USF Libraries
USF Digital Collections

Teachers' sense of efficacy and use of behavioral interventions

MISSING IMAGE

Material Information

Title:
Teachers' sense of efficacy and use of behavioral interventions consultation effects and sustainability
Physical Description:
Book
Language:
English
Creator:
Tanner, Brandi L
Publisher:
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:

Subjects

Subjects / Keywords:
Student behavior
Teacher behavior
Treatment integrity
Teacher beliefs
Intervention training
Dissertations, Academic -- Psychological and Social Foundations -- Doctoral -- USF   ( lcsh )
Genre:
non-fiction   ( marcgt )

Notes

Abstract:
ABSTRACT: The amount of services delivered by school psychologists through consultation is increasing as is the number of students with challenging behaviors in the classroom. In this type of delivery model, the school psychologist works as a consultant to the teacher who will actually deliver the intervention to the student. The purpose of this research was to examine the relationship between participation in a tertiary level behavior intervention program and teacher efficacy, confidence in dealing with challenging behavior, and implementation of behavior interventions. Two studies were conducted to examine a series of research questions. Study 1 used archival data to examine the influence of teacher efficacy before participation in a tertiary level behavior intervention on the amount of coaching necessary for a teacher to implement an intervention with an acceptable level of integrity. To explore sustainability, Study 2 used a survey of teachers who had participated in a tertiary behavior intervention as well as teachers who had not, to determine if they differed on teacher efficacy, confidence in dealing with challenging behaviors, and use of recommended behavioral strategies. Teacher efficacy was not found to be a statistically significant predictor of the amount of coaching time necessary for the teacher to implement the intervention with integrity. It is hypothesized that other factors such as readiness to change may be contributing to the model. PTR participants did not significantly differ from non-participants in any of the proposed areas. It is possible that non-participants over-estimated their knowledge and abilities. Future research should continue to explore the effects of consultation and its sustainability and while considering these additional factors.
Thesis:
Dissertation (Ph.D.)--University of South Florida, 2009.
Bibliography:
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Brandi L. Tanner.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 105 pages.
General Note:
Includes vita.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002028697
oclc - 436297169
usfldc doi - E14-SFE0002805
usfldc handle - e14.2805
System ID:
SFS0027122:00001


This item is only available as the following downloads:


Full Text

PAGE 1

TeachersSenseofEfficacyandUseofBehavioralInterventions:ConsultationEffectsAndSustainabilitybyBrandiL.TannerAdissertationsubmittedinpartialfulfillmentoftherequirementsforthedegreeofDoctorofPhilosophyDepartmentofPsychologicalandSocialFoundationsCollegeofEducationUniversityofSouthFloridaMajorProfessor:HaroldKeller,Ph.D.LindaRaffaele-Mendez,Ph.D.JohnFerron,Ph.D.CarieEnglishPh.D.DateofApproval:March9,2009Keywords:studentbehavior,teacherbehavior,treatmentintegrity,teacherbeliefs,interventiontraining,interventioncoaching,performancefeedbackCopyright2009,BrandiL.Tanner

PAGE 2

iTableofContentsListofTablesiiiAbstractivChapter1Introduction1ConsultationinSchools1TreatmentIntegrity3BehavioralConsultationModel4TeacherFactorsAffectingTreatmentIntegrity6PurposeandSignificanceoftheStudy7Chapter2ReviewoftheLiterature9ConsultationinSchools9HistoryofTreatmentIntegrityLiterature10MeasuringTreatmentIntegrity14ConsultantFactorsInfluencingTreatmentIntegrity16SkillinConsultation16TreatmentAcceptabilityandtheBehaviorSupportPlan16InterventionTrainingProcedures17PerformanceFeedback18SocialInfluence19Negativereinforcement23SummarizationofConsultantFactorsInfluencingTreatmentIntegrity24ConsulteeFactorsInfluencingTreatmentIntegrity25AdultBehaviorChange26TeacherBeliefs27TeacherKnowledgeandUseofBehavioralPrinciplesandInterventions27TeacherEfficacy29TeacherEfficacyandPerceptionsofStudentOutcomes30TeacherEfficacyandTeacherBehaviors31TeacherEfficacyandExpectanciesforConsultation33SustainabilityandGeneralizationofInterventionUsage34PurposeoftheStudy36

PAGE 3

iiChapter3Method37ContextoftheStudy37PreventTeachReinforce(PTR)Overview37BaselinePhaseProcedures39InterventionTrainingandCoachingProcedures39Post-testPhaseProcedures41Study141DataCollection41Recruitment41DataCollectionProcedures43Participants43Instruments46TeachersSenseofEfficacyScale46IntegrityChecklists47ConsultantCostAnalysisSheet47Study249DataCollection49Recruitment49DataCollectionProcedure50Participants51GroupingofParticipants51ParticipantDemographics53Instruments56QuestionnaireaboutTeachersandChallengingBehaviors56Summary58Chapter4Results59Study159DataAnalysis59DescriptiveStatistics60DataScreening61Results62Study265DataAnalysis66DescriptiveStatistics66DataScreening68MANOVAResults68ANOVAResults69Summary72Chapter5Discussion72Summary72Results74ComparisonofStudy1toPreviousResearch75LimitationsofStudy178ComparisonofStudy2toPreviousResearch79

PAGE 4

iiiLimitationsofStudy281ImplicationsandFutureResearch81References84Appendices91AppendixA:TeachersSenseofEfficacyScale92AppendixB:RecruitmentE-mailforStudy292AppendixC:Study2GroupAssignmentGuidelines94AppendixD:Study2Survey95AbouttheAuthorEndPage

