self Rating Depression Scale(Zung 1965). Quinn and Shepard reported that this 10item subscale correlates 0.95 with the original scale.We then added two other self-report items to thisscale. The first asked the extent to which the employeefelt happy or sad at work. The second asked employeesto report their level of pep and energy on a seven-stepladder. We included this item because feeling energeticand alert rather than sluggish indicates absence ofdepression. In addition to the 12 self-report items, weincluded three items from the trained observers indi-cating how often the employee smiled, laughed or saidsomething funny. These observational data were addedbecause they provided an independent assessment ofthe employee's positive emotion. Combined together,these 15 items formed a reasonably reliable scale(Cronbach's alpha equaled 0.74), designed to assesspositive emotion in the work role. The scale assessedthe tendency to feel and display positive rather thannegative emotion on the job at time 1.Dependent Variables. The dependent variables inthis study included two measures of work achievement(supervisor evaluations at time 2 and pay at time 2),one measure of job enrichment (job characteristics attime 2), and two measures of the employee's socialenvironment (supervisor social support at time 2 andcoworker social support at time 2).The same instrument was used to measure supervisorevaluation at both time 1 and time 2. Supervisor evalu-ation at time 1 was used as a control variable in thelongitudinal analysis to help partial out the effects ofthe supervisor's opinion at time 1 on his or her opinionat time 2. Because of employee turnover and changesin supervisors, complete evaluation forms at both time1 and time 2 were available for only 60 employees.Supervisors were asked to rate their subordinates oneight dimensions: work quality, work quantity, creativ-ity, lateness, dependability, affinity for working, desirefor responsibility, and getting along with others. Theeight items on this rating form were combined into asingle index. Cronbach's alpha at both time 1 and time2 was 0.92.Pay from the job (rather than income from allsources) was measured on the interview with the fol-lowing question: "How much does your income fromyour job figure out to be a year, before taxes and other

deductions are made?" The same measure of pay wasused on the interview at time 2. Pay at time 1 was usedas a control variable in the analysis. Archival informa-tion at time 2 was not available for use in this study.However, using the archival data available at time 1,Gupta and Beehr (1982) conducted a study of thecorrespondence between archival and self-report paydata. They reported that the correlation between thetwo indicators of income was 0.71.The measure of job enrichment used data gatheredby the trained observers. This measure was based onscales developed by Jenkins et al. (1975) and Glicket al. (1986) in prior studies using the Effectiveness inWork Roles data set. This measure operationalizesfour dimensions identified by Hackman and Oldham(1980) as features of motivating jobs: task feedback,variety, autonomy and meaning. Task identity, the fifthof Hackman and Oldham's (1980) proposed list of jobcharacteristics, was not included because the observa-tional data did not include measures of task identity atboth time 1 and time 2. Items measuring these four jobcharacteristics were combined to form a measure ofthe overall motivating potential of each employee's job.Cronbach's alpha was 0.96 at time 1 and 0.90 at time 2.Supervisor support is the extent to which an em-ployee receives emotional and tangible assistance fromhis or her immediate supervisor. The same scale wasused to measure supervisor social support at both time1 and time 2. Supervisor social support at time 1 wasused as a control variable in the analysis. This scaleincluded three items from the supervisor support scaledeveloped by Beehr (1976). The fourth item asked howtrue it was that the employees' supervisor went out ofhis/her way to praise good work. Cronbach's alphawas 0.84 at time 1 and 0.79 at time 2.Coworkersupport is the extent to which an employeereceives emotional and tangible assistance from themembers of his or her work group. The same scale wasused to measure coworker support at both time 1 andtime 2. Coworker support at time 1 was used as acontrol variable in the analysis. This five-item scale wascomposed of three items adapted (and modified) fromSeashore (1954) and two items developed for the Ef-fectiveness in Work Roles study. Cronbach's alpha was0.76 at time 1 and 0.68 at time 2.Control VariablesIn addition to measures of the five dependent variablesat time 1, the other control variables used in this studywere education, age, gender and rated intelligence.These variables were all measured at time 1. Age wasmeasured during the interview. We controlled for age




