Stress and Social Support

Stress and Social SupportAmong Online DoctoralPsychology StudentsJason A. KaufmanABSTRACT.Thisstudyexploredtherelationshipbetweenperceived stressandperceivedneedforsocialsupportamongdoctoralpsychology studentsattendingadistanceeducationuniversity.Althoughsmall samplesizeconceivablypreventedidentificationofadirectcorrelation betweenthesetwofactors,femaledoctoralstudentsperceivinggreater levelsofstressdemonstratedlessaffiliativeandmoreaggressiveneeds. Perceivedstressdidnotcorrelatewithcounselingreadinessamongthe studentsample.Thesefindingssuggestthatfemaledoctoralstudents engagedinonlinestudymayresembletheirtraditionalcampuspeers. [ArticlecopiesavailableforafeefromTheHaworthDocumentDeliverySer- vice:1-800-HAWORTH.E-mailaddress: Website: 2006byTheHaworthPress,Inc. Allrightsreserved.] KEYWORDS.Stress,socialsupport,distanceeducation,doctoral, graduate, studentsJasonA.Kaufman,PhD,isInstructorofPsychology,InverHillsCommunity College,HeritageHall302d,250080thStreetEast,InverGroveHeights,MN55076. ThisstudywasconductedwhiletheauthorwasstudyingatCapellaUniversity, Minneapolis,MN. Journal of College Student Psychotherapy, Vol. 20(3) 2006Available online at http://www.haworthpress.com/web/JCSP 2006 by TheHaworthPress, Inc. All rights reserved. doi:10.1300/J035v20n03_0779INTRODUCTIONItisoftenacceptedatfacevaluethatdoctoralstudyisapersonallyand academicallychallengingendeavor.Suchanacademicendeavormay teststudentsregardingperceptionsofself,relationships,finances,and opportunitiesforfutureemployment.Doctoralstudentsindeedface enormousdemandsupontheirtime,energy,intelligence,endurance, patience,andorganizationalskills(CommitteeontheCollegeStudent, 2000,p.1).Consequently,itistypicaltoexperienceincreasedstress resultantofthedemandsandexpectationsinherentinthegraduate schoolprocess(Toews,Lockyer,Dobson,&Brownell,1993).This stressmayprohibitdoctoralstudentsfrombenefitingfromthestress- alleviatingeffectsofsocialsupport.Theissuebecomesmoreproblem -aticstillwhenappliedtothecaseofdoctoralstudentsattendinga distanceeducationuniversity.Itisliterallyunknowntowhatextent characteristicsofdistanceeducationmayaffectperceptionsofstress andsocialsupportamonggraduatestudents.However,initialexplora- tion can begin with a consideration of the relevant topics. StressAccordingtothecognitive-relationaltheoryofstress(Lazarus,1990), stressisanongoingtransactionbetweenindividualandenvironmentin whichtheindividualappraisestheenvironmentasposingachallenge. Stressisthereforeapost-appraisalstate,aperceptionofacontinuous person-environmentinteraction.Consideredanotherway,stressisan internalstateoftensionresultantofperceptions,anexampleofclassical conditioningacrossthepsychologicalandphysiologicaldomains.In -deed,stressmayimpairthefunctionsofthemind-bodyandthusinterfere withanabilitytoattendtodoctoralstudies(Ray,2004).Perceptionsof chronicstressmayevennegativelyaffectlifespan(Epeletal.,2004). Anumberofstudiesduringthepastdecadehavesoughttoex -ploreperceptionsofstressamonggraduatestudents(e.g.,Nelson, DellOliver,Koch,&Buckler,2001;Rocha-Singh,1994;Soares, Prestridge,&Soares,1992;Soares,Soares,&Ward,1987).Morespe -cificfocushasevenbeendevotedtoquestionsregardingstressand