PAGE 5

ivListofTablesTable1OverviewofPTRActivitiesandDataCollectionbyPhase40Table2Study1ParticipantCredentialsandExperience44Table3Study1ParticipantTeachingAssignment45Table4Study2ParticipantCredentials53Table5Study2ParticipantExperience54Table6Study2ParticipantTeachingAssignment55Table7Study1DescriptiveStatisticsforTeacherEfficacyandCoachingTimeVariables60Table8Means,StandardDeviations,andIntercorrelationsamongTeacherEfficacyandCoachingTimeVariables61Table9RegressionResultsforNumberofSessionsandEfficacySubscales63Table10RegressionResultsforConsultantCoachingTimeandEfficacySubscales64Table11SummaryofResultsofRegressionAnalyses65Table12Study2MeansandStandardDeviationsforDependentVariablesbyGroup67Table13ResultsofANOVAforEachDependentVariable70

PAGE 6

vTeachersSenseofEfficacyandUseofBehavioralInterventions:ConsultationEffectsandSustainabilityBrandiL.TannerABSTRACTTheamountofservicesdeliveredbyschoolpsychologiststhroughconsultationisincreasingasisthenumberofstudentswithchallengingbehaviorsintheclassroom.Inthistypeofdeliverymodel,theschoolpsychologistworksasaconsultanttotheteacherwhowillactuallydelivertheinterventiontothestudent.Thepurposeofthisresearchwastoexaminetherelationshipbetweenparticipationinatertiarylevelbehaviorinterventionprogramandteacherefficacy,confidenceindealingwithchallengingbehavior,andimplementationofbehaviorinterventions.Twostudieswereconductedtoexamineaseriesofresearchquestions.Study1usedarchivaldatatoexaminetheinfluenceofteacherefficacybeforeparticipationinatertiarylevelbehaviorinterventionontheamountofcoachingnecessaryforateachertoimplementaninterventionwithanacceptablelevelofintegrity.Toexploresustainability,Study2usedasurveyofteacherswhohadparticipatedinatertiarybehaviorinterventionaswellasteacherswhohadnot,todetermineiftheydifferedonteacherefficacy,confidenceindealingwithchallengingbehaviors,anduseofrecommendedbehavioralstrategies.

PAGE 7

viTeacherefficacywasnotfoundtobeastatisticallysignificantpredictoroftheamountofcoachingtimenecessaryfortheteachertoimplementtheinterventionwithintegrity.Itishypothesizedthatotherfactorssuchasreadinesstochangemaybecontributingtothemodel.PTRparticipantsdidnotsignificantlydifferfromnon-participantsinanyoftheproposedareas.Itispossiblethatnon-participantsover-estimatedtheirknowledgeandabilities.Futureresearchshouldcontinuetoexploretheeffectsofconsultationanditssustainabilityandwhileconsideringtheseadditionalfactors.

PAGE 8

1Chapter1IntroductionConsultationinSchoolsConsultationisincreasingasaservicedeliverymodelforpsychologistspracticinginschoolstodayandisexpectedtoexpandsignificantlyoverthenextfewdecades(Wilkinson,2006).Whenschool-basedconsultationemergedinthe1970sandearly1980s,itwasviewedasavoluntarycollaborativerelationshipbetweenco-equalprofessionalswithteachershavingtherighttorejectormodifytheconsultantssuggestions(Martens&Ardoin,2002).Inresponsetochangesinfederalregulationsandparadigmshiftsinthefieldofspecialeducationbeginninginthe1980s,statesbegantomandatepre-referralinterventionswithstudentsreferredforspecialeducation.Withthepassageofthe1997amendmentstotheIndividualswithDisabilitiesEducationACT(IDEA1997),consultationhasalsobecomeamodelthroughwhichpositivebehavioralsupportservicesareprovidedforstudentswithidentifieddisabilities.Oneofthebiggestchallengesfacingschoolsprovidingservicesthroughaconsultativeservicedeliverymodelisthefailureoftheschoolstafftoimplementandsustaininterventionsdevelopedinconsultation.Theproblemisnotthattheconsultationprocessisflawed;rathercurrentschoolsystemsarenotdesignedtosupporttheconsultativeprocess(Foxx,1996;Lewis&Newcomer,2002).Contextualfactors(e.g.

PAGE 9

2administrativepriorities,timeconstraints,scheduling)withintheschoolinfluencetheeffectsofconsultation.Wheninterventionisnotvaluedbytheschoolsleadershipandresourcesarescarce,eventhemostpowerfulconsultationproceduresmaynotproducethedesiredresults.Ifteachersareheldaccountablebyadministratorsandinterventionimplementationistiedtoevaluationsorotherjobperformancefactors,theyshouldbemorelikelytoimplementinterventions.However,consultantsrarelyhaveadministrativeauthorityovertheteacher(Martens&Ardoin,2002;Noell&Witt,1999).Talk,intheformofinterviews,collaboration,makingrecommendations,andattendingmeetingsdescribeswhatconsultationtypicallylookslikeinschools.Inhiscommentary,Witt(1997)challengesthevalueoftalkbyquestioningthefunctionalrelationshipamongthreekeyparts:(a)whatissaidtotheteacherduringconsultation,(b)whetherthattalkistranslatedintobehaviorchangefortheteacher,(c)ifthebehaviorchangeoftheteacheriscorrelatedwithbehaviorchangeinthechild.Wittarguesthatifthetalkdoesnotleadtoachangeinbehaviorbytheteacherandsubsequentchangeinbehaviorbythechild,wearewastingtimeandresourcesonineffectiveactivity.Inadditiontothefinancialcost,suchasystemisprovidingadisservicetochildrenbyassumingthattalkfunctionstoaffordthestudentastandardofcare.Insteadofrelyingmerelyontalkandself-reportofinterventionimplementation,Wittadvocatesforaccountabilitythroughtrainingtheteacherontheintervention,monitoringimplementation,usingdirectobservationorpermanentproductstomeasuretreatmentintegrity,andmeasuringstudentperformance.Sincetheconsultationprocessistimeconsumingforboththeconsultantandtheteacher,accountabilitymeasuresshouldbein

