Antecedents of workplace emotional labor dimension

Antecedentsofworkplaceemotionallabordimensionsandmoderatorsoftheire€ectsonphysicalsymptomsJOHNSCHAUBROECK*1ANDJAMESR.JONES21DepartmentofManagement,LeBowCollegeofBusiness,DrexelUniversity,33rd&ArchStreets,Philadelphia,PA19104,U.S.A.2DepartmentofMarketing/Management,CollegeofBusinessAdministration,UniversityofNebraskaatOmaha,Omaha,NE,U.S.A.SummaryThepresentstudydistinguishedbetweentwomodalemotionaldisplayrules,demandstoexpresspositivee€erenceanddemandstosuppressnegativee€erence,thatpartiallyconstitutetheworkrolesofmanyemployees.Perceiveddemandstoexpresspositiveemotionwerepositivelyrelatedtohealthsymptomsprimarilyamongthosereporting:(1)loweridenti®cationwiththeorganization;(2)lowerjobinvolvement;and(3)loweremotionaladaptability.Thee€ectsofvariouspersonalitytraitsandsituationalvariablesonperceivedemotionallabordi€ereddependingonthenatureoftheemotionallabor.The®ndingsarediscussedintermsofimplicationsofemotionallaborforhealthandpracticesthroughwhichorganizationsmightintervenetominimizeitsunhealthfulconsequencesamongemployees.Wealsoattempttoreconcilethe®ndingswithsomeoftherelatedresearchinpsychologysuggestingthatsomeformsofrequirede€erencemayhavesalutaryphysiologicalconsequences.Copyright#2000JohnWiley&Sons,Ltd.IntroductionMuchresearchinvestigatingtherelationshipbetweenjobdemandsandhealthhasemphasizedthekeypsychosocialnatureofthedemands.Pressuresareplacedonindividualsinasocialcontextandtheirbehavioralresponsestothesepressureshavesociallyderivedimplicationsforthem.Beingrequiredtomodulatetheexpressionofone'sownemotionsinparticularwaysisasigni®cantcomponentoftheworkroleformanypersons.Thistypeofrolerequirementhasbeencalled`emotionallabor'(cf.WhartonandErickson,1993,p.458).TheconceptofemotionaldisplayruleswasintroducedbyPaulEkman(e.g.,Ekman,1972).Emotionaldisplayrulesrefertonormsaboutappropriateemotionalexpressionforspeci®csituations.Personswhohavemuchcustomerorclientcontact(e.g.,salespersons,nurses)areseentobesubjecttostrongeremotionaldisplayrules(Sutton,1991;SuttonandRafaeli,1988).Severalauthorshaveobservedthatsuchdisplayrulesmaybeexpectedtocompromisethepsychologicaland/orphysicalhealthofworkersCCC0886±9383/2000/020163±21$17.50Copyright#2000JohnWiley&Sons,Ltd.JournalofOrganizationalBehaviorJ.Organiz.Behav.21,163±183(2000)*Correspondenceto:JohnSchaubroeck,DepartmentofManagement,LeBowCollegeofBusiness,DrexelUniversity,33rd&ArchStreets,Philadelphia,PA19104,U.S.A.E-mail:[e-mail]oadisturbingdisequilibrium(or`dissonance')betweenfeltemotionsandtheemotionsonemustexhibit(AshforthandHumphrey,1993;MorrisandFeldman,1996;RafaeliandSutton,1987;WhartonandErickson,1993).Indeedthereisconsiderableevidencefromoutsidetheworkspherethatchronicallyexperiencingcon¯ictamongone'semotionshasnegativehealthconsequences(seereviewsbyFriedman,1989;KingandEmmons,1990).Emotionallaborisaconstructofincreasinginterestamongtheoristsinthemanagementliterature(cf.AshforthandHumphrey,1993;MorrisandFeldman,1996;RafaeliandSutton,1987;WhartonandErickson,1993).Fromanempiricalpointofview,however,`theexpressionofsociallyappropriateemotionconstitutesaneglectedformofroledemand'(AshforthandHumphrey,1993,p.110).AsnotedbyWhartonandErickson(1993),`virtuallyallpreviousstudies[ofemotionallabor]havebeencasestudies'(p.462).Toourknowledge,onlytwoquantitativestudiesofemotionallaborintheworkplacehavebeenpublishedintheliterature.Aftersamplingworkersintwocompanies,Wharton(1993)constructedadummyvariableofemotionallaborbyusingaschemedevelopedbyHochschild(1983)toclassifysomeofthejobsashighinemotionallabor.Wharton(1993)observednosigni®cantcorrelationbetweenthisvariableandemotionalexhaustion,andasigni®cantnegativecorrelationwithjobsatisfaction.MorrisandFeldman(1997)developedascalemeasureofemotionallabordi€erentiatedalongthreedimensions:durationofcontactwithothers,frequencyofcontactwithothers,andemotionaldissonanceinspiredbysuchcontact.Theirstudysupportedhypotheseslinkingemotionallabortocharacteristicsofthejobaswellasemotionalexhaustion.DespitetheclaimsofHochschild(1983)andothersthatemotionallaborisrelatedtophysicalsymptoms,thereisnoresearchrelatingemotionallabortophysicalsymptoms.Aswereviewbelow,laboratorystudiesfrompsychologyaresuggestiveofthepossibilitythatcertaintypesofemotionallabormayinfacthavesalutaryin¯uencesonindividual'sphysiologicalwell-beingwhereasotherformsmaybetoxictohealth.Wedistinguishinthisstudybetweendemandstosuppressnegativeemotion,whichmaygenerallybeexpectedtobeunhealthful,anddemandstoexpresspositiveemotion.Second,weidentifykeyindividualandsituationalmoderatorsofthisemotionallaborÐsymptomsrelationship.Third,itisnotclearwhetherortowhatextentemotionallaborre¯ectsdispositionaltendenciesand/orobjectivesituationalexposures.Accord-ingly,weexaminehowindividualdi€erencesandsituationalvariablesdeterminedtwodimen-sionsofemotionallaborandalsohowthesevariablesmoderatedrelationshipsbetweenemotionallaborandhealthsymptoms.Figure1depictstheframeworkforthisstudy.Objectivecharacteristicsoftherole(i.e.,requirementsforinterpersonalinteraction)