socialsupportamonggraduatestudentsstudyingpsychology(e.g., Cahir&Morris,1991;Hudson&ORegan,1994).Thesestudieshave demonstratedtheproblemsinherenttoperceivedstressintheabsence oftheperceptionofsufficientsocialsupportamonggraduatestudentsin psychology and other disciplines.80JOURNAL OF COLLEGE STUDENT PSYCHOTHERAPY Social SupportSocialsupportreferstothoseactualandperceivedinterpersonal resourcesavailableinanindividualsenvironment.Mallinckrodtand Leong(1992)reviewedtwoprimarymodelsofhowsocialsupportmay havebeneficialeffects.Themoderatingeffectshypothesis,proposedby CohenandWills(1985),suggeststhatsocialsupportmayfunctionas abufferagainststress(Koopmanetal.,2000;Mallinckrodt&Leong, 1992).Thisprotectiveeffectofsocialsupportisthoughttooperateby contributingpreventiveexternalresourcestoindividualsunderper -ceivedstress.Forexample,LawsonandFuehrer(2001)demonstrated thatperceptionsofsocialsupportmoderatedtherelationshipbetween stress and satisfaction with graduate school.Conversely,themediatingeffectshypothesismaintainsthatactualor perceivedsocialsupportaffectstheperceptionofstress(Mallinckrodt& Leong,1992)bydirectlyinterferingwiththestressresponse(cf.Lazarus, 1990).Evidencesuggeststhatthegendersmaydifferentiallyexperience suchmediatingeffects.MunirandJackson(1997),forexample,found thatfemaledoctoralstudentsmostfrequentlycitedfriendsastheirgreat- estsourceofsocialsupport,whereasadvisorswerecitedtheleast. MallinckrodtandLeong(1992)furthermorefoundevidencethatmale andfemalegraduatestudentsmayperceiveavailabilityofsocialsupport differently.Resultsoftheirstudyrevealedthatfemalegraduatestudents, ascomparedwiththeirmalecounterparts,perceivedsignificantlyless supportfromtheiracademicdepartmentsandfamilialenvironments. Femalegraduatestudentsadditionallyexperiencedmoderatingeffects fromfamilialsupport,butnomediatingeffectsonperceptionsofstress fromsuchsupport.DayandLivingstone(2003)recentlysupportedthis ideaamongasampleofuniversitystudents.Intheirstudy,femaleuniver -sitystudentsweremorelikelytoreportseekingsocialsupportfromfam -ilyandfriendsduringtimesofperceivedstress.Questionthusremainsto whatextentperceptionsofstressandsocialsupportrelate,especially amonggraduatestudentsattendingdistanceeducationuniversities. HypothesesThecurrentstudyconsequentlysoughttotestanumberofhypotheses regardingonlinedoctoralpsychologystudents.First,itwashypothe -sizedthatoverallratingofperceivedstressasmeasuredbythePerceived StressScale(PSS;Cohen,Kamarck,&Mermelstein,1983)wouldnot Jason A. Kaufman81 differbygender.Thishypothesiswasbasedonthelackofgenderdiffer -encefoundinthePSSnormativesample.Second,itwashypothe -sizedthatoverallratingofperceivedstressasmeasuredbythePSS wouldbepositivelycorrelatedwithperceivedneedforsocialsupport asmeasuredby10oftheneedsscales(i.e., Achievement,Dominance,Intraception,Affiliation,Heterosexuality,Exhibition,Aggression,Suc-corance,Abasement,andDeference )oftheAdjectiveCheckList (ACL;Gough&Heilbrun,1983).Third,itwashypothesizedthatover -allratingofperceivedstressasmeasuredbythePSSwoulddirectly correlatewithreportedreadinessforcounselingasmeasuredbythe ACL scale,Counseling