PAGE 10

3placetomakesurethattheconsultationactivitiesleadtothedesiredoutcomes(Foxx,1996;Witt,1997).Witt(1997)referstoafunctionalrelationshipinvolvingwhattheconsultantsaysinconsultation,teacherbehaviorchange,andstudentbehaviorchange,inwhichtheteachermustimplementinterventionswithintegrity.Inorderforthetimespentinconsultationdevelopingabehaviorinterventionplantopayoff,anadditionalinvestmentmustbemadeintrainingtheteachertoimplementtheinterventionsandmonitoringthatitisdonewithahighdegreeoftreatmentintegrity.Teacherefficacyandbeliefsmayalsobesettingeventsorpreimplementationfactorsthataffectthisfunctionalrelationship(Han&Weiss,2005).Theteachermustbelievethatheorsheiscapableofimplementingtheinterventionandhisorherbehaviorinimplementationwillleadtothedesiredchangeinthestudent.Thus,teacherefficacyandbeliefsmayimpacttheamountoftimetheconsultantmustinvestincoachingtheteacherinimplementingtheinterventionwithahighleveloftreatmentintegrity.Consequently,therelationshipbetweenteacherefficacyandtreatmentintegritywarrantsinvestigation.Likewise,theproductofinterventionpracticesshouldbeevaluatedattheteacherlevel,aswellasthestudentlevel,aschangesintheteachersbeliefsandbehaviorareanecessarypartoftheprocess.TreatmentIntegrityWiththepassageoftheNoChildLeftBehindAct,(U.S.DepartmentofEducation,2002)teachersarebeingheldaccountablefortheirpracticesmorenowthananytimeinthepast.Thisaccountabilityisforcoreinstructionaswellasstrategicandintensiveinterventions.Whenschool-basedinterventionsaredesigned,implementation

PAGE 11

4istoooftenassumedtobeaccurate.Treatmentintegrityrefersthedegreetowhichinaninterventionisimplementedasdesigned(Gresham,1989;Gresham,Gansle,Noell,Cohen,&Rosenblum,1993).Thisconstructneedstobeassessedinsteadofassumedifwearetodrawaccurateconclusionsaboutinterventionoutcomesasteachersarenotalwaysaccurateinself-reportingtheirleveloftreatmentintegrity(Wickstrom,Jones,LaFleur,&Witt,1998).Demonstratingtreatmentintegrityrepresentsoneofthemajoraspectsofscientificinvestigationandpracticalapplicationsofinterventionsinschoolsettings.Treatmentintegrityshouldbeconsideredanimportantlinkbetweentheuseandeffectivenessofinterventionsinschoolsettings(Gresham,etal.,1993).Aclearunderstandingofhowtheinterventionwasappliedisnecessaryfortheteacherortheotherprofessionalstomakevalideducationaldecisions.Theeffectivenessofaninterventioncannotbejudgediftheinterventionisnotimplementedcorrectly(Gable,Hendrickson,&VanAker,2001).Ifthestudentsbehaviordoeschange,theconsultantmustdetermineiftheeffectsarebecauseoftheinterventionsorduetootherextraneousfactors.Ifthedesiredbehaviorchangesdonotoccur,itmaybebecauseofinappropriatetreatment,oraninappropriatelyimplementedintervention.Thus,knowledgeoftreatmentintegrityisnecessarytomakethedistinctionbetweenanineffectiveinterventionandapotentiallyeffectiveplanthatwaspoorlyimplemented(Gable,etal.,2001;Sterling-Turner,Watson,Wildmon,Watkins,&Little,2001).Anadditionalpracticalconsiderationisthatifpeoplefailtofullyandconsistentlyimplementanintervention,thetargetbehaviorwillpersistandislikelytobecomemoreresistanttoextinction(Gable,etal.,2001).

PAGE 12

5BehavioralConsultationModelTodate,behavioralconsultationhasbeenconsistentlyidentifiedasthemosteffectiveandpreferredmodelforschool-basedconsultation(Sheridan,Welch,&Orme,1996;Wickstrom,etal.,1998;Wilkinson,2006).ThismodeldevelopedbyBerganandKratochwill(1990)istypicallydescribedasafour-stageproblem-solvingprocess,withinterviewsincludingproblemidentification,problemanalysis,treatmentimplementation,andtreatmentevaluation.Previousresearchhasevaluatedbehavioralconsultationintermsofteacherperceptionoracceptability;rarelyhasitbeenevaluatedthroughmeasurementofinterventionimplementationandtreatmentintegrity(Wickstrom,etal.,1998).AlthoughBerganandKratochwill(1990)originallyemphasizedverbalinstructionbytheconsultantpriortoimplementationasameansofpromotingtreatmentintegrity,datasuggestthatteachersfailtoimplementorsustaintreatmentplanswithoutongoingsupportfromtheconsultant(DiGennaro,Martens&McIntyre,2005).ReferredtoastheconsultandhopestrategybyWilkinson(2006),thismethodologyimpedesourabilitytoestablishfunctionalrelationshipsbetweenthetreatmentandtheoutcomesandthereforeisproblematic.Forsometime,criticshavequestionedtheutilityofthebehavioralconsultationmodelinensuringtreatmentintegrity.Specifically,thereislackofconclusiveevidencethatusingbehavioralconsultationprocedureswillleadtheteachertoreturntotheclassroomandengageintheinterventionprocedureswithahighdegreeoffidelitythatwillresultinchildbehaviorchange(Witt,Gresham,&Noell,1996).Morerecently,Wilkinson(2006)describedaproposedexpansionofthetraditionalbehavioralconsultationprocesstoprovideconsultantswithamethodof