12/22

BARRYM. STAW,ROBERT I. SUTTON AND LISA H. PELLED EmployEmotion and Favorable Outcomes

because it is related to job satisfaction (Janson andMartin 1982), and thus may be related to generalaffectiveresponseson the job. Moreover,a meta-analy-sis by Waldmanand Avolio (1986) suggeststhat super-visors tend to give lowerperformanceratings to olderemployees. Education was measured through self-report data from the interview.We controlled for thisvariable because more educated employees may beviewed more favorably by their supervisors and co-workers,and be paid more, regardlessof the employ-ees' felt andexpressedemotion. Genderwas measuredthrough the observationsrecorded at the end of theface-to-face interview. We controlledfor gender be-cause research suggests that women convey morewarmth han men (Deaux 1985).We used a one-item measure of rated intelligencedevelopedby Quinnand Shepard 1974).This five-pointscale was completed by the professional interviewerfrom the SurveyResearchCenter after the 90 minuteinterviewwith the employee.Quinn and Shepard 1974)reported that this one item scale correlated 0.72 withintelligence as measured by the Ammons Quick Test(Ammonsand Ammons1962, Trauband Spruill 1982).We controlledfor rated intelligencebecause employ-ees who appear to be intelligent may also be morelikely to receive higher ratings and pay from theirsupervisors.

ResultsTable 1 reports means,standarddeviations,and inter-correlations or all measuresused in this study.The three hypothesesreflected our general assump-tion that positiveemotionleads to favorableoutcomesfor employees.Our longitudinaldesign enabled us toconduct analysesconsistentwith this causal assump-tion. We examinedthe relationshipbetween positiveemotion on the job at time 1 and five favorableout-comes on the job at time 2: 18 to 20 months ater. Yet,even using this longitudinaldesign, an alternativeex-planationis that a relationshipbetween positive emo-tion and a favorable outcome occurs because of theenduringeffects of that favorableoutcome on positiveemotion. In order to help control for the prior andconcurrenteffects of favorable outcomes on positiveemotionat time 1, we used a multiple regressionpro-cedure similarto that employed by Nolen-Hoeksemaandher colleagues (1986)in theirlongitudinal tudyofthe effects of depression and explanatory style onachievement n school children.For each of the fiveequations predictinga favorableoutcome variable attime 2, we introduceda measure of that outcome at

time 1 as a control. If positive emotionat time 1 was astatistically ignificantpredictorof a favorableoutcomeat time 2 over and above the effects of that dependentvariable at time 1 (and of the other four control vari-ables), then support or our underlying ausalassump-tions would be found (Pedhauzer1982).It could be argued that the 18 to 20 month timeintervalof this study was not theoreticallyideal fortestingthe effects of positiveemotionupon workout-comes. Logically,a 6 to 12 month intervalwould haveallowed enough time for pay, task design or socialsupport o have been influencedby positive affect.Thelonger interval of this study probably increased thenumber of exogenous and random influences on thedependent variables,making t harder to find any sig-nificant effects of emotion.Thus, one should view theresults that follow as a fairly conservative est of theresearchhypotheses.

The first hypothesis was that employees who hadmore positive emotion at time 1 would receive morefavorableevaluations rom their supervisors nd higherpay at time 2. The longitudinal indings presented inTable 2 support this hypothesisfor both supervisorevaluationsand pay. Positive emotion at time 1 had afairly strong effect on supervisor ratings at time 2(Beta = 0.31, p < 0.01, one-tailed) and had a weak,but significant, ffect on pay time2 (Beta = 0.05, p


13/22

C\l

C\l I OI- C\lC

C?J 01

0 C5 t

l ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~o' 5 O'o) N

10O';- C 00

(lo00o o 00

0C\ C\J CO O- 10

O~~~~~~~~~~~~~~~~~~~~~~ (l (l N (D m It0) 0 0 0 6 0~~~~~~~~~~~~~~~~~05O'C 5 3OII I

> 0 006 666\ l

0) (D5 C C\l C C O0

0 co_0 10 CO C\l O O O O

n~~~~~~~~~~~c (. co C\ (. rl "I co co m co co "0) ) 0 C\l C\l-O C\OX t I ~~~C\l C\l C\l O O O -~

Cl) 10 CMO C\l C -C\lo ( 6 6 6 6 6 6 6 6 6o o

co D CO C\l (D [ - 0 0 CO0)C0 COO C\l S 0 CO 0D O C\lO 0C\l~---

h co CO Cl_ 10) rl Cm C_ CO l-CO0)10oCO CO sO C\lC\lO- C\l O -,O

0O.0 ['- 10 S- ~- ~ -~co cOL0 O~~ c~~~~ ~~~~-~~~ 6 6 6 6~~~~~~~~~0~~~~~~~~~~~~~~~~~~~~~~~~~~~0c0 0 CO 0 (D\OC\l-COO0 U *O10 10 coc\CO cC co - (Ci6c t Oa c> CD- CO

C o o c )cn cn E E~~~~~~-n - , t t -CCo 0D O ,to " DE Ea a 0 ~~~~~~0)\(U Q W) UL) C/) C1 ) C ) = T- Q

