Content uploaded by Kathryn Von Treuer
Author content
All content in this area was uploaded by Kathryn Von Treuer on Jul 17, 2016
Content may be subject to copyright.
Thisisthepublishedversion:
Lough,JonathanandVonTreuer,Kathryn2013,Acriticalreviewofpsychologicalinstruments
usedinpoliceofficerselection,Policing,vol.36,no.4,pp.737‐751.
AvailablefromDeakinResearchOnline:
http://hdl.handle.net/10536/DRO/DU:30059560
Reproducedwiththekindpermissionofthecopyrightowner.
Copyright:2013,EmeraldGroupPublishing
Acriticalreviewofpsychological
instrumentsusedinpolice
officerselection
JonathanLoughandKathrynVonTreuer
Psychology,DeakinUniversity,Burwood,Victoria,Australia
Abstract
Purpose–Thepurposeofthispaperistocriticallyexaminetheinstrumentsusedinthescreening
process,withparticularattentiongiventosupportingresearchvalidation.Psychologicalscreeningis
awell‐establishedprocessusedintheselectionofemployeesacrosspublicsafetyindustries,
particularlyinpolicesettings.Screeninginandscreeningoutarebothpossible,withscreeningout
beingthemostcommonlyusedmethod.Littleattention,however,hasbeengiventoevaluatingthe
comparativevaliditiesoftheinstrumentsused.
Design/methodology/approach–Thisreviewinvestigatesliteraturesupportingtheuseofthe
MinnesotaMultiphasicPersonalityInventory(MMPI),theCaliforniaPersonalityInventory(CPI),the
InwaldPersonalityInventory(IPI),theAustralianInstituteofForensicPsychology’stestbattery
(AIFP),andsomeotherlessresearchedtests.Researchsupportingthevalidityofeachtestis
discussed.
Findings–Itwasfoundthatnotestpossessesunequivocalresearchsupport,althoughtheCPIand
AIFPtestsshowpromise.Mostformalresearchintothevalidityoftheinstrumentslacksappropriate
experimentalstructureandisthereforelesspowerfulas“evidence”oftheutilityofthe
instrument(s).
Practicalimplications–Thisresearchraisesthenotionthatmanycurrentscreeningpracticesare
likelytobeaddingminimalvaluetotheselectionprocessbywayofusinginstrumentsthatarenot
“cutout”forthejob.Thishasimplicationsforpolicyandpracticeattherecruitmentstageofpolice
employment.
Originality/value–Thisresearchprovidesacriticaloverviewoftheinstrumentsandtheirvalidity
studiesratherthanexaminingthegeneralprocessofpsychologicalscreening.Assuch,itisusefulto
thoseworkinginselectionwhoarefacingthechoiceofpsychologicalinstrument.Possibilitiesfor
futureresearcharepresented,anddevelopmentopportunitiesforabestpracticeinstrumentare
discussed.
KeywordsManagement,Selection,Psychology,Screening,Validity
PapertypeLiteraturereview
Generalpractice:psychologicalscreeninginparamilitarysettings
Policeandsimilarorganizationsaroundtheworldfacethechallengeofselectingapplicantswhowill
becomesuccessfulon‐the‐jobperformers.Societyexpectspoliceofficers(andotherpublicsafety
personnel,suchasfirefightersandcorrectionalofficers)tobehaveaccordingtocertainstandards.
Thenatureofpolicingmakesthejobdangerous,whilealsoprovidingofficerswithuniqueauthority.
Itisgenerallyacknowledgedthatpsychologicalscreening,orprofiling,asapersonnelselection
processfortheseorganizationsisaviablemeansofassessingcandidatesforemployment.Job‐
personfitisparticularlyimportantinapublicsafetyposition.Useofprofilingforselectioninthe
policeindustryiswellestablished,particularlyintheUSA(Guller,1993,1994a;Ho,2001;White,
2008).Thebenefitsofprofilingincludereducedstaffturnoveranddecreasedlikelihoodofpoorjob
performance,whichleadtoconsiderablefinancialbenefit(e.g.Choy,1998).However,the
instrumentsusedforprofiling,andtheprocessesandresearchrelatedtousingthem,needtobe
validinorderforthemtoservetheirpurpose.Itistheaimofthisreviewtoexaminetheinstruments
usedinthescreeningprocesswiththeaimofidentifyingwhichinstrument,ifany,providesthe
greatestbenefitfortheselectionprocess.
Itshouldbenotedthatthevastmajorityofresearchinthisfieldhasbeenconductedwithpolice
officersandpolicecandidates;assuch,thisreviewwillfocusonresearchfrompolice‐relatedstudies.
Researchinvolvingotherjurisdictionsdoesexist–forexample,Loughetal.(2007)examinedthe
differencesbetweenscreenedandunscreenedcorrectionalofficers’jobperformances.However,as
themainbodyofworkrelatestopolicework,thisreviewwillbeconfinedtothediscussionofthat
sphere.Thisissupportedbytheargumentthatpolice‐relatedfindingsmaybepartlygeneralizableto
otherjurisdictions(e.g.Barretetal.,1999;Brough,2005).
Howpsychologicalscreeningworks
Thebenefitsofpsychologicalscreeningthatgoesbeyondthestraightforwardinterviewprocessare
wellestablished(Blau,1994).Objectivepsychologicalassessmentreducesriskfromphenomena
suchasinterviewerbias,inter‐interviewdisagreement,andhaloeffects.Thecapacityof
standardizedinstrumentstoallowbenchmarkingtoanappropriatenormativepopulationfurther
increasestheirusefulness.
Theoperationalaspectsofpsychologicalassessmentarewelldocumented(e.g.Murphyand
Davidshofer,1994).Essentially,theprocessinvolvestheadministrationofoneormorevalid
psychologicaltestsinordertoassessaperson’ssuitabilitytoataskorsituation.Thetests’findings
areevaluatedbyapsychologist,wholooksforevidenceormarkersintherespondent’sprofilethat
suggestsuitabilityorunsuitabilityforaparticularrole.Psychologicalscreeninginanappliedsetting
isthepracticalapplicationofsuchassessments–thetestee(anapplicanttothepolicedepartment)
undertakestheassessmentaspartoftheirapplication.Theresultingdataandreportarethenused
bypolicepsychologiststoassistindeterminingthesuitabilityoftheapplicantforpoliceduties.
Testingusuallyinvolvestheadministrationofaformalinstrument(suchastheMinnesota
MultiphasicPersonalityInventory(MMPI))and,inmostcases,aninterview(oftenwitha
psychologist;Blau,1994).Thisreviewfocussesonthefirstpartofthisprocess–theformal
instrument.Thefunctionalgoalofthescreeningprocessisanimportantelementinpsychological
screening.Comparingthedifferentmethodsof“screeningin”(selectionbasedondesirabletraits)vs
“screeningout”(eliminationbasedonundesirabletraits)isanimportantissue.Itisacknowledged
(e.g.HoggandWilson,1995;Meieretal.,1988)thatscreeninginisamoredifficultprocess.
However,itisgenerallysuggestedthatbothapproachesshouldbeusedintandemforoptimal
resultsinthepoliceenvironment(e.g.Hensonetal.,2010;Metchik,1999).Measuresfocussedon
detectingpathology,suchastheMMPI,aremoreusefulasscreen‐outmechanisms,whilesome
personalityinstrumentsandscales–particularlythosemeasuringdesirableorpositivetraits–are
potentiallysuitedtoscreeningin(e.g.Guller,2003).