PAGE 13

6assessingandmonitoringtreatmentimplementationandintegritywithinterventionplans.Thismodificationincludestheadditionofasteptoreviewstudentprogress,identifybarrierstoimplementationandmodifytheplanifneeded.Thisstepalsoincludesaperformancefeedbacksessiontoanalyzetreatmentintegritydatawithpraiseorfeedback,determinetheneedforfurthertrainingorsupport,andtoreaffirmtheconsulteescommitmenttoimplementingtheplan.TeacherFactorsAffectingTreatmentIntegrityStudentsarriveintheclassroomwithcharacteristicsthatwillrequiredifferenttypesandlevelsofteachersupport.Similarly,teachersmayalsohavesomecharacteristicsthatwillrequiredifferentialtypesandlevelsofconsultantsupporttoimplementaninterventionwithintegrity.Thesecharacteristicsmayincludeskilllevel,motivation,ortheperceivedandactualeffectivenessoftheintervention(Lane,etal.,2004).Teachercompliancewiththetasksrelatingtotheconsultationitselfsuchascollectingdata,activelyparticipatingininterviews,andimplementingtheinterventionwithintegritymayalsobepredictiveoftheleveloftreatmentintegrityandpotentialstudentoutcomes(Lewis&Newcomer,2002).Teacherefficacyreferstotheteachersjudgmentabouthisorhercapabilitiestobringaboutdesiredoutcomesofstudentengagementandlearning,evenamongthosestudentswhomaybedifficultorunmotivated(Bandura,1977;Tschannen-Moran&Hoy,2001).Teacherefficacyhasbeenfoundtoberelatedtotheteachersattitudesandbehaviorintheclassroomandtostudentoutcomessuchasachievementandthestudentsownsenseofefficacy(Tschannen-Moran&Hoy,2001).HanandWeiss(2005)notethat

PAGE 14

7beliefsandtheoreticalorientationareimportantbecausetheyaffecttheteachersmotivationandeffortduringimplementation.Teacherfactorsarealsoevidentaftertheconclusionoftheconsultation.Studiesonmaintenanceandgeneralizationofbehavioralinterventionshaveproducedmixedresults.Ingeneral,datacollectedatfollow-upfrequentlydonotindicatecompletemaintenanceofskillsastreatmentintegritylevelsfrequentlydropafterremovaloftheconsultant(Rose&Church,1998).However,useofperformancefeedbackandpracticeintheteachersclassroomisassociatedwithhigherlevelsofmaintenancethanothertypesoftraining.Directresearchonmaintenancewiththisstrategyismixed.PurposeandSignificanceoftheStudyThepurposeofthisresearchwastoexaminetherelationshipbetweenparticipationinatertiarylevelbehaviorinterventionprogramandteacherefficacy,confidenceindealingwithchallengingbehavior,andimplementationofbehaviorinterventions.Twostudieswereconductedtoexamineaseriesofresearchquestions.Study1usedarchivaldatatoexaminetheinfluenceofteacherefficacybeforeparticipationinatertiarylevelbehaviorinterventionontheamountofcoachingnecessaryforateachertoimplementaninterventionwithanacceptablelevelofintegrity.Toexploresustainability,Study2usedasurveyofteacherswhohadparticipatedinatertiarybehaviorinterventionaswellasteacherswhohadnot,todetermineiftheydifferedonteacherefficacy,confidenceindealingwithchallengingbehaviors,anduseofrecommendedbehavioralstrategies.Arelationshipbetweenteacherefficacyandtreatmentintegrityhasmultipleimplicationsforpractice.Ifconsultantswereabletoidentifyteacherswhoweremore

PAGE 15

8likelytoimplementinterventionsandsustainahighdegreeoftreatmentintegrity,theconsultantcouldquicklybegininterventiontrainingreducingtheamountofcoachingtimeneeded.Forteacherslesslikelytoimplementinterventionscorrectlyandsustainintegrity,consultantscouldusesomeconsultationtimeearlyintheprocessaddressingpreimplementationfactorsandprogrammingforsuccessfulimplementation(Han&Weiss,2005).Thesoonertheteachercorrectlyimplementstheintervention,thesoonerstudentoutcomescanimprove.Ineithercasescenario,thestudentreceivesacorrectlyimplementedinterventionmoreefficiently,freeingupmorevaluabletimefortheconsultanttoengageinconsultationwithmoreteachersandstudents.Additionally,ifalinkbetweenteacherefficacyandtreatmentintegritywerefound,schoolscouldchoosetouseasystemsapproachintoaddressteacherefficacyinadditiontotargetingindividualstudents(Meijer&Foster,1988).Ingeneral,schoolsandconsultantscouldusethisinformationtomaximizetheeffectivenessandefficiencyofconsultationinschoolsandtoimproveoutcomesforstudents.