F 2c,2 a- a- a- NO 1,e bruar)0 nJ 0 0 0 C -TJ) )C(U ~~~~~~-~c~j CI5 1f i6 c6 r-- c6 i I C\, CI LO



14/22

BARRY M. STAW, ROBERT I. SUTrON AND LISA H. PELLED EmployEmotion and Favorable Outcomes

Table 2 Work AchievementResults of longitudinal multiple regression: Beta weightsof positive emotion, workachievement variables, and controlvariables at time 1 as predictors of work achievementat time 2

Predictors Supervisor Evaluation Pays(n = 191)Positive Emotion 0.31*** 0.05**Dependent Variable

at Time 1 0.38*** 0.94***Education 0.14 0.05*Age 0.23 -0.02Gendera -0.04 0.02Rated Intelligence 0.08 0.01Adjusted R Square 0.41*** 0.90***aGender is coded: 0 = males, 1 = females.***p < 0.01.** p < 0.05.

p


15/22

BARRYM. STAW,ROBERT 1. SUTTON AND LISA H. PELLED Employee Positive Emotion and Favorable Outcomes

knowledge about the consequences of job attitudes canbe enhanced by going beyond traditional correlates ofjob satisfaction, such as absenteeism and turnover. Thefindings also suggest that affective measures other thanjob satisfaction can be useful for predicting importantindividual outcomes. We used theory and research onconstructs related to positive emotion such as opti-mism, positive and negative affect, and depression toidentify a wider range of psychological and social pro-cesses likely to bring favorable outcomes to the em-ployee. This wide-ranging literature, along with thepreliminary empirical support found for the modelproposed here, suggests that an expanded search forthe consequences of emotion should continue alongseveral lines of inquiry.We examined only five dependent variables as indi-cators of the three categories of favorable outcomes atwork. The effect of positive emotion on a wider rangeof outcomes might be examined. For example, ourarguments suggest that, all other factors being equal,employees who feel and express positive emotion onthe job may receive more frequent and more substan-tial promotions, may generate more favorable impres-sions among clients, may receive more outside offersfor employment, and may be trusted more by peers andsupervisors. Further, although the model presentedhere focused on the link between positive emotion andoutcomes that are favorable to the employee, many ofour arguments could easily be extended to explainoutcomes that are also favorable to the organization.Certainly, the literature that shows people with positiveemotion working more persistently, more quickly, andwith more creativity implies a direct link between affectand organizational outcomes. However, as Staw et al.(1986) note, a crucial moderator may be whether thework role can be best served through enthusiasm andaction versus a more critical (and perhaps negative)posture. This may be an important dilemma for futureresearch to solve empirically.The model developed here suggests that organiza-tional researchers may also have devoted insufficientattention to the influence of emotion on social rela-tions at the workplace. Job satisfaction research typi-cally considers how employees act in isolation in thework context. In contrast, the emphasis here on thebroader literature on positive emotion, along with ourfindings that employees with positive emotion receivedgreater support from supervisors and coworkers 18months later, suggests that emotion shave a significantinfluence on peers and superiors. This research sug-gests that positive emotion may bring about tangiblerewards, at least in part, through its effects on relation-

ships with others. It is, of course, not yet clear whetherpositive emotion yields interpersonal benefits becauseof real or imagined contributions by the individual.Additional research is needed to examine whetherpositive emotion actually improves group and interper-sonal functioning in organizations, as well as the partic-ular types of organizations in which such effects mightbe greatest.The basic psychological literature on emotion notonly suggests a wider range of consequences for jobattitudes; it also suggests a wider set of determinants.Already, the broader formulation of job attitude re-search mentioned at the outset of this paper has openedorganizational research to enduring (Staw et al.,1986) and even genetically-based dispositions (Arveyet al. 1989), along with social information processing(Salancik and Pfeffer 1978), as predictors of positiveemotion.3 However, the broad psychological literatureon positive emotion suggests still other important de-terminants that have been given little empirical ortheoretical attention by organizational researchers. Inspite of the early work by Roethlisberger and Dickson(1946) and Herzberg (1966), organizational researchershave devoted only modest attention to physical charac-teristics of the workplace as predictors of employeeoutcomes. The psychological literature suggests thatsimple physical variables such as noise, crowding andtemperature are powerful determinants of mood. Forexample, there is consistent evidence that high temper-atures bring about a negative mood (Anderson 1989,Zajonc et al. 1989). These findings, when combinedwith the model presented here, suggest that simplephysical variables (or what have been labeled "hygienefactors") may ultimately have an overlooked influenceon the outcomes that employees receive from theworkplace. Supporting this position have been recentstudies on the effects of physical space on job attitudes(e.g., Oldham and Rotchford 1983, Hatch 1987), andsome provocative research by Baron (1992) on theconsequences of both lighting and scent. When peoplework under lighting conditions that are favorable interms of illuminance and spectral distribution, theyappear to set higher work goals, resolve conflicts morecooperatively, and exhibit more creative thinking.Baron (1992) also reports that similar results have beenfound when individuals work in the presence of pleas-ant rather than unpleasant scents.Our general perspective emphasized that employeeswith positive emotions will be more successful in orga-nizational life than employees with negative emotions.We do not, however, wish to convey the impressionthat negative emotions will always lead to unfavorable