Thetimingofpsychologicalassessmentsinapotentialemployee’sapplicationprocessisan
importantvariablethatcandifferaccordingtonationallaws.Forexample,intheUSA,the
introductionoftheAmericanswithDisabilitiesActin1992effectivelyeliminatedpurepre‐
employmentscreeningusingtestsofpsychopathology(suchastheMMPI).Thishasresultedinthese
testsbecomingadministrableonlyonceaconditionalofferofemploymenthasbeenmade.Thishas
hinderedresearchintotheeffectivenessofmanyinstrumentsthatuseAmericanpolicesamples,and
hasreducedthereturnoninvestment(ROI)benefitsconferredbytesting(WeissandInwald,2010).
Bycontrast,nosuchlegalrequirementexistsinAustralia,wherepsychologicalscreeningisgenerally
consideredastandardcomponentofthepre‐employmentselectionprocess.
Theusefulnessofaninstrumentinthescreeningprocessultimatelyfocussesononecentral
question:howwelldoestheinstrumentpredictperformance–particularlypoorperformance?The
answertothisquestionliesintheresearchsurroundingtheinstrument.Thisreviewwillnowfocus
onthemostfrequentlyusedinstruments,andwillcriticallyexaminetheresearchunderpinningtheir
useinappliedsettings.Thereaderwillnotethatsingleinstruments,suchastheMMPI,are
comparedwithtestbatteriessuchastheMatrix‐PredictiveUniformLawEnforcementSelection
Evaluation(M‐PULSE)andAustralianInstituteofForensicPsychology(AIFP)system.Whilethismay
seemunfaironthesurface–abattery,orsuiteoftests,shouldlogicallycovermoregroundthana
singleinstrument–thecomparisonisvalidonthebasisofcurrentpractices.Thatis,manypolice
departmentsusesingleinstrumentssuchastheMMPIasthesolepsychologicalscreeningtool
(Aamodt,2010);itisthereforefairtocomparesingleinstrumentswithtestbatteries.
TheMMPI
Instrumentoverview
TheMMPIhasbeenusedinmorepolice‐relatedselectionresearchthananyotherinstrument
(Aamodt,2004).TheoriginalMMPIwasdesignedin1940asameasureofpsychological
maladjustmentforassessingpatientsseekingpsychiatrichelp(Butcher,2005).Thetestwasrevised
in1989andre‐issuedastheMMPI‐2.Bothversionsconsistof567itemstowhichtherespondentis
requiredtoindicateeither“true”or“false.”Theinstrument’stenclinicalscalesmeasurevarious
typesofpathology.Any“T‐score”(standardizedscore)above70(orabove65intheMMPI‐2)
indicatesthelikelihoodofpsychologicalmaladjustment.Threeothernon‐clinicalscalesindicatehow
honestthesubjectwasinansweringthequestions.Thesescalesseektoindicateiftherespondent
wasattemptingtoappearbetteradjustedthanheorsheactuallywas(fakinggood);was
manufacturingorexaggeratingsymptoms(fakingbad,ormalingering);orwasrespondingcandidly.
Morerecently,anumberofRestructuredClinical(RC)scalesweredevelopedtocorrectforsome
overlapintheoriginalclinicalscales(Graham,2006).ThiswasfollowedbytheMMPI‐2Restructured
Form(RF).Consistingof338items,thisisdescribedasmorestatisticallyrigorousandisconsidered
tobebasedonwell‐establishedtheoreticalmodelsofpersonalityandpathology(Ben‐Porathand
Tellegen,2008).Atthisstage,theRCscalesandtheMMPI‐2RFhavehadminimalresearchattention
inthefieldofpoliceapplicantscreening(Sellbometal.,2007).
EarlyreviewsofMMPIresearch
BurbeckandFurnham(1985)reviewedfivestudiesontheMMPIandpoliceselectionconducted
between1964and1980.Theyconcludedthat“ithasnotbeenpossible[y]toshowwhether
psychologicaltesting[theMMPI]candiscriminatebetweenpoliceofficersandmembersofthe
public,orbetweengoodandbadpoliceofficers[y]psychologicaltestingmaybeusefulforselecting
outpeoplesufferingfromsomementalabnormality.”Similarly,inreviewing18studiesonthe
MMPI,Simmersetal.(2003,p.287)notedthattheMMPIwasoriginallydesignedtopredict
psychopathologyandsubsequently“doesnotappeartobeausefulgaugeforidentifyingcertain
behaviorsinthelesspathologicalrange.”Theyarguedthatmanyskillsthatareessentialforsuccess
asapoliceofficermaygoundetected.
Ameta‐analysisthatincludedover300studiesofMMPIresearchdatabyAamodt(2004)observed
thatcorrelationsbetweenMMPIscalesandmeasuresofacademyandpatrolperformancewerelow,
andthatthemajoritywerenotstatisticallysignificant.CorrelationsbetweenMMPIscalesand
disciplineproblemsandcommendationsalsoshowednocorrelation.Aamodtconcludedthat
“becausetheFscaleiscomprisedofitemsfromtheotherclinicalscalesandbecauseitis
significantlyrelatedtobothacademyperformance(R+‐0.11)andsupervisorratingsofperformance
(R=‐0.09),itisprobablythemostusefulindividualMMPIscale”(p.94).Thisconclusionsaysagreat
dealabouttheutilityoftheotherscales.Whilethesecorrelationsmaybestatisticallysignificant,
theyareverysmall–theF‐scaleaccountsforlessthan1percentofthevarianceofsupervisor
ratingsofperformance.Thus,thescalesareunlikelytobeofanyclinicalorpracticalassistance.
MoredetailedexaminationofMMPIresearch
WhileAamodt’s(2004)meta‐analysisprovidedarelativelyrecenttechnicaloverviewofMMPIdata,
amorequalitativereviewofMMPIresearch,itself,iswarranted.WeissandWeiss(2010a)discussed
thegeneralstateofMMPIresearch,andconcludedthattheMMPIhasmuchtooffertothe
disciplineofpolicepsychology.However,theystatedthattheMMPIis“notrecommendedasa
stand‐aloneforscreening,butitcanbeusedinconjunctionwithotheroneormoreother
instruments”(p.69).Thisstatementappearscongruentwiththebulkofpublishedliterature.This
reviewwillnowdiscussarepresentativesampleofMMPI‐relatedresearch.Acompletely
comprehensivediscussionofallrelevantstudies,giventhesizeoftheliteraturebase,isbeyondthe
scopeofthispaper.
Inconclusivestudies.ManystudiesexaminingtheefficacyoftheMMPIasaselectiontoolhave
yieldednegativeorinconclusivefindings.Azenetal.(1973)studied100newlyhiredsheriffsinLos
Angeles,andconcludedthatMMPIscoreswereunsuccessfulinpredictingdropouts.Researchby
Schoenfeldetal.(1980)andMerianetal.(1980)comparedgroupsofproblempoliceperformers
withacontrolgroup.NosignificantdifferencesbetweentheMMPIprofilesofthetwogroupswere
detectedineitherstudy.DralleandBaybrook(1985)usedasimilarapproachandreportedsimilar
results–nosignificantMMPIscalecorrelationswerepresentamonganyperformancecriteriafor
thesampleof356policeapplicants.