PAGE 16

9Chapter2ReviewofLiteratureThisliteraturereviewwillbeginwithadiscussionofschool-basedconsultationandtheaccompanyingchallengesthisservicedeliverymodelposesforconsultants.Next,thehistoryoftreatmentintegrityinschool-basedconsultationispresented.Advantagesandlimitationsofdifferentmethodstoassesstreatmentintegrityarehighlightednext.Then,literatureregardingstrategiestheconsultantmayusetoincreasetreatmentintegrityisdiscussed.Consulteevariables,includingteacherefficacy,whichalsomayaffecttreatmentintegrityarepresentednext.Thereviewcloseswithadiscussionofsustainabilityofinterventions.ConsultationinSchoolsDuetotherecentparadigmshiftsandlegislationdesignedtoeducatemorestudentsingeneraleducationsettings,manyservicesprovidedbyspecialistssuchasschoolpsychologistsandbehaviorspecialistsarebeinghandledthroughconsultation.Thoughspecialistsmayassistindevelopinginterventionplans,consultationisaservicedeliverymodelinwhichtheteacheristheprimarytreatmentagent.Likewise,asconsultationisaservicedeliverymodel,itisnotaprocedurewhichhasbeentestedandproventhroughabodyofempiricalevidence(Lewis&Newcomer,2002).Whenmeasuringtheeffectivenessofschoolconsultation,itisimportanttomeasurethreeareas:(a)changesintheteachersknowledgeandorattitudes,(b)changesintheteachers

PAGE 17

10classroombehavior,and(c)changesinthestudentsbehaviorintheclassroom(Witt,1997).Inorderforthestudenttobenefitfromtheeducationalconsultation,theteachermustchangehisorherbehaviorasafunctionoftheconsultationcontact.Thus,interventionimplementationisacrucialchallengeforconsultantsworkinginschools.Treatmentintegrityisthedegreetowhichaninterventionisimplementedasdesigned(Gresham,1989;Gresham,etal.,1993).Thoughprogressisbeingmade,researchtodatehasindicatedthatrelativelylittleisknownabouttheextenttowhichteachersimplementtheinterventionsfollowingconsultation.HistoryofTreatmentIntegrityLiteratureItappearsacceptabletoreasonthatpoorlyimplementedinterventionsarelesslikelytobeeffective(Han&Weiss,2005).Althoughanintuitiverelationexistsbetweentreatmentintegrityandtreatmentoutcomes,researchexaminingthisphenomenonislimited(Noell,etal.,2005).Thissectionwillpresentaseriesofmeta-analysesthatchroniclethehistoryoftreatmentintegrityandoutcomeresearchinconsultation.Ameta-analysiswasconductedspecificallyexaminingtreatmentintegrityinschool-basedconsultation(Gresham,etal.,1993). Thisarticlereviewed181experimentalinterventionstudieswhichtookplaceinschoolsettingspublishedfromasampleofbehavioraljournalsbetween1980and1990.Treatmentintegritywascodedintooneofthreecategories.AstudywasassignedtotheYescategoryifitassessedtreatmentintegrityexpressedasapercentage,Noifthestudymadenomentionoftreatmentintegritychecks,andMonitoredifitmentionedtreatmentintegritybutprovidednonumericalindex.Only14.9%(27studies)systematicallymeasuredandreported

PAGE 18

11levelsoftreatmentintegrityfortheinterventionsimplemented.Anadditional9.9%(18studies)monitoredtreatmentintegritybutdidnotprovidedataonit.Theremainingthree-quartersofstudies(135studies)didnotaddresstreatmentintegrity.Ofthosestudiesreportingtreatmentintegritydata,meanintegritylevelsrangedfrom75%to100%withameanof96.92%consideredtobeconsistentwithpreviousstudies.Dataindicatedthathighertreatmentintegritywasassociatedwithlargereffectsizes.TheworkofGresham(1993)wasrecentlyextendedbyfifteenyearstoreviewstudiespublishedintheJournalofAppliedBehaviorAnalysis(JABA)between1991and2005(McIntyre,Gresham,DiGennaro,&Reed,2007).Atotalof152studiesin142articleswerereviewed.Treatmentintegritydatawasprovidedintheformofpercentageofimplementationin30%ofthestudies(46studies)withanadditional9%(13studies)reportingmonitoringoftreatmentintegrity.Neithertreatmentintegritylevelsnormonitoringoftreatmentintegritywerereportedin61%(93studies).Thoughthepercentageofstudiesreportingtreatmentintegritydatawassomewhathigherinthisreview(30%)ascomparedtotheprevioussampleofstudiesinthisseries(20%),treatmentintegrityreportinghasgenerallyremainedstableovertheyears.Sheridanandcolleagues(1996)examinedtheliteratureonschool-basedconsultationresearchfrom1985to1995.Thepurposesofthismeta-analysisweretodeterminehowmuchempiricalliteraturehadbeenproduced,whatcommonmethodologicalfeatureswerepresent,andtoexamineconclusionsregardingoutcomeresearch.Forty-sixstudiesmetinclusioncriteria.Schoolconsultationprovedatleastpartiallyeffectiveinaboutthree-fourthsofstudies.Manystudiesreviewedhowtheconsultationservicesweredelivered;however,fewmonitoredtheintegritywithwhich