64 ORGANIZATION SCIENCE/VO1.5, No. 1, February 1994



16/22

BARRYM. STAW, ROBERT I. SUTTON AND LISA H. PELLED EmployEmotion and Favorable Outcomes

outcomes. Research is needed on the conditions underwhich employees' negative emotions benefit both them-selves and their organizations. It seems plausible, asnoted by Staw et al. (1986), that negative emotions maybe functional in a job in which critical evaluation is animportant part of the role (e.g., an inspector or aresource allocator). Likewise, the use of heuristics andthe speed of response often associated with positiveaffect (Isen and Means, 1983, Isen et al., 1982) may notbe applicable in situations in which cautious and delib-erate decision-making is needed.There also may be hidden costs to conveying positiveemotion in interactions with subordinates, peers andsuperiors. Employees who respond to interruptionsfrom others by being positive or friendly may reinforcesuch behavior, and thus be interrupted with increasingfrequency. As a result, warm and friendly employeesmay be unable to get their work done, while negativeor hostile employees-whom others may dread inter-rupting-may be more productive because they workwith fewer diversions (Pfeffer, 1989). Dissatisfactionand negative emotion may also lead to constructiveconflict. Bell and Staw (1992), for example, found thatstudents who were the least satisfied with their MBAprogram were the ones who made the greatest numberof suggestions for the program's improvement.Similarly, Janis (1989) has emphasized that poordecisions may be made by groups composed of mem-bers who have strong interpersonal attraction, thatexpress positive feelings to one another, and avoidunpleasant open conflict. He describes this situation asgroupthink, a decision-making pathology of highly co-hesive groups that suppress dissent and have the opti-mistic illusion that they are invulnerable. His worksuggests that groups that are characterized by excessivepositive emotion will make poor decisions becausemembers do not critically evaluate each decision andits consequences. Janis posits that encouraging argu-mentation, and thus a moderate level of expressednegative emotion, will help groups engage in criticalthinking.Finally, it should be noted that there are someoccupations in which employees who express negativeemotions garner social and financial rewards from theorganization. For example, bill collectors and policeinterrogators are paid, in part, for the expression ofnegative emotions that degrade the self-esteem of oth-ers, so as to convince them to pay their overdue billsor to confess a crime (Hochschild 1983, Rafaeli andSutton 1989). Nonetheless, even in occupations wherepeople are paid for acting in a negative manner, acloser examination reveals that employees are re-

warded for being friendly to their superiors andcoworkers, and that acting nasty is only one of themany social influence strategies used during interac-tions with a recalcitrant public (Rafaeli and Sutton1991, Sutton 1991).

ConclusionThis study was not meant to provide a definitive an-swer to how emotion shapes behavior and outcomes inorganizations, but to help open the topic for seriousresearch. We have attempted to map the likely conse-quences of positive affect in organizations and to pointto its most plausible mediatory processes. This studyshould therefore be viewed as an early step in a broadcampaign-a long-term program of research that mayultimately link the myriad of human emotions with aset of relevant outcomes for both individuals and workorganizations.This research cast its conceptual net beyond the jobsatisfaction literature. The perspective proposed hereimplies that organizational researchers may have de-voted too much effort in defining and measuring jobsatisfaction and not enough time exploring the broaderimplications of emotion in the workplace. The presentstudy focused on what is probably the most widely-studied, and most widely agreed-upon dimension ofhuman emotion, the positive-negative continuum(Zajonc et al. 1989). But other possible emotions arealso candidates for study in the organizational context.For example, it may be fruitful to delineate reactionsto stressful jobs into specific emotions such as fear,frustration, dread and fatigue, since the consequencesof these specific emotions may differ from that of themore global construct of distress. Likewise, instead ofusing broad concepts of job involvement such as orga-nizational identification or internalization, it may beimportant to assess specific emotions that occur atwork such as pride, joy, hope and excitement. Finally,rather than treating the emotions of guilt and jealousyas (usually unmeasured) mediating processes in equitymodels (Adams 1965), it may be useful to considerthese emotions as interesting variables in their ownright. Just as Stearns and Stearns' (1986) writings implythat the experience and control of anger may have animportant influence on employee success, the occur-rence and management of many emotions may havedirect effects on work outcomes.From the above discussion it should be clear thatpositive affect, though important, is not the only emo-tional state of relevance to organizational behavior.Ideally, future research will start to explore the full