Inwald(1988)used16predictorvariables,includingavarietyofpsychologicaltests,aswellasclinical
judgmentsstudiedinvaryingcombinations.Thesewereappliedto219“publicsafetyofficers”who
hadremainedonthejobforfiveyears.Amongthatgroup,28werefiredforcause.Inthiscase,
clinicalinterpretationoftheMMPIwastheleastaccuratepredictorofperformanceissues.A
computer‐generatedpredictionofterminationorretentionbasedonearlierMMPIresearch
(Shusmanetal.,1984)misidentifiedthegreatestpercentageofapparentlysuccessfulofficersby
predictingtheywouldbeterminated.Thismethodincorrectlyrejected36percentofthegroupof
219(78individuals)–aTypeOneerror.
OtherstudieshavefocussedonmorespecificaspectsoftheperformanceofMMPIscales.Hargrave
etal.(1988)foundnopredictiveutilityonaderivedaggressionindex,asdevelopedfromthree
MMPIscales.SimilarfindingsfromacommensuratemethodologywerereportedbyCastoraetal.
(2003).Costelloetal.(1996)examineddisciplinarysuspensiondaysamongservingpoliceofficers,
andfounda0.223correlationwiththeF+Pd+Mascale.Theyconcludedthattherelationship
“appearsunimpressive”(p.302)andthatfurtherconstructvalidationisnecessarytodeterminethe
potentialusefulnessoftheindex.
DanielsandKing(2002)notedthatMMPIscaleswereunabletomakeanydistinctionbetween
successfulandunsuccessfulperformers,asdeterminedbysupervisorratings.Bartoletal.(1992)
reportedthattheMMPIwasnotastrongpredictorofself‐reportedstress.Byusingperformance
ratingsandMMPItocompareapplicantsintwomajorUScities,Winters(1990)concludedthatthere
wasnocorrelationbetweentestsscoresandjobperformance.Thesameconclusionwasreachedby
Wrightetal.(1990),whostudiedpoliceperformanceattheendofa14‐weektrainingcourse.
Surretteetal.(2004)cametoasimilarconclusion.TheyexaminedMMPIspecialscalescoresand
supervisorratingsof129experiencedpoliceofficers,anddeclaredthat“scoresontheGood
Cop/BadCop,HusemannIndex,GoldbergIndex,andGonderIndexwerenotsignificantlyrelatedto
supervisorratings”(p.71).
Studieswithpositivefindings.Alargenumberofstudieshavereportedpositivedataregardingthe
MMPI’sabilitytoaddpredictivepowertopoliceselection.However,manyofthesestudieshad
designissuesthatunderminedanyresearchevidenceoftheMMPI’sutility.Thissectionpresentsa
representativeselectionofthesestudies.
Severalauthors(e.g.Bartol,1982;Bernsteinetal.,1982)reportedsignificantfindingsthatwere
basedonstatisticalcalculationsofrawscores.Thisstudymethodcanbemisleadingbecause,inthe
actualpracticeofMMPIinterpretation,rawscoresareneverused.Interpretationofthetestresults
isalwaysundertakenfromthescaledscores,asthebodyofvalidityresearchsupportingtheMMPIis
allbasedonscaled–notraw–scores(Butcher,2005).Further,inthesestudies,whenthereported
rawscoreswereconvertedtoT‐scores,thestatisticallysignificantdifferencesdisappeared,andthe
meanT‐scoreswereallinthenormalrange.Thismadeitimpossibletoobtainanyclinical
differences.Bernsteinetal.(1982)reportedfollowupdataon120policewhocompletedtheMMPI
priortobeingemployed.SignificantcorrelationsbetweenMMPIrawscoresandfieldperformance
werereported.However,whentherawscoreswereconvertedtoT‐scores,allwereintheaverage
range.Undertheseconditions,itisunlikelythatapsychologistusingthetestforscreening
applicantswouldfindanymeaningfuldata.
Bartol(1982)administeredtheMMPIto102malepoliceapplicantsand,usingrawscores,reported
astatisticallysignificant“elevation”onfiveMMPIscalesforthoseofficersrated“belowaverage”
throughsupervisorperformanceratings.Whentherawscoresforeachofthethreegroupsofpolice
weretranslatedtoT‐scores,allwereintheaveragerange,whicheliminatedtheabilityto
differentiatebetweenthegroups.Thissuggeststhatalloftheapplicantslooked“normal”according
totheusualcriteriaoftestinterpretation.
SimilarissueswerepresentinBartol’s(1991)investigationof600policeofficers.Bartolfoundthat
officerswhowere“elevated”onthePd(PsychopathicDeviate),Ma(Hypomania),orL(Lie)scales
werelaterdescribedasimmaturebysupervisors.
Combiningthesethreescoresresultedinanimmaturityindex;however,therewasnodescriptionof
whatconstitutedanelevatedscore.Thedifferenceinimmaturityscoresbetweenthosewhowere
terminatedorforcedtoresignandthosewhoretainedemploymentwasdescribedashighly
significant.However,whenthereportedrawscoreswereconvertedtoT‐scores,thedifferences
betweenthetwogroupsdisappeared,andallsubjectsscoredwithintheaveragerange.Therefore,
theindexappearstohavelittleornoclinicalutility.
Severalstudies(e.g.Beutleretal.,1985;HiattandHargrave,1988ab)reportedstatistically
significantintergroupdifferencesbetweenoneormorescoresonMMPIscales,withthesuggestion
thatsuchfindingsshowtheMMPIcandiscriminatebetweengoodandpoorperformers.Upon
inspection,itwasnotedthattwoissuesrecurredinthesestudies.First,thedifferencesinscale
scoresweresosmallthattheywouldbeunhelpfulintheactualtaskofscreeningpoliceapplicants.
Second,despitethedifferences,thescoresforgoodandpoorperformerswereentirelywithinthe
normalrange.SaxeandReiser(1976)acknowledgedthislatterdifficultyinpracticalapplication.
Theyreportedsignificantdifferencesbetweenseveralgroups,butaddedthat“thesedifferencesare
allwithinthe“normal”rangeandaretoosmall[y]tohavemeaningfulutilityinclinical
differentiationofsuccessfulandunsuccessfulpoliceapplicants”(p.424).
Somestudies(e.g.HookeandKrauss,1971;Blauetal.,1993;BrewsterandStoloff,1999)reported
positivefindingsfortheMMPIandpoliceselectionbyusingsamplesofalreadyemployedofficers.
Theproblemwiththisapproachwasthatthosewhohadbeenhiredrepresentedavery
homogeneousgroup–bydefinition,thestudysubjectsshouldhavebeen“thebestofthebest.”
Furthermore,thelongerthesesubjectshadbeenemployedaspoliceofficers,thelesslikelyitwas
thattheyrepresentednewapplicants.Researchthatusesthismethodhasutilityonlyifthefindings
arereplicatedwithastudyofpoliceapplicants.
Sellbometal.(2007)acknowledgedtheself‐limitingnatureofusinganalreadyemployedpolice
sample,andsubsequentlycorrectedtheiranalysistocompensateforthisrangerestriction.The
resultinganalysisindicatedmoderatepredictiveabilityofcertainMMPIscales(RCscales),andlower
thannormalT‐scoreelevations.AsimilarmethodwasadoptedbyCaillouetetal.(2010)intheir
examinationofthePSY‐5scalesandfacets.Adichotomousoutcomevariable(ceasedemploymentvs
stillemployed)wasusedasthepredictedmeasure.Modestpredictivepowerwasfound,albeitwith
caveats(rangerestrictionsandscoresmodifiedbyimpressionmanagementlevels),andtheauthors
acknowledgedthattheirmodel’sfindingsdid“notrepresentaclearlysignificantadvancefor
predictinglawenforcementofficerperformance”(p.236).
MMPIresearchhasalsousedPoliceAcademyperformanceasanoutcomemeasure.