PAGE 19

12theinterventionwasimplemented.Treatmentintegritywasassessedinonly26%ofthestudies.Gansle(2005)focusedspecificallyonstudiesofschool-basedangerinterventionprogramsandusedacodingschemesimilartothestudybyGreshamandcolleagues(1993)mentionedaboveforintegrity.Ofthe20articlesreviewed,onlytwoarticlesmeasuredintegrity,onebyself-reportandanotherbyacombinationofmeasurementtypes.Forthesetwostudies,themeanintegritywas92.5%.Sixstudiesmentionedtheimportanceoftreatmentintegrity,while60%(12studies)didnotmentionoraddresstreatmentintegrity.Theoverallmeandifferenceeffectsizewas.31indicatingadifferenceof.31standarddeviationsbetweentreatmentandcontrolgroupsorpre-andpost-intervention.Theseoutcomesweresimilartooutcomesofotherstudiesinvolvingangerinterventions.CochraneandLaux(2008)recentlyconductedasurveyof806professionalsholdingtheNationallyCertifiedSchoolPsychologist(NCSP)credentialregardingthemeasurementoftreatmentintegrityanditsimportance.Inthissurvey,11.3%indicatedthattheyalwaysmeasuredtreatmentintegrityand41.6%saidtheysometimesmeasuredtreatmentintegrity.Ofthoseindicatingtheyalwaysorsometimesmeasuredtreatmentintegrity,directobservationwasreportedtobeused25.3%ofthetimewithinterviewingasthemajorityused60.6%ofthetime.Theauthornotedinthediscussionthatinherexperience,interviewingmeantonlyaskingtheteacherattheteammeetingifheorsheimplementedtheinterventionasitwaswrittenintheinterventionplan(p.504).

PAGE 20

13Despitethefactthat33.5%ofthosesurveyedindicatedtheynevermeasuredtreatmentintegrity,56.2%statedtheystronglyagreedthattreatmentintegrityisakeyfactortoconsiderinevaluatingthesuccessofaninterventionand97.6%agreedorstronglyagreedthatitwascriticaltoincludetreatmentintegrityinformationwhendeterminingspecialeducationeligibility.Seventypercentoftheparticipantschosetoansweranoptionalopen-endedquestionattheendofthesurveytoexplaintheiropinionregardingtheimportanceoftreatmentintegrityinschool-basedinterventions.Participantsindicatedthatlackoftime,training,administrativesupport,andteacherresistancetomeasurementasthereasonsthattreatmentintegrityisnotmeasured.AuthorsofthisstudynoteasalimitationthattheuseoftheNCSPdatabaseasaconveniencesampleandthattheresultsofthoseholdingthiscredentialmaydifferfromnon-NCSPschoolpsychologists.Likewise,theresponseratewas11.6%anditispossiblethatthosewhochosenottoparticipatedidnotunderstandtreatmentintegrityorbelievethatitisimportant.Insummary,theliteratureexaminingtreatmentintegrityfromthepasttwodecadesindicatesalackofsignificantattentiontothisimportanttopic.Onlyaboutone-fourthofstudieshaveassessedandmeasuredtreatmentintegritywithslightlymorediscussingitsimportance.Themajorityofstudiesofschool-basedconsultationhavenotaddressedtreatmentintegrity.Thoughschoolpsychologists,indicatethattreatmentintegrityisimportant,measurementisstilllow.Consultationserviceshaveyieldedatleastpartiallyeffectiveoutcomesforstudentswithhigherdegreesoftreatmentintegritybeingassociatedwithlargereffectsizes.

PAGE 21

14MeasuringTreatmentIntegrityAssuringplanimplementationinconsultationandbehaviortherapymayfrequentlybemorechallengingthanthedevelopmentofthetreatmentplans(Foxx,1996;Noell,etal.,2005).Asnotedintheabovesection,interventionimplementationintheearlyliteraturewasnotaprimaryfocusofinvestigationinconsultation.Thisassumptioncanbesupportedbythelackofdirectmeasurementoftreatmentintegritywithrelianceinsteadonsatisfactionratingsandinformantreportstoassesstreatmentintegrity(Gable,etal.,2001;Noell&Witt,1999).Satisfactionreportsprovideevidenceoftheconsulteesperceptionoftheconsultationprocess,notwhattheyactuallydidasaresultoftheconsultation.Whileteachersatisfactionisdesirable,itdoesnotdemonstratethattheconsultationledtoeffectivedeliveryoftheinterventiontargetedforthestudent.Thereareseveraldifferentmethodswhichconsultantscanusetodocumentthedegreeoftreatmentintegrity(Wilkinson,2006).Systematicobservationofinterventionimplementationisthemostdirectmeansofassessingtreatmentintegrity.Lessdirectmethodsincludeconsulteeself-reportsandpermanentproducts.Eachmethodofintegritymeasurementhasadvantagesanddisadvantages.Observationisthemostdirectmeansofassessingtreatmentintegrityandthusprovidesthemostaccurateaccountofinterventionimplementation(Noell&Witt,1999;Wickstrometal.,1998).Thelimitationsofthismethodarethatitcanbelaborintensiveandnotfeasiblewithrealworldtimeconstraints.Thisprocedureisalsovulnerabletoreactivityeffectsastheconsulteeknowstheyarebeingobserved.Self-reportsincludemeasuressuchasLikert-typescalesorchecklistsfortheconsulteetocompletetoindicateifthestepsoftheinterventionwerefollowed.Though