17/22

BARRY M. STAW, ROBERT I. SUTTON AND LISA H. PELLED Employee Positive Emotion and Favorable Outcomes

etiology of emotions at work and match these withlikely organizational and personal consequences. Asecond challenge will be to explore the transmission ofvarious emotional states throughout the organization.Already there is social psychological research on theinterpersonal transmission of emotions such as fearand anxiety (Kerckhoff and Back 1968, Barley andKnight 1992). Negative emotion may likewise spreadrapidly in organizations, creating job dissatisfactionthat is endemic to the setting. Positive emotion, incontrast, may be slower to build and more dependenton objective improvements in outcomes. By under-standing the transmission of emotion in organizations,we can start to identify how people determine theinterpersonal environments in which they work(Schneider 1987) and how psychological states mayultimately influence organizational norms and struc-tures (Staw and Sutton 1992).In closing, however, it is interesting to view theimplications of the theory and data presented here forthe vast job satisfaction literature. Although it is awidely accepted truism that job satisfaction has little orno impact on employee performance, our conceptualperspective and supporting evidence suggests that thistruism may require some modification. First, becausethe concept of positive emotion is a broader (andperhaps more dispositional) construct than job satisfac-tion, it is likely that its effects are more widespread andenduring than those of satisfaction. Second, when oneshifts the dependent variables of interest from organi-zational to employee outcomes, as we have done in this

research, the effects of job attitudes may become larger.This shift has particularly striking implications for theliterature on employee performance evaluations. Whenone adopts the employee's perspective instead of theorganization's perspective on performance evaluations,job satisfaction is transformed from a source of un-wanted measurement error to a way of getting ahead atwork. More generally, our theory and data imply thateven if employee attitudes have little or no impact onaspects of performance that benefit the organization,displaying and feeling positive attitudes on the job maycause organizations to perform in ways that benefitupbeat employees.AcknowledgementsWe wish to thank Sigal Barsade, Steven Cardoze, Dick Daft, JenniferHalpern, James Jucker, Roderick Kramer, Sherry Nelson, MarinaPark, Jeffrey Pfeffer, Lorna Peden, Anat Rafaeli, Wilfred Zerbe, andthree anonymous Organization Science reviewers for their contribu-tions to this manuscript.

AppendixWelistbelow the itemscomposinghe scales used in this study.Eachitem is followed by parentheses.The first symbol in parenthesesindicates he source of data for an item:I = respondent elf-reportduring he 90 minute structurednterview; O = observationsmadeof the respondentby the interviewerat the end of the structuredinterview;TO = structured bservationsmade by trained nterview-ers; and S = evaluationsreported by the respondent's mmediatesupervisor.The second symbol n parentheses ndicates he numberof responseanchorsused for the item. The thirdsymbol n parenthe-ses indicates whether the item was positivelyor negativelycoded.Positive (+) coding means that larger values reflect more of theconstruct; egative -) codingmeans hatlargervaluesreflect ess ofthe construct.

Items Measuring Positive EmotionHowoften do you feel thiswayat work?I feel down-hearted ndblue. (I, 4, -)I get tiredfor no reason. I, 4, -)I findmyselfrestlessandcan't keep still. (1,4, -)I find it easyto do the thingsI used to do. (I, 4, +)Mymind s as clearas it used to be. (I, 4, +)I feel hopefulaboutthe future. I, 4, + )I find it easy to makedecisions. I, 4, +)I am more irritable hanusual.(I, 4, -)I still enjoythe thingsI used to. (I, 4, +)I feel thatI amusefulandneeded.(I, 4, +)How do you see yourself n yourwork?Sad/Happy (I,7, +)Here is [a pictureof a] ladder.[It] describeshow much pep andenergya personhas.Thetopof the ladder ndicatesalwaysbeingfullof pep and energy,and the bottom of the ladderrepresentsneverhavingany pep or energy.Please tell me whichstep on the ladderindicateshow muchpep andenergyyou'vehadlately.(I, 7, +)Check he column hat indicateshow often the employeedideach ofthe following:Smiled. TO, 4, +)Chuckled r laughed. TO, 4, +)Said something hat he/she mighthave expectedto "get a laugh"fromsomeone. TO, 4, +)