ProblemswiththisapproachhavebeennotedbyKleimanandGordon(1986),whoobservedthat
thisresearchreliesontrainingperformanceasavalidationmeasure.Theauthorsdiscussedalarge
numberofrelevantlegalcases,andnotedthatvirtuallyallcasesjudgedthattherewas“insufficient
evidencetosubstantiatethevalidityoftheselectiondevice”(p.93).Burkhart(1980)agreed,noting
that“trainingacademydatacannotbeconsideredanidealchoiceforcriteriabecausesuchdatado
notinvolvetheofficers’responsestotherealjobdemands”(p.123).Anumberofstudiesbased
uponMMPIandtrainingperformancefacedtheissuedescribedabove(e.g.GonderandGilmore,
2004;Gough,1950;Hargraveetal.,1986;InwaldandShusman,1984).
Macintyreetal.(2001)usedasophisticatedmethodtodevelopaframeworktopredictproblem
performersamongpoliceofficers.Ascale‐combinationandrecalculationmodelwasconstructed
fromexistingMMPIdata.Thesampleincludedtwomatchedgroupsofmaleofficers.Onegroup
consistedofproblematicperformers,andtheothergrouphadnoindicationsofpoorperformance.
Thestudy’spredictivemodelwasmoresuccessfulinidentifyingproblemperformersthangood
performers,despitethefactthatallcaseswereclassifiedbythatdichotomy.Themodelappearedto
find“addedvalue”residingintheMMPIdata,whichcouldbeusefulatthepersonnellevel.In
practicalterms,thiswouldrequiresubmittingeachMMPIprofiletoarelativelycomplexmodel
developedfromafewscales.Whilecumbersome,thismethodhaspromiseforresearchwithpolice
applicants.
TheCaliforniaPsychologicalInventory(CPI)andtheInwaldPersonalityInventory(IPI)
CPI:instrumentoverviewandreviewofresearch
TheCPIsharesalmosthalf(194)ofits434itemswiththeMMPI.However,thisisfocussedon
commonpersonalityfactors,ratherthanpsychopathology.Itemsaretrueorfalse,andthe
instrumentisscoredon18scales(threeofwhicharetest‐takingvalidityscales).Duetoitsshared
heritagewiththeMMPI,andageneralpredispositionamongpractitionerstoprefertheMMPI,
muchCPIresearchtendstobementionedalongsideMMPIstudies.ThishasresultedinCPIandits
potentialcontributiontopoliceselectionreceivingdilutedattention.
Ingeneral,CPI‐relatedresearchmimicsthebodyofresearchusingtheMMPI;someresearchers
reportsignificantrelationshipsbetweenCPIscalesandaspectsofpoliceperformance,whileother
researchersdonot.Pugh(1985)andHogan(1971)bothreportedthatseveraloftheCPIscaleswere
significantlycorrelatedwithavarietyofperformancemeasures,includingevaluationsbyfield
commanders.However,Hargrave(1985)andSarchioneetal.(1998)werenotsuccessfulinlinking
CPIscaleswithaspectsofpoliceperformance.
Aamodt’s(2004,2010)landmarkmeta‐analysisofpolicepsychologyselectiondatashowedsome
supportfortheCPI’spredictivevalidity.Aamodt’sanalysisalsorevealedthatanumberofscales
werepredictiveofacademyperformance.ThesescalesincludedIntellectualEfficiencyandCapacity
forStatus.However,thetolerancescalewasparticularlypredictive.Aamodtnotedthatthe
tolerancescale“seemstoprovidethebestcombinationofpredictingsupervisorratingsof
performance,disciplinaryproblems,andacademyperformance”(Aamodt,2010,p.240).
IPI:instrumentoverviewandreviewofresearch
TheIPIisa310‐item(trueorfalse)inventorydesignedtoassessarangeofbehaviouraland
personalitycharacteristics.Theinventoryhasafocusonadmittedbehaviourpatterns,suchas
troublewiththelaw,anddrugandalcoholuse.SimilartotheMMPI,italsoassessesother
personalitycharacteristicsandclinicalindicatorsofpsychopathology(Inwald,2010).
ResearchsupportforthepredictivevalidityoftheIPIasapoliceselectiontoolislimited.Inwald
(2010)assertedthattheIPIis“abenchmarkfornewertestsintheselectionfield[y]theIPIhas
demonstratedthatitidentifiesantisocialbehaviorpatternsandisausefulpredictorofpoliceand
publicsafetyofficerjobperformance”(p.91).PositivefindingshavebeenreportedbyDetrickand
Chibnall(2002)whostatedthatsomeoftheIPIscales–notably,familyconflicts,guardedness,and
drivingviolations–werepredictiveofon‐the‐jobperformanceattheone‐yearevaluationpoint.
Similarly,Inwald(1988)reportedthattheIPIwasaccurateinidentifyingwhichapplicants(afterone
yearofperformance)werelikelytobeterminated.However,ahighfalsepositiverate(36percent)–
aTypeOneerror–wasalsonoted.
OtherresearchershavereportedmixedfindingswiththeIPI.Cortinaetal.(1992)administeredthe
IPI(andtheMMPI)tonewlyhiredpoliceofficers.Ignoringtheproblemofrangerestriction(thatis,
theyassumedthatsuccessfulapplicantswerethebestofthebest),theIPI,althoughadequately
measuringthe“BigFive”personalitydimensions,addedlittletonoincrementalvalidity(abovethe
entryexams)forarangeofperformancemeasures,includingacademyperformanceandturnover.
Similarly,MufsonandMufson(1998)observedthatfouroftheIPI’sscaleswerepredictiveofpoor
performanceratingsandeventualtermination;however,asmallsamplesize(n=33)andtherange
restrictionagainlimitedthestudy’sgeneralizability.
Otherpublishedtests
Asmallgroupofotherpsychologicalinstrumentshavealsobeenappliedtothepoliceselection
environment.ThePersonalityAssessmentInventory(PAI)hasbeenusedinanumberofstudies,with
varyingdegreesofsuccess.WeissandWeiss(2010b)discussedthePAIindetail,andreportedthat,
whilesomescales,suchastheInterpersonalandTreatmentscalesshowed“considerablepotential”
(p.87),muchresearchremainstobedone,particularlyconcerningthecriterionvalidityofsomeof
theclinicalscales.Moreover,thePAIisfundamentallyatestofpsychopathology,andhasmany
similaritiestotheMMPI,whichismuchmorewidelyused(Super,2006).AswiththeMMPI,there
remainstheissueofusingatestofpsychopathologyinsteadofassessing“normal”ornon‐clinical
aspectsofpersonalitythatmayhaveasignificanteffectonjobperformance.
TheM‐PULSEisamorerecentlydevelopedinstrument,whichwasfirstintroducedin2008(Davis
suchastheMMPIand16PF,andassessesarangeofattitudesandbeliefsspecificallyrelatedto
policeemploymentduties.Initialresearchsuggeststhatthistestmayhavepromiseforpolice
selectionpurposes(DavisandRostow,2010);however,muchmoreresearchisrequiredbeforethis
instrumentcanbeconsideredempiricallyvalidated.