PAGE 22

15thesereportssavetimefortheconsultant,theymayproduceinaccuratereportsandoverestimatetheleveloftreatmentfidelity.Withpermanentproductassessment,aproductisgeneratedthatistheresultoftheinterventionandevaluatedtodeterminetheextenttowhichthecorrespondinginterventionwasimplemented.Whilethismethodofassessingtreatmentintegrityalsosavestimeandreducesreactivity,limitationshavebeenexpressedbecauseitispossibletohavesomepermanentproductseventhoughtheinterventionmaynothavebeenimplementedcorrectly(Noell,etal.,2002).Limitationsofself-reportmeasureoftreatmentintegrityhavebeendocumentedbymultiplestudies.RobbinsandGutkin(1994)conductedamultiple-baselineacrosssubjectsdesignstudywiththreesecondgradeteachers.Inpost-treatmentinterviewswiththeconsultant,allthreeteachersself-reportedthattheyimplementedtheplannedintervention(positiveverbalizations).However,observationaldatadidnotsupportthisconclusion.Noneofthethreeteachersevidencedanysubstantialincreaseintheuseofpositiveteacherverbalizationsdirectedtowardthetargetchilddespiteverylowlevelsduringbaseline.Authorsconcludedthatconsulteesself-reportsoftreatmentimplementationmaynotbeadequateformeasuresofactualbehavior.Wickstorm,andcolleagues(1998)investigateddifferencesintreatmentintegritymeasureswith29elementaryteacherswithstudentswithdisruptivebehaviors.Alltreatmentplansweredevelopedtoincludeanobservablecomponenttoenabledirectmeasurementofintegrity.Themeanintegrityofrecordkeepingorselfreportwas54%.Thepresenceofinterventionproductswasobservedin62%ofcases.Whenmeasuredbydirectobservation,teachersonlyimplementedtheprogrammedconsequencetoproblem

PAGE 23

16behaviorin4%ofinstances.Theseresultsfurtherreducethecredibilityofmeasuringtreatmentintegritythroughself-reportorthroughpermanentproduct.Insummary,researchhasnotsupportedtheuseofindirectmeasuresoftreatmentintegrityincludingself-reportorpermanentproducts.NoellandWitt(1999)arguethatduetothelackofcorrespondencebetweenwhatteacherssayanddoandpooragreementbetweenteacher-reportedlevelsofimplementationandwhatisobserved,itisinappropriatetoacceptanindirectindicatorforsomethingthatcanbemeasureddirectly.ConsultantFactorsInfluencingTreatmentIntegritySkillinconsultation.Somefactorsaffectingefficacyofschool-basedconsultationarerelatedtotheconsultantincludingconsultantstyleandconsultationmodelused.Theconsultantcanmitigatethesefactorsbypossessingalevelofskillinbothcontentoftheproblemandinconsultationskill(Lewis&Newcomer,2002).Recentliteraturesuggeststhattheconsultantbeanactiveparticipantnotonlyinthedevelopmentoftheintervention,butalsointrainingteachersonhowtousetheinterventionandinmonitoringimplementation.Treatmentacceptabilityandthebehaviorsupportplan.Identifyinganeffectivetreatmentisanecessarybutnotsufficientstepforchildbehaviorchange.Factorsrelatedtotheinterventionitself,suchascomplexity,andthebehaviorsupportplansalsomayinfluencecompliancewiththeinterventionstrategiesandtreatmentintegrity(Weigle&Scotti,2000).Whenselectingtheintervention,theconsultantshouldtaketreatmentacceptabilityintoconsideration.Iftheteacherdoesnotliketheselectedintervention,heorshewillbelesslikelytoimplementitcorrectly.Interventioncomplexity,thetimeand

PAGE 24

17materialsrequiredforimplementation,andtherequirementofextrapersonnelaffecttreatmentacceptability(Lane,Bocian,MacMillan&Gresham,2004).Interventiontrainingprocedures.Becauseteachersmaynothavetherequisiteskillsorknowledgeneededtoimplementtheinterventionasprescribed,theconsultantisoftenneededtotrainothersonhowtoimplementthebehaviormanagementintervention(Sterling-Turner,etal.,2001).Consultantstypicallyunderestimatetheamountoftrainingandsupportthatwillberequiredtosuccessfullychangethebehavioroftheteachers,resultinginthedesiredstudentbehavior(Martens&Ardoin,2002).Manystrategiestotrainteacherstoimplementinterventionsandtoprovidefollowupormonitorinterventionimplementationhavebeensuggested.Someexamplesincludesocialinfluencestrategies,settinggoalsforteacherbehavior,modeling,coaching,performancefeedback,anduseofimplementationprotocols.Othersuggestionsincludepreciselydelineatingwhoisresponsibleforeachpartoftheplananddescribingwhatconditionstheplanshouldbeusedunder(Gable,etal.,2001).Scriptingouteachcomponentoftheplanandrehearsingtherolesalsoisbelievedtoincreasetheaccuracyofimplementation.Inaclinicalstudy,researchersmanipulatedthetypeoftrainingproceduresusedtomeasuretheeffectontreatmentintegrity(Sterling-Turner,etal.,2001).Participantswere64undergraduatestudentsinformedthattheywouldbetrainedtoimplementatreatmentprotocolwithaclientwithafacialtic(actuallyaconfederateofthestudy).Participantswererandomlyassignedtooneofthreeexperimentalconditions.Inthedidactictrainingcondition,thetreatmentprocedureswereverballyexplainedtotheparticipantalongwithdemonstrationoftargetedbehaviors.Inthemodelingcondition,participantswatcheda