Items Measuring SupervisorEvaluationCheck he boxon each line that best describes the personwho reports o you]:Does very high qualitywork/Does very ow qualitywork. S, 7, -)Does a largeamountof work/Does very ittlework. S, 7,-)Very dependable/Veryundependable. S, 7, -)Alwaysarriveson time/Alwayslate. (S, 7, -)Very creative/Not at all creative. S, 7,-)Likesworking ery much/Dislikesworking erymuch. S, 7, -)Enjoyshavingresponsibility/Avoids avingresponsibility.S, 7, -)Gets alongwell with otherpeople/Doesn't get alongwell with otherpeople. (S, 7, -)

66 ORGANIZATION SCIENCE/Vol. 5, No. 1, February 1994



18/22

BARRYM. STAW, ROBERT I. SUITON AND LISA H. PELLED EmployEmotion and Favorable Outcomes

Items Measuring Supervisor SupportPlease tell [the interviewer]how true each statement is of [yoursupervisor]:Standsup to outsiders or the people he/she supervises. I, 4, +)Takes a personal nterest n those he/she supervises. I, 4, +)Keepsinformedabout the way his/her people think and feel aboutthings. (I,4, +)Goes out of his/her wayto praise goodwork. I, 4, +)

Items Measuring Coworker SupportHow good are [yourcoworkers] bout givingyou the help you needto do your ob? (I, 4, + )How well do you feel that [your coworkers] hare with you andamong themselves news about important things that happen at(STUDY EMPLOYER)? I, 4, +)How well do you feel that [your coworkers]get along together?(I,4, +)How well do you feel that [yourcoworkers] tick together? I, 4, +)How readyare [yourcoworkers] o defend each other from outsidecriticism? (I,4, +)

Items Measuring Observed Job CharacteristicsHow muchvariety s there in the job? (TO,7, +)The job requiresan individual o do the same thingsover and overagain. (TO, 6, -)The job providesan individual he opportunityo do a numberofdifferentkinds of thingsat work. TO,6, +)Howmuch autonomys there in the job? (TO,7, +)The job denies the individualany chance to use his/her personalinitiativeor discretionat work. TO,6, -)He/She is givenenoughfreedom o decidehow to do his/her work.(TO, 6, +)The job allows an individual o make a lot of decisionson his/herown. (TO, 6, +)To what extent does doing thejob itself provide he employeewith"feedback" bouthowwell he/she is doing? TO,7, +)Just doing the workrequiredbythejob providesmanyopportunitiesfor a personto figureout how well he/she is doing.(TO,6, +)Thejob is meaningful. TO,6, +)

Item MeasuringRated IntelligenceRate [respondent's] pparent ntelligence. IO, 5, +)Endnotes1 The more supportive ocial contextmay be limited to representa-tives of the formalorganization.Employeeswho persevereand useenhancedcognitive unctioning o do exceptionalwork may be criti-cized and shunnedby their peers because they are "rate busters"(Lawler1973).2 There is some evidence that the effects of negativeand positivemood on helpingare not symmetrical.Researchby Cialdiniand hiscolleagues suggeststhat people induced to be in bad moods as aresult of some transgressione.g., knockinga stack of the investiga-tor's computer cardson the floor) are more prone to be help aperson in need (see Cialdiniet al. 1982). Cialdini developedthe

Negative State Relief model to explain these findings:Altruismpromotes positive feelings, and thus is one of many methodsthatpeople can use to return to a good moodwhen they are in a badmood. His research ndicates hat people who are in badmoods canbe returned o a good mood by encounteringany reinforcer.Thussubjectswho commit ransgressions, ut are given ice creambeforebeing givena chanceto help someone, are no more likelyto engagein altruismthan subjects in neutral moods. The implicationsofCialdini's esearch or the relationshipbetweennegativeaffect andaltruism n organizationalettingsare unclear.Employeesmay usealtruismnow and then to breakout of foul moods. But we see noreason to expect that employeeswho generallyare dissatisfiedordepressedwill be more helpfulto othersat the workplace.Depres-sion is associatedwith less activityn generaland less socialinterac-tion in particular.Thus it seems reasonable o expect that peoplewho generally ravel throughorganizationalife in a bad mood willhave less energy or helpingothers andwill encounter ewer oppor-tunities to do so.3 The presentstudy focused on the consequencesrather than thecauses of positive emotion. Yet, as with prior researchsuggestingthat job attitudes stem partly from stable dispositions Stawet al.1986), we did uncover evidence of stability n employee positiveemotionduring he courseof this longitudinaltudy.The threeitemsmeasuringdisplayedpositiveemotionwere not measuredat time 2.As a result,we could only calculate correlationsbetween the 12self-report positive emotion items at time 1 and at time 2. Thecorrelationbetween positiveemotion at time 1 and at time 2 was0.49 (n = 267, p < 0.001).In addition,a t-test indicatesno signifi-cant difference p = 0.85)between the level of positiveemotionattime 1 (mean = 3.53) and at time 2 (mean = 3.54).