Readiness.METHODSParticipantsPsychologydoctoralstudentsenrolledinanationaldistanceeduca- tionuniversitywereinvitedtoparticipateinthecurrentstudyviagen- eralannouncementspostedaton-campusseminarsaswellasonline. Forty-sixstudentsrespondedintheaffirmativetorequestsforparticipa- tion,41ofwhomcompletedandreturnedtheresponsematerials.The differencebetweengenderswasnotsignificant( 2[1,N=41]=0.989, ns),supportingthefirsthypothesis.However,34ofthe41respondents werefemale.Onlyfemaleparticipantswereconsequentlyincludedin thesubsequentanalysessoasnottoskewresults.Datawerethuscol- lected33femaleparticipants,ranginginagefrom27to60(meanage= 40.52years).Allparticipantsweretreatedinaccordancewiththecode of ethics of the American Psychological Association (2002). InstrumentsDemographicSurvey. Participantswereaskedtocompletea demographicsurveyregardingthefollowingtopics:(1)gender,(2)age, (3)maritalstatus(yes/no),(4)presenceofchildreninthehome(yes/no), (5)employmentstatus(yes/no),(6)numberofquartersintheprogram, and(7)enrollmentstatus(full-timeyes/no). PerceivedStressScale. ThePerceivedStressScale(PSS;Cohen, Kamarck,&Mermelstein,1983)isavalidated14-iteminstrument designedtoassessthedegreetowhichlife-eventsareappraisedas stressful,andisbasedonLazarus(1990)cognitive-relationaltheoryof 82JOURNAL OF COLLEGE STUDENT PSYCHOTHERAPY stress.Itemsareratedvia5-pointLikert-typescales,yieldinganadditive totalscore(sevenitemsarereverse-scored).The PSShasbeen describedasausefulresearchinstrument(cf.Cohen,2004).Ithas demonstratedacceptabletest-retestreliability(0.85)amongacollege studentsample,withmeansof23.18( SD=7.31)and23.67( SD=7.79) across two separate college samples.AdjectiveCheckList. Originallydevelopedin1949,theAdjective CheckList(ACL;Gough&Heilbrun,1983)isaninstrumentthat measuresself-awarenessbyrequestingparticipantstocategorize300 adjectivesalong37scales.Anonthreatening,nonpathologizinginven -tory,theACLcanbeusedtoexplorenormalpsychologicalfunctioning. Forthecurrentstudy,10oftheACLneedsscaleswereutilized(i.e., Achievement,Dominance,Intraception,Affiliation,Heterosexuality,Exhibition,Aggression,Succorance,Abasement,andDeference ).This measure,withits10needsscales,wasspecificallychosenforthe currentstudyduetoitsabilitytoelucidatenonpathologicalbehavior. AlthoughstatisticallysignificantscoresonACLscalesmaysuggestan atypicalresponsestylerelativetothoseofthenormativepopulation, suchscoresdonotspecificallyindicatepsychopathology.Instead,sig- nificantscoresonthe10needsscalesindicateproblemswithmeeting ones social needs. Design and ProcedurePsychologydoctoralstudentsatamajordistanceeducationuniversity wereinvitedviageneralannouncementstoparticipateinthecurrent study.Announcementsweremadethroughapostingwiththeuniver -sitysonlinepsychologyclub,whichatthetimemaintainedapproxi -mately30members,andwerefollowedbysubsequentandidentical onlinepostings31and60daysaftertheoriginalannouncement.Printed fliersofthesameinvitationtoparticipatewerepostedattheuniversity servicesheadquartersduringasummerseminar.Thespecificnumberof studentsattendingthesummerseminarwasunavailablefromtheuni -versity.However,anecdotalinformationsuggestedthepresenceofap -proximately 400 students.Studentswhorespondedintheaffirmativetotheannouncements weremailed:(1)apacketcontaininganinformedconsentformdescrib -ingthestudy;(2)acoverletterinstructingparticipantstocompleteall formsinthepacketandtoreturnthemtotheinvestigatorintheenclosed stamped,self-addressedenvelope;aswellas(3)thedemographicsur -vey;(4)thePerceivedStressScale(Cohen,Kamarck,&Mermelstein, Jason A. Kaufman83 1983);and(5)theAdjectiveCheckList(Gough&Heilbrun,1983). Participantswerealsoaskednottoputtheirnamesonanymaterialsin thepacket,otherthanontheinformedconsentform,inordertomain -taintheirconfidentiality.Participantswhodidnotreturntheirresearch packetswithinapproximatelyonemonthofthemail-outdateweresent a follow-up e-mail requesting that they do so.Bivariatecorrelations. Thestrengthofassociationbetweenpercep -tionofstressandperceivedneedforsocialsupportwasexplored viaasetofbivariatecorrelationsbetweenPerceivedStressScale (PSS;Cohen,Kamarck,&Mermelstein,1983)scoreandeachofthe10 AdjectiveCheckList(ACL;Gough&Heilbrun,1983)needsscales.A bivariatecorrelationwasalsousedtoexplorethedifferencebetween PSS score and score on the ACL scale,Counseling Readiness.RESULTSResponsestothedemographicquestionnaireindicatedthatthetypical femaleparticipantwasinadoctoralpsychologystudentinalong-term relationship(84.85%),withoutchildrenpresentinthehome(45.45%), employed(81.82%),inthe3rdquarter(21.2%)ofstudies,andenrolled asafull-timestudent(81.82%).ResponsestothePerceivedStressScale (PSS;Cohen,Kamarck,&Mermelstein,1983)suggestedperceptions ofstressamongtheparticipantstobecongruentwiththoselevelsprevi- ouslyreportedintheliteratureforadultresponders(mean=24.73, SD=8.23).Thus,theparticipantsdidnotappeartodemonstrateanelevated levelofperceivedstress.Meanresponsestotheelevenscalesofinterest ontheAdjectiveCheckList(ACL;Gough&Heilbrun,1983)were largelywithinthenormalrange,withallbutoneT-scorefallingwithin onestandarddeviationofscalemeans:(1) Achievement=56.30( SD=6.97),(2) Intraception=49.79( SD=12.21),(3) Affiliation=48.70( SD=9.27),(4) Heterosexuality=47.73( SD=9.26),(5) Exhibition=53.21 (SD=7.83),(6) Aggression=54.58( SD=10.85),(7) Succorance=44.55( SD=7.08),(8) Abasement=44.30( SD=8.15),(9) Deference=45.52( SD=10.12),and(10) CounselingReadiness =54.52( SD=11.54).ThemeanT-scoreforthe Dominancescale(57.27, SD=6.44) wassignificantlyelevated.TheACLmanual(Gough&Heilbrun,1983) definesindividualselevatedonthisscaleasastrong-willed,ambitious, determined, and forceful individual (p. 9).84JOURNAL OF COLLEGE STUDENT PSYCHOTHERAPY Bivariate Correlational AnalysisThecurrentstudyhypothesizedthatscoresobtainedfrom10ofthe needsscalesoftheAdjectiveCheckList(ACL;Gough&Heilbrun, 1983)wouldcorrelatewiththatobtainedwiththePerceivedStress Scale(PSS;Cohen,Kamarck,&Mermelstein,1983).Furthermore,it washypothesizedthatscoresonthe CounselingReadiness scalewould correlatewithPSSscores,demonstratingagreaterwillingnesstoen -gageincounselingamongparticipantsreportingperceptionsofhigher stress.Totestthesehypotheses,Pearsonproduct-momentcorrelations werecomputedforallvariables.Asevincedbythecorrelationmatrix (seeTable1),onlyoneofthetwohypothesestestedwaspartially supported.ParticipantsdemonstratedsignificantcorrelationsbetweenPerceived StressScale(PSS;Cohen,Kamarck,&Mermelstein,1983)scoreand theAffiliation(r= 0.379,p