TheAIFPtestbattery
Instrumentoverview
TheAIFPprofilingsystemisacollectionofinstrumentsthathavebeencombinedandadministered
byAIFPsincetheearly1990s.Itsancestor–theIFPprofilingsystem–hasbeenusedwithsuccessin
theUSA,withsomesuccesssupportedbyresearch(Guller,1994b;Guller,2003).TheAIFPsystem
consistsofsixseparatepsychologicaltests:
.CandidateandOfficerPersonnelSurvey(COPS):atestdesignedbyAIFP,viaresearch,for
thespecificpurposeofselectingpublicsafetyofficerapplicants.Thiscomponentisthe
flagshipofthetestbattery.Scalesinclude,butarenotlimitedto,factorssuchaslegal
difficulties,alcoholanddruguse,aggression,andimpulsivity.
.EdwardsPersonalPreferenceSchedule(EPPS):abroadmeasureofpersonalitythat
assessescharacteristicsdirectlyrelatedtopoliceperformance.Scalesincludefactorssuchas
aggression,needtodominateothers,capacitytolearnfrommistakes,abilitytofitintoa
team,andabilitytotolerateroutinetasks.
.LocusofControl(LOC):ameasureoftheextenttowhichapersonfeelsincontrolofhisor
herlife.Thisisadefactomeasureofmaturity.
.HowSuperviseScale(HS):ameasureofjudgmentininterpersonalsituations,and
awarenessofbehavioralsensitivitywhendealingwithothers.
.ShipleyInstituteofLivingScale:astandardized,well‐researchedtestofgeneralintelligence.
.TheOpinionSurvey:ameasurethatassessesattitudestowardenforcingthelaw,ranging
frombeingrigidandunyieldingtobeingsoftandnaive.
Thefulltestconsistsof540items,withamixoftrueorfalseitems,forcedchoiceitems,andLikert
scaleitems.Itisintendedtobeusedasascreen‐outtool.
ReviewofAIFPresearch
SmythandByrne(1994)appliedtheAIFPprofilingsystemto51VictorianPolicerecruitsontheirfirst
dayattheVictorianTrainingAcademy,andfoundthattheAIFPpredictionsweresignificantly
accurateinpredictingtrainees’performancecategories.Theyalsofoundthat,overtime,the
predictionsweresignificantlyaccurateindeterminingrecruits’sickleaveandtheirlikelihoodof
beinginvolvedininvestigationsbytheInternalInvestigationsUnit(IIU).Thatis,priortocommencing
training,thesystemwasabletosuccessfullypredictanddistinguishbetweengoodandpoor
performers,highandlowsickleaveusers,andrecruits’highandlowlikelihoodsofbeinginvolvedin
IIUinvestigations.
Morerecently,LoughandRyan(2005,2006)examinedtheperformanceofAIFPscreenedand
unscreenedpoliceconstablesafteroneyearofservice.Thescreenedgroupdemonstrablyout‐
performedtheunscreenedgroupacrossmostperformancevariables.Thescreenedgrouptook11
percentlesssickleave,had58percentlessinjuryclaims,were84percentlesslikelytohavebeenin
amotorvehicleaccidentwhiledrivingapolicecar,andweremorethanthreetimeslesslikelyto
incurcomplaintsfromthegeneralpublic.
LoughandRyan(2010)furtherextendedtheirexaminationofpolicecohortsoveraperiodofthree
years.TheAIFPscreenedgroupconsistentlyout‐performedtheunscreenedgroup;theAIFPgroup
tooksignificantlylesssickdays,hadsignificantlyfewernon‐stress(injury)claims,andhad
significantlylessmotorvehicleaccidents.Additionally,thescreenedgrouphadalowerdropoutrate,
andwerelesslikelytomakeseriouson‐the‐joberrorsthatresultedinformaldisciplinaryactionor
investigation.However,aswiththeearlierstudies,individualofficerperformanceratings(e.g.
supervisorreports)werenottakenintoaccount.
OtherresearchsupportingthevalidityoftheAIFPprofilingsysteminAustralianconditionshasbeen
conductedinthecorrectionscontext.Choy(1998)comparedNewSouthWales(NSW)corrections
officersscreenedforselectionbytheAIFPbatterywithanunscreenedgroup,andreported
significantlylowerratesofattritionandsickleaveusageamongtheAIFPscreenedcohortaftertwo
yearsofservice.ChoyestimatedthattheNSWcorrectionsdepartmentsickleavecosthadreduced
by39percentasaresultofthescreening.
SimilarfindingswerereportedamongQueensland(Byrne,2001)andSouthAustraliancorrections
officers(Caseyetal.,2001),withscreenedofficerstakingapproximatelyhalfthenumberofsickdays
oftheirunscreenedcounterpartsduringtheirfirsttwoyearsofemployment.ROIcalculationsupon
Caseyetal.’sdatarevealedamonetarysavingofover$850,000foraninitialoutlayof$22,500–an
ROIfigureof3,815percent.
Loughetal.(2007)trackedthesickleaveoftwocohortsofcorrectionalofficers(AIFPscreenedand
unscreened)overafour‐yearperiod.Theyfoundthatthescreenedgrouptooksignificantlylesssick
leaveineachofthefirsttwoyears,and30percentlessintotalacrossthefouryearsofthestudy.
Theyalsonotedasignificantlylowerdropout(terminationsplusresignations)amongtheAIFP
screenedgroup.Theauthorsconcludedthat“psychologicalprofilingasamethodofselecting
correctionalofficersappearstobeaworthwhileandvalidcourseofaction”(Loughetal.,2007,p.6).
Byrneetal.(2002)summativeinvestigationofAIFPtestingwithinthecorrectionscontextreinforced
theutilityoftheprocessanditsassociatedfinancialbenefits.Forexample,theyfoundthatadeeper
analysisofChoy’s(1998)datarevealedthatthescreeningprocessyieldedareturnoninvestmentof
approximately3,000percent.Inadditiontomonetarybenefits,thereareotherbenefitsthatare
difficulttoquantify–thosederivedfromhiringthemostsuitablecandidatesforthejob,which
improvesstaffmoraleandgeneralefficiency,andreducesthelikelihoodoffuturelegalaction.
FurtherexaminationofAIFPresearch
TheresearchevidencefortheAIFPtestdiffersfromagreatdealofresearchofothertests,
particularlytheMMPI,becauseofthefrequentpresenceofcontrolgroupsthatallowfordirect
comparisonbetweenselectedandnon‐selectedcohorts.Inmostpreviousstudies,thecontrol
groupswerehiredimmediatelypriortotheintroductionofthepsychologicalscreeningprogram
(and,therefore,thecommencementoftheexperimentalgroup),whichtendedtomitigateany
potentialmaturation‐basedresearchartifacts.Thisapproachisparticularlystrongforexaminingthe
effectofthetestasawhole,andalsoallowsforlongitudinalobservation.
AlthoughresearchsupportingthegeneralvalidityoftheAIFPappearssound,thereisnopublished
evidenceofanyincrementalvalidity.Noneoftheindividualcomponents,oranyofthesubscales
withinthosecomponents,havebeenlinkedorcorrelatedwithperformanceoutcomes.Such
researchwouldbevaluablegiventhatthemajorityofscale‐basedresearchfortheMMPIandCPI
tendstodivergeonthepredictivevalidityofindividualscales(althoughAamodt’s,2004findingthat
thetolerancescaleoftheCPIisavalidpredictorisnoted).
Inadditiontotheabove,oneotherfactorconfoundstheAIFPfindings.Researchtodatehasbeen
drawnfromboththeAIFP(Australian)andIFP(American)data.Whilethetwoinstrumentshavea
lotofcommoncontent–includingtheCOPStest–theyarenotthesameinstrument.Research
evidencesupportingtheIFPtestinstrumentdoesnotnecessarilyapplytothepotentialvalidationof
theAIFPtestinstrument,andviceversa.Futureresearchneedstoconcentrateitseffortsona
particulariterationofthetestbattery.