ReferencesAbramson,L. Y., M.E.Seligman ndJ.D. Teasdale 1987),"LearnedHelplessness n Humans:CritiqueandReformulation,"ournalof Abnormal Psychology, 87, 32-48.

Adams,J. S. (1965),"Inequalityn SocialExchange,"n L. Berkowitz(Ed.), Advances in Social Psychology, 2, 267-299.

Alexander,E. R. and R. D. Wilkins 1982), "PerformanceRatingValidity:The Relationshipof Objectiveand SubjectiveRatingsof Performance," Group and OrganizationStudies, 7, 485-496.

Alloy,L. B. andL. Y. Abramson1979),"Judgmentsf Contingencyin Depressedand Non-depressed tudents:Sadderbut Wiser?"Journal of Experimental Psychology:General, 108, 441-485.

____ - and D. Viscusi (1981), "Induced Mood and the Illusionof Control," Journal of Personality and Social Psychology, 41,1129-1140.

Ammons,R. B. and C.H. Ammons 1962),TheQuickTest,Missoula,MT:Psychological est Specialists.Anderson,C. A. (1989),"Temperature ndAggression:UbiquitousEffectsof Heat on the Occurrenceof HumanViolence," Psy-chological Bulletin, 106, 74-96.

Arvey,R. D. and J. E. Campion 1982),"The Employment nter-view:A Summary nd Review of Recent Research,"PersonnelPsychology, 35, 281-322.




19/22

BARRY M. STAW,ROBERT I. SUTTONAND LISA H. PELLED Employee Positive Emotion and Favorable Outcomes

Arvey, R. D., T. J. Bouchard Jr., N. L. Segal and L. M. Abraham(1989), "Job Satisfaction: Environmental and Genetic Compo-nents," Journal of Applied Psychology, 74, 187-192.

Asch, S. E. (1946), "Forming Impressions of Personality," Journal ofAbnormal and Social Psychology, 41, 258-290.

Bandura, A. (1982), "Self-efficacy Mechanism in Human Agency,"American Psychologist, 37, 122-147.(1991), "Social Cognitive Theory of Self-regulation," Organiza-tional Behavior and Human Decision Processes, 50, 248-287.

Barley, S. R. and D. Knight (1992), "Toward a Cultural Theory ofStress Complaints," in B. M. Staw and L. L. Cummings (Eds.),Research in Organizational Behavior, Vol. 14, 1-48, Greenwich,CT: JAI Press.

Baron, R. A. (1993), "Affect and Organizational Behavior: Whenand Why Feeling Good (or Bad) Matters," in K. Murnighan(Ed.), Social Psychology in Organizations:Advances in Theoryand Research, Englewood Cliffs, NJ: Prentice-Hall, 63-88.

Beehr, T. A. (1976), "Perceived Situational Moderators of the Rela-tionship Between Subjective Role Ambiguity and Role Strain,"Journal of Applied Psychology, 61, 35-40.

Bell, N. M. and B. M. Staw (1992), "The Functions of Dissatisfac-tion," Working paper, Johnson School of Management, CornellUniversity, Ithaca, NY.

Bell, P. A. (1978), "Affective State, Attraction, and Affiliation: Mis-ery Loves Happy Company, Too," Personality and Social Psy-chology Bulletin, 4, 616-619.

Berkowitz, L. (1978), "Do We Have to Believe We Are Angry withSomeone in Order to Display "Angry" Aggression toward thePerson?" in L. Berkowitz (Ed.), Cognitive Theories in SocialPsychology, New York: Academic Press.

and W. H. Conner (1966), "Success, Failure, and Social Re-sponsibility," Journal of Personality and Social Psychology, 4,664-669.

Caplan, R. D., C. Cobb, J. R. P. French Jr., R. V. Harrison and S. R.Pinneau Jr. (1980), Job Demands and Worker Health, AnnArbor, MI: Institute for Social Research.

Cardy, R. L. and G. H. Dobbins (1986), "Affect and AppraisalAccuracy: Liking As an Integral Dimension in Evaluating Per-formance," Journal of Applied Psychology, 71, 672-678.