Discussionanddevelopmentopportunity
Fromthematerialpresentedinthispaper,itisapparentthatnobestpracticeinstrumentforthe
selectionofpoliceofficersexistsatthistime.Eachoftheinstrumentsexaminedhaveissuesintheir
supportingresearch.TheMMPIresearchisequivocalatbest,researchwiththeCPIandIPIislimited,
andtheAIFPresearchlacksscale‐levelincrementalvalidity.Otherinstruments,suchasthePAIand
M‐PULSE,aresimplynotusedoftenenoughtogeneratetheresearchrequiredtounderpintheir
endorsement.
Therefore,itappearsthatthereisopportunityforthedevelopmentofabestpracticeinstrument.
Researchdevotedtothisdevelopmentneedstoconsiderthefactorsthatensuretheutilityand
validityoftheinstrument’screationandtesting.Futureresearchshouldensure:
.thedevelopmentofaninstrumentthatcanassessaswidearangeofattitudesand
behaviorsasisreasonable,ratherthanbeingfocussedsolelyonpathologies;
.asufficientsamplesize(>100);
.thatappropriateoutcomevariables,suchason‐the‐jobperformance,areobtainedin
additiontoperformanceduringtraining(e.g.academyperformance);
.thatacontrolgroupormatchedcohortiscomparedtoanygroupselectedusinga(new)
instrument;and
.thatallcomponentsofaninstrument,downtothescalelevel,areadequatelyinvestigated
andexaminedforvalidationpurposes.
Othercriteria,suchastheexaminationofnon‐successfulapplicants’testscores,shouldalsobe
considered(althoughthismaynotbefeasibleinAmericanjurisdictionsduetotheAmericanswith
DisabilitiesAct).Ultimately,theidealinstrumentwouldbeatestthatispurpose‐builtforpolice
selection,ratherthanatestthathasbeenadaptedforuseinthepoliceenvironment.
Ifaninstrumentweretobeselectedasastartingpointforthedevelopmentofabestpractice
instrument,theAIFPtestbatteryappearstobeasoundoption.TheAIFPinstrumentfulfillsmostof
theabovecriteria,withtheexceptionofavailabilityofincrementalvalidationdata.Giventhecurrent
author’sexperiencewiththeAIFPinstrument,itisrecommendedthatathoroughforensic
examinationofthattest,withtheaimofbuildingapsychometricallysoundinstrument,wouldbea
logicallaunchingpadforthedevelopmentofabestpracticeinstrument.
References
Aamodt,M.G.(2004),ResearchinLawEnforcementSelection,BrownwalkerPress,BocaRaton,FL.
Aamodt,M.G.(2010),“Predictinglawenforcementofficerperformancewithpersonalityinventories”,inWeiss,
P.(Ed.),PersonalityAssessmentinPolicePsychology:A21stCenturyPerspective,CharlesC.Thomas,
Springfield,IL,pp.229‐259.
Azen,S.P.,Snibbe,H.M.andMontgomery,H.R.(1973),“Alongitudinalpredictivestudyofsuccessand
performanceoflawenforcementofficers”,JournalofAppliedPsychology,Vol.57No.2,pp.190‐192.
Barret,G.,Polomsky,M.andMcDaniel,M.(1999),“Selectiontestsforfirefighters:acomprehensivereview
andmeta‐analysis”,JournalofBusinessandPsychology,Vol.13No.4,pp.507‐514.
Bartol,C.R.(1982),“Psychologicalcharacteristicsofsmall‐townpoliceofficers”,JournalofPoliceScience&
Administration,Vol.10No.1,pp.58‐63.
Bartol,C.R.(1991),“PredictivevalidationoftheMMPIforsmall‐townpoliceofficerswhofail”,Professional
Psychology,Vol.22No.2,pp.127‐132.
Bartol,C.R.,Bergen,G.T.,Volckens,J.S.andKnoras,K.M.(1992),“Womeninsmall‐townpolicing:job
performanceandstress”,CriminalJusticeandBehavior,Vol.19No.3,pp.240‐259.
Ben‐Porath,Y.S.andTellegen,A.(2008),“EmpiricalcorrelatesoftheMMPI‐2RestructuredClinical(RC)scales
inmentalhealth,forensic,andnonclinicalsettings:anintroduction”,JournalofPersonalityAssessment,Vol.90
No.2,pp.119‐121.
Bernstein,I.H.,Schoenfeld,L.S.andCostello,R.M.(1982),“Truncatedcomponentregressionmulticollinearity
andtheMMPI’suseinapoliceofficerselectionsetting”,MultivariateBehaviorResearch,Vol.17No.1,pp.99‐
116.
Beutler,L.E.,Storm,A.,Kirkish,P.,Scogin,F.andGaines,J.A.(1985),“Parametersinthepredictionofpolice
officerperformance”,ProfessionalPsychology,Vol.16No.2,pp.324‐335.
Blau,T.H.(1994),PsychologicalServicesforLawEnforcement,JohnWileyandSons,NewYork,NY.
Blau,T.H.,Super,J.T.andBrady,L.(1993),“TheMMPIgoodcop/badcopprofileinidentifyingdysfunctional
lawenforcementpersonnel”,JournalofPoliceandCriminalPsychology,Vol.9No.1,pp.2‐4.
Brewster,J.andStoloff,M.L.(1999),“Usingthegoodcop/badcopprofilewiththeMMPI‐2”,JournalofPolice
andCriminalPsychology,Vol.14No.2,pp.29‐34.
Brough,P.(2005),“Acomparativeinvestigationofthepredictorsofwork‐relatedpsychologicalwell‐being
withinpolice,fireandambulanceworkers”,NewZealandJournalofPsychology,Vol.34No.2,pp.127‐134.
Burbeck,E.andFurnham,A.(1985),“Policeofficerselection:acriticalreviewofliterature”,JournalofPolice
Science&Administration,Vol.13No.1,pp.58‐69.
Burkhart,B.(1980),“Conceptualissuesinthedevelopmentofpoliceselectionprocedures”,Professional
Psychology,Vol.11No.1,pp.121‐129.
Butcher,J.N.(2005),ABeginner’sGuidetotheMMPI‐2,2nded.,AmericanPsychologicalAssociation,
Washington,DC.
Byrne,K.(2001),ResearchintotheEffectivenessoftheAIFPCriticalCharacterAssessmentSystemfor
ScreeningNewCustodialCorrectionalOfficers,AIFPAustralia,CliftonHill.
Byrne,K.,Guller,I.andGuller,M.(2002),“Slashingsickleave/attritionratesthroughnewrecruitscreening”,
CorrectionsToday,Vol.64No.5,pp.92‐95.
Caillouet,B.,Boccaccini,M.T.,Varela,J.G.,Davis,R.andRostow,C.(2010),“PredictivevalidityoftheMMPI‐2
PSY‐5scalesandfacetsforlawenforcementofficeremploymentoutcomes”,CriminalJusticeandBehavior,
Vol.37No.2,pp.217‐238.
Casey,S.,Dollard,M.andWinefield,T.(2001),SelectionofCorrectionalServiceOfficers.Work&Stress
ResearchGroup,UniversityofSouthAustralia,Adelaide.
Castora,K.,Brewster,J.andStoloff,M.L.(2003),“PredictingaggressioninpoliceofficersusingtheMMPI‐2”,
JournalofPoliceandCriminalPsychology,Vol.18No.1,pp.1‐8.