Carnevale, P. J. D. and A. M. Isen (1986), "The Influence of PositiveAffect and Visual Access on the Discovery of Integrative Solu-tions in Bilateral Negotiation," Organizational Behavior andHuman Decision Processes, 37, 1-13.

Cialdini, R. B. (1984), Influence: How and Why People Agree toThings, New York: Morrow.

,_ D. T. Kenrick and D. J. Baumann (1982), "Effects of Mood onProsocial Behavior in Children and Adults," in N. Eisenberg(Ed.), The Development of Prosocial Behavior, New York: Aca-demic Press, 339-359.

Cobb, S. and S. V. Kasl (1977), Termination: The Consequencesof JobLoss, HEW Publication No. (NIOSH) 77-224, Washington, DC:U.S. Department of Health, Education and Welfare.

Cooper, W. H. (1981), "Ubiquitous Halo," Psychological Bulletin, 90,218-244.

Coyne, J. C. (1976), "Depression and the Response of Others,"Journal of Abnormal Psychology, 85, 186-193.

Crowne, D. P. and D. Marlowe (19640, The Approval Motive: Studiesin Evaluative Dependence, New York: Wiley.

Deaux, K. (1985), "Sex Differences," Annual Review of Psychology,36, 49-82.

Drachman, D., A. DeCarufel and C. A. Insko (1978), "The ExtraCredit Effect in Interpersonal Attraction," Journal of Experi-mental Social Psychology, 14, 458-465.Dunning, D. and A. L. Storey (1991), "Depression, Realism, and theOverconfidence Effect: Are the Sadder Wiser When PredictingFuture Actions and Events?" Journal of Personality and SocialPsychology, 61, 521-532.

Ebbson, E., B. Dunean and V. Konecni (1975), "Effects of Contextof Verbal Aggression on Future Verbal Aggression: A FieldExperiment," Journal of Experimental Psychology, 11, 192-204.

Eder, R. W. and G. R. Ferris (1989), The Employment Interview:Theory, Research, and Practice, Newbury Park, CA: Sage Publi-cations.

Ekman, P. (1982), "Methods for Measuring Facial Action," in K. R.Scherer and P. Ekman (Eds.), Handbook of Methods in Nonver-bal Behavior Research, Cambridge: Cambridge University Press,45-90.(1985), "Telling Lies," New York: Berkeley.

Fiske, S. T. and S. E. Taylor (1991), Social Cognition, New York:McGraw-Hill.

Foa, U. G. (1971), "Interpersonal and Economic Resources," Sci-ence, 171, 345-351.

Forbes, R. J. and P. R. Jackson (1980), "Non-verbal Behavior andthe Outcome of Selection Interviews," Journal of OccupationalPsychology, 53, 65-72.

Ganster, D. C., M. R. Fusilier and B. T. Mayes (1986), "Role ofSocial Support in the Experience of Stress at Work," Journal ofApplied Psychology, 71, 102-110.

George, J. M. (1989), "Mood and Absence," Journal of AppliedPsychology, 74, 317-324.

and A. P. Brief (1992), "Feeling Good-Doing Good: A Concep-tual Analysis of Mood at Work-Organizational SpontaneityRelationship," Psychology Bulletin, 112, 310-329.

Glick, W. H., G. D. Jenkins Jr. and N. Gupta (1986), "Method versusSubstance: How Strong Are Underlying Relationships BetweenJob Characteristics and Attitudinal Outcomes?" Academy ofManagement Journal, 29, 441-464.

Gouaux, C. (1971), "Induced Affective States and InterpersonalAttraction," Journal of Personality and Social Psychology, 20,37-43.

Gouldner, A. W. (1960), "The Norm of Reciprocity: A PreliminaryStatement," American Sociological Review, 25, 161-178.

Gupta, N. and T. A. Beehr (1982), "A Test of the CorrespondenceBetween Self-reports and Alternative Data Sources about WorkOrganizations," Journal of VocationalBehavior, 20, 1-13.

Hackman, J. R. and G. R. Oldham (1980), "Work Redesign,"Reading, MA: Addison-Wesley.

Hatch, M. J. (1987), "Physical Barriers, Task Characteristics, andInteraction Activity in Research and Development Firms," Ad-ministrativeScience Quarterly, 32, 387-399.

Herzberg, F. (1966), Work and the Nature of Man, Cleveland, OH:World.

68 ORGANIZATION SCIEfNCE/VO1.5, No. 1, February 1994

http://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/page/info/about/policies/terms.jsphttp

Motivation 1.pdf

Documents

Transcript of Motivation 1.pdf