Choy,J.(1998),ReducingSickLeaveinCorrectionalOfficers:TheRoleofPsychologicalAppraisal,NewSouth
WalesDepartmentofCorrectiveServices,Sydney.
Cortina,J.M.,Doherty,M.L.,Schmitt,N.,Kaufman,G.andSmith,R.G.(1992),“The“bigfive”personality
factorsintheIPIandMMPI:predictorsofpoliceperformance”,PersonnelPsychology,Vol.45No.1,pp.119‐
140.
Costello,R.M.,Schneider,S.L.andSchoenfeld,L.S.(1996),“Validationofapre‐employmentMMPIindex
correlatedwithdisciplinarysuspensiondaysofpoliceofficers”,Psychology,CrimeandLaw,Vol.2No.1,pp.
299‐306.
Daniels,S.andKing,E.(2002),“ThepredictivevalidityofMMPI‐2contentscalesforsmall‐townpoliceofficer
performance”,JournalofPoliceandCriminalPsychology,Vol.17No.2,pp.54‐60.
Davis,R.andRostow,C.(2008),M‐PULSEinventory:Matrix‐predictiveuniformlawenforcementselection
evaluationinventory.TechnicalManual,MHSInc,Toronto.
Davis,R.andRostow,C.(2010),“TheuseoftheM‐PULSEinventoryinlawenforcementselection”,inWeiss,
P.A.(Ed.),PersonalityAssessmentinPolicePsychology:A21stCenturyPerspective,CharlesC.Thomas,
Springfield,IL,pp.5‐28,132‐158.
Detrick,P.andChibnall,J.(2002),“PredictionofpoliceofficerperformancewiththeInwaldpersonality
inventory”,JournalofPoliceandCriminalPsychology,Vol.17No.2,pp.9‐17.
Dralle,P.andBaybrook,R.M.(1985),“Screeningofpoliceapplicants:areplicationofa5‐itemMMPIresearch
indexvaliditystudy”,PsychologicalReports,Vol.57No.3,pp.1031‐1034.
Gonder,M.andGilmore,D.C.(2004),“Validitystudy:personalityprofilesofpoliceofficerswhosuccessfully
completeacademytraining”,AppliedHRMResearch,Vol.9No.2,pp.59‐62.
Gough,H.G.(1950),“TheFminusKDissimulationIndexfortheMMPI”,JournalofConsultingPsychology,Vol.
33No.1,pp.247‐255.
Guller,I.(1993),AFollow‐UpStudyofAppointedPublicSafetyOfficersandRelationshipofPerformanceto
MeasuredPsychologicalVariables,InstituteforForensicPsychology,Oakland,NJ.
Guller,I.(1994a),EvolutionofaBio‐DataInstrumentfortheScreeningofPoliceandotherPublicSafety
Personnel,InstituteforForensicPsychology,Oakland,NJ.
Guller,I.(1994b),AFollow‐UpStudyofAppointedPublicSafetyOfficersandRelationshipofPerformanceto
MeasuredPsychologicalVariables,paperpresentedattheAnnualMeetingoftheNewJerseyChiefsofPolice,
AtlanticCity,NJ,June22.
Guller,M.(2003),PredictingPerformanceofLawEnforcementPersonnelusingtheCandidateandOfficer
PersonnelSurveyandotherPsychologicalMeasures,unpublisheddoctoraldissertation,SetonHallUniversity,
SouthOrange,NJ.
Graham,J.R.(2006),MMPI‐2:AssessingPersonalityandPsychopathology,OxfordUniversityPress,NewYork,
NY.
Hargrave,G.E.(1985),“UsingtheMMPIandCPItoscreenlawenforcementapplicants:astudyofreliability
andvalidityofclinicians’decisions”,JournalofPoliceScienceandAdministration,Vol.13No.3,pp.221‐224.
Hargrave,G.E.,Hiatt,D.andGaffney,T.W.(1986),“AcomparisonofMMPIandCPItestprofilesfortraffic
officersanddeputysheriffs”,JournalofPoliceScience&Administration,Vol.14No.3,pp.250‐258.
Hargrave,G.E.,Hiatt,D.andGaffney,T.W.(1988),“Fþ4þ9þCn:anMMPImeasureofaggressioninlaw
enforcementofficersandapplicants”,JournalofPoliceScience&Administration,Vol.16No.3,pp.268‐273.
Henson,B.,Reyns,B.,Klahm,C.andFrank,J.(2010),“Dogoodrecruitsmakegoodcops?Problemspredicting
andmeasuringacademyandstreet‐levelsuccess”,PoliceQuarterly,Vol.13No.1,pp.5‐26.
Hiatt,D.andHargrave,G.E.(1988a),“MMPIprofilesofproblempeaceofficers”,JournalofPersonality
Assessment,Vol.52No.4,pp.722‐731.
Hiatt,D.andHargrave,G.E.(1988b),“Predictingjobperformanceproblemswithpsychologicalscreening”,
JournalofPoliceScience&Administration,Vol.16No.2,pp.122‐125.
Ho,T.(2001),“Theinterrelationshipsofpsychologicaltesting,psychologists’recommendations,andpolice
departments’recruitmentdecisions”,PoliceQuarterly,Vol.4No.3,pp.318‐342.
Hogan,R.(1971),“Personalitycharacteristicsofhighlyratedpolicemen”,PersonnelPsychology,Vol.24No.1,
pp.679‐686.
Hogg,A.andWilson,C.(1995),IsthePsychologicalScreeningofPoliceApplicantsaRealisticGoal?TheSuccess
andFailuresofPsychologicalScreening,NationalPoliceResearchUnit,Payneham.
Hooke,J.F.andKrauss,H.H.(1971),“Personalitycharacteristicsofsuccessfulpolicesergeantcandidates”,
JournalofCriminalLaw,CriminologyandPoliceScience,Vol.62No.1,pp.104‐106.
Inwald,R.E.(1988),“Five‐yearfollow‐upstudyofdepartmentalterminationsaspredictedby16pre‐
employmentpsychologicalindicators”,JournalofAppliedPsychology,Vol.73No.4,pp.703‐710.
Inwald,R.(2010),“UseoftheInwaldPersonalityInventory,HilsonTests,andInwaldSurveysforselection,
fitnessfordutyassessment,andrelationshipcounselling”,inWeiss,P.A.(Ed.),PersonalityAssessmentinPolice
Psychology:A21stCenturyPerspective,CharlesC.Thomas,Springfield,IL,pp.5‐28,91‐131.
Inwald,R.E.andShusman,E.J.(1984),“TheIPIandMMPIaspredictorsofacademyperformanceforpolice
recruits”,JournalofPoliceScience&Administration,Vol.12No.1,pp.1‐11.
Kleiman,L.S.andGordon,M.E.(1986),“Anexaminationoftherelationshipbetweenpolicetrainingacademy
performanceandjobperformance”,JournalofPoliceScience&Administration,Vol.14No.4,pp.293‐299.
Lough,J.andRyan,M.(2005),“PsychologicalprofilingofAustralianpoliceofficers:anexaminationofpost‐
selectionperformance”,InternationalJournalofPoliceScienceandManagement,Vol.7No.1,pp.15‐23.
Lough,J.andRyan,M.(2006),“PsychologicalprofilingofAustralianpoliceofficers:alongitudinalexamination
ofpost‐selectionperformance”,InternationalJournalofPoliceScienceandManagement,Vol.8No.2,pp.143‐
152.
Lough,J.andRyan,M.(2010),“Researchnote:psychologicalprofilingofAustralianpoliceofficers:athree‐year
examinationofpost‐selectionperformance”,InternationalJournalofPoliceScienceandManagement,Vol.12
No.3,pp.480‐486.
Lough,J.,Wald,E.,Byrne,K.andWalker,G.(2007),“TheimpactofpsychologicalprofilingofAustralian
correctionalofficers”,CorrectionsCompendium,Vol.32No.4,pp.1‐6.
Macintyre,S.,Ronken,C.andPrenzler,T.(2001),“TheMMPI‐2asatoolforpreventingpolicemisconduct:a
Victorian(Australian)policestudy”,InternationalJournalofPoliceScience&Management,Vol.4No.3,pp.
213‐229.
Meier,R.D.,Farmer,R.E.andMaxwell,D.(1988),“Psychologicalscreeningofpolicecandidates:current
perspectives”,JournalofPoliceScienceandAdministration,Vol.15No.3,pp.210‐215.
Merian,E.M.,Stefan,D.,Schoenfeld,L.S.andKobos,J.(1980),“Screeningofpoliceapplicants:a5‐itemMMPI
researchindex”,PsychologicalReports,Vol.47No.1,pp.155‐158.
Metchik,E.(1999),“Ananalysisofthe“screeningout”modelofpoliceofficerselection”,PoliceQuarterly,Vol.
2No.1,pp.79‐95.
Mufson,D.W.andMufson,M.(1998),“PredictingpoliceofficerperformanceusingtheIPI:anillustrationfrom
Appalachia”,ProfessionalPsychologyResearchandPractice,Vol.29No.1,pp.59‐62.
Murphy,K.andDavidshofer,C.(1994),PsychologicalTesting:PrinciplesandApplications,PrenticeHall,
EnglewoodCliffs,NJ.
Pugh,G.(1985),“TheCaliforniapsychologicalinventoryandpoliceselection”,JournalofPoliceScienceand
Administration,Vol.13No.2,pp.172‐177.
Sarchione,C.,Muchinsky,P.,Nelson,R.andCuttler,M.(1998),“Predictionofdysfunctionaljobbehaviors
amonglawenforcementofficers”,JournalofAppliedPsychology,Vol.83No.6,pp.904‐912.
Saxe,S.andReiser,M.(1976),“AcomparisonofthreepoliceapplicantgroupsusingtheMMPI”,Journalof
PoliceScience&Administration,Vol.4No.4,pp.419‐425.
Schoenfeld,L.S.,Kobos,J.andPhinney,I.R.(1980),“Screeningpoliceapplicants:astudyofreliabilitywiththe
MMPI”,PsychologicalReports,Vol.47No.1,pp.419‐425.
Sellbom,M.,Fischler,G.L.andBen‐Porath,Y.S.(2007),“IdentifyingMMPI‐2predictorsofpoliceofficer
integrityandmisconduct”,CriminalJusticeandBehavior,Vol.34No.8,pp.985‐1004.
Shusman,E.J.,Inwald,R.E.andLanda,B.(1984),“CorrectionofficerjobperformanceaspredictedbytheIPI
andMMPI”,CriminalJusticeandBehavior,Vol.11No.3,pp.309‐329.
Simmers,K.,Bowers,T.andRuiz,J.(2003),“Pre‐employmentpsychologicaltestingofpoliceofficers:theMMPI
andtheIPIaspredictorsofperformance”,InternationalJournalofPoliceScienceandManagement,Vol.5No.
4,pp.277‐294.
Smyth,B.andByrne,K.(1994),“TheUseofSpecializedPsychologicalAssessmentMethodsintheSelectionof
ApplicantstotheVictoriaPolice”,unpublishedmanuscript,AustralianInstituteofForensicPsychology,
Melbourne.
Super,J.T.(2006),“Asurveyofpre‐employmentpsychologicalevaluationtestsandprocedures”,Journalof
PoliceandCriminalPsychology,Vol.21No.2,pp.83‐90.
Surrette,M.A.,Aamodt,M.G.andSerafino,G.(2004),“UsingMMPIspecialscaleconfigurationstopredict
performanceratingsofpoliceofficersinNewMexico”,AppliedHRMResearch,Vol.9No.2,pp.71‐72.
Weiss,P.A.andInwald,R.E.(2010),“Abriefhistoryofpersonalityassessmentinpolicepsychology”,inWeiss,
P.A.(Ed.),PersonalityAssessmentinPolicePsychology:A21stCenturyPerspective,pp.5‐28,CharlesC.
Thomas,Springfield,IL.
Weiss,P.A.andWeiss,W.U.(2010a),“UsingtheMMPI‐2inpolicepsychologicalassessment”,inWeiss,P.A.
(Ed.),PersonalityAssessmentinPolicePsychology:A21stCentury
Perspective,CharlesC.Thomas,Springfield,IL,pp.5‐28,59‐71.Weiss,W.U.andWeiss,P.A.(2010b),“Useof
thepersonalityassessmentinventoryandpoliceandsecuritypersonnelselection”,inWeiss,P.A.(Ed.),
PersonalityAssessmentinPolicePsychology:A21stCenturyPerspective,CharlesC.Thomas,Springfield,IL,pp.
5‐28,72‐90.
White,M.(2008),“Identifyinggoodcopsearly:predictingrecruitperformanceintheacademy”,Police
Quarterly,Vol.11No.1,pp.27‐49.
Winters,C.A.(1990),“ProblemsofvarianceintheutilityoftheMMPIintheselectionofmetropolitanpolice”,
PoliceJournal,Vol.63No.2,pp.121‐128.
Wright,B.S.,Doerner,W.G.andSpeir,J.C.(1990),“Pre‐employmentpsychologicaltestingasapredictorof
policeperformanceduringanFTOprogram”,AmericanJournalofPolice,Vol.9No.4,pp.65‐84.
Abouttheauthors
JonathanLoughisaPhDCandidateattheDeakinUniversity,Melbourne,Australia.HeisalsothePrincipalof
theLoughResearchServices,andaPsychologistandanIndependentResearchConsultantinMelbourne,
Australia.Hespecializesinresearchdesign,analysisandstatistics,surveywriting,andpsychological
assessment.Overthelasttenyears,hehasworkedwitharangeofprivateandgovernmentclients,including
ANZBank,ClemengerBBDO,andtheAustralianInstituteofForensicPsychology.PriortoestablishingLough
ResearchServices,JonathanwasaResearchPsychologistattheMonashUniversity,workingintheareaof
publicsafetyandinjuryprevention.JonathanLoughisthecorrespondingauthorandcanbecontactedat:
jlough@optusnet.com.au
DrKathrynvonTreueristheAssociateHeadoftheSchool(PartnershipsandDevelopment)andtheCourse
ChairfortheOrganisationalPsychologyProgramattheDeakinUniversity.Sheutilizes20yearsofindustry
experienceinorganisationalpsychologyandseniormanagementtoinformherresearch,teachingandservice
rolesattheUniversity.Sheishighlyregardedandsought‐afterbyindustrygroups,andisanationalleaderin
areasoforganisationalpsychologypracticeandresearch.DrKathrynvonTreuercompletedpostgraduate
studiesinpsychology,businessandeducation.Herdoctoraldissertationwasawardedtheprestigious
“ResearchExcellence”awardfromtheSwinburneUniversityin2006,forinvestigatingorganisationalfactors
thatfacilitateworkplaceinnovation.Shewasrecentlyinvitedtoprovideakeynoteaddressonorganisational
resilience,andistheManagingEditoroftheAustralianandNewZealandJournalofOrganisationalPsychology.