PAGE 25

18videotapeofatreatmentsessionwithverbalexplanations.Intherehearsalfeedbackcondition,theparticipantreceivedtrainingwiththeexperimenterandtheconfederateincludingverbalprompts,correctingofmistakes,andcontingentpraise.Regardlessofcondition,alltrainingsessionslastedfiveminutes.Participantswhoreceivedthemostdirecttraining(rehearsalfeedbackcondition),implemented84%oftreatmentcomponentscorrectly,higherthanparticipantstrainedusingindirectprocedures.Participantsinthemodelingconditionachievedameanof70%oftreatmentcomponentsimplementedcorrectly,whilethedidacticconditionachievedanaverageof51%.Theauthorssuggestthatfailuretoachievethedesiredstudentgoalsduringconsultationmaybearesultofinadequatetrainingoftheconsulteebytheconsultant. SimilarresultswerefoundbyRoseandChurch(1998)inameta-analysisinvolving49studiesthatuseddirectobservationtoprovidedataontheeffectsofprocedurestochangeteachingbehaviors.Allstudiesthatusedroleplaywithpeersproducedchangeinteacherbehavior.Practicewithfeedbackwasconsistentlyfoundtoproducethestrongesteffects.Noevidencewasfoundtosuggestthatdidacticinstructionalonewillproducechangesinteacherbehavior.Performancefeedback.Researchershavenotedtheimportanceofperformancefeedbacknotingthatthestrongesteffectsoftrainingarefoundwhenclassroompracticeiscombinedwithperformancefeedback(Han&Weiss,2005).Noellandcolleaguesconductedaprogramofresearchinvestigatingtheeffectsofdifferenttypesoffollow-upstrategiestobeusedinbehavioralconsultation(Noell,etal.,1999;Noell,etal.,2000;Noell,etal.,2002;Noell&Witt,2005;Witt,etal.,1997),primarily,performance

PAGE 26

19feedback.Performancefeedbackconsistsofmonitoringabehaviorthatisthefocusofconcernandprovidingfeedbacktotheindividual.Goalsetting,performancecontingencies,andgraphicdisplayshavebeenfoundtoenhancetheefficacyofperformancefeedback(Noell,etal.,2005).Theresearchteamreportedontheimplementationofabehaviormanagementinterventionbyfourgeneraleducationteachersforeightstudentswithdisruptiveandchallengingbehavior(Noell,etal.,2002).Thestudyexaminedimplementationofawholeschooldaybehavioralintervention.Teacherswereinvolvedinthedevelopmentoftheplan,trainedintheintervention,andprovidedwithmaterialsincludingarewardbox.Allstudentsinterventionplansincludedmonitoringtargetbehaviors,recordingtheiroccurrenceoftargetbehaviors,goalsetting,andrewardsformeetinggoals.Thebehaviormonitoringrecordsrequiredtheteachertorecordthestudentstargetbehavioroccurrencesforkindergartenstudentsorindicatewhethersheagreedwiththestudentsself-monitoringforolderstudents.Theteacheralsorecordedwhetherornotthestudenthadearnedarewardforeacheligibleperiodoftheday.Planimplementationwasassessedbypermanentproductsasaresearchassistantcollectedthestudentsbehaviormonitoringrecordsattheendoftheday.Thepercentageofstepsimplementedeachdaywascalculatedanddividedbythetotalstepspossibleforthatdaytogetatreatmentintegrityscore.Amultiplebaselinedesignwasusedtoevaluatetheimpactoftheconsultationproceduresontheteachers.Inthefirstphase,datareview,theconsultantinitiated3-5minutereviewmeetingseachmorningbeforeschoolwiththeteachertoreviewthe

PAGE 27

20behaviormonitoringrecordfromthepreviousday.Thefrequencyofthemeetingswasgraduallyfadedtoeveryotherday.Next,aperformancefeedbackphasewasinitiatedforteacherswhoseimplementationdidnotstabilizeenoughtofadeduringdatareview.Performancefeedbackmodifiedthedatareviewprocedurebyfocusingthemeetingontwographspreparedbytheconsultant.Thefirstgraphdepictedthenumberofstarsthestudentearnedonthebehaviormonitoringrecordwhilethesecondgraphdepictedthepercentageoftreatmentstepsimplementedbytheteachereachday.Performancefeedback,inwhichtheconsultantprovidedplanimplementationdata,wasassociatedwithhighandstableinterventionimplementation.Inthisstudy,teachersdidnotmaintainaccurateimplementationintheabsenceofprogrammedfollow-up.Datareviewwasonlyclearlyeffectiveforoneofthefourteachers,thuswasnotassociatedwithsignificantimprovementsinimplementation.Intheperformancefeedbackphase,theconsultantsnotedthattheteachersfocusedonthestudentperformanceandgavelessconsiderationtohowtherecordsreflectedonimplementationoftheplan.Theuseofpermanentproductstoassessinterventionimplementationwasregardedasalimitationbytheauthorsbecauseitispossiblefortheteachertocompletethebehaviormonitoringrecordwithoutimplementingtheintervention.Socialinfluence.TheNoellteaminvestigationoftreatmentintegritywasextendedtocomparethreefollowupstrategies(Noell,etal.,2005).Thepurposeofthestudywastoexaminehowfollow-upandsocialinfluencemoderateimplementationofinterventionplans,andhowthat,inturn,moderatesstudentoutcome.Theparticipantsin

PAGE 28

21thestudywere45teachersofgeneraleducationstudentsinkindergartenthroughfifthgradesatsixurbanelementaryschoolsintheSoutheast.TheschoolspopulationsweremajorityAfrican-American(96%)andlowsocioeconomicstatus(90%ofstudentsreceivedfreeorreducedlunch).Consultationserviceswereprovidedbysevendoctoralstudents.Participationwasinitiatedwhentheteachersreferredstudentstoaschoolbasedteamthatprovidedconsultativepsychologicalservicesandinterventionplanningforstudentswithacademicorbehavioraldifficulties.Teacherswererandomlyassignedtooneofthethreetreatmentconditions:weeklyfollow-up,socialinfluence,andperformancefeedback.Weeklyfollow-upconsistedofabriefmeetingthatwasstructuredasaPlanEvaluationInterview(PEI).Inthiscondition,theconsultantaskedabouttheextenttowhichtheplanwasimplementedandonlyreviewedmaterialswhentheteacheraskedtheconsultanttoexaminethem.Notreatmentintegrityinformationwassharedwiththeteacher.Inthecommitmentemphasiscondition,allelementsofweeklyfollow-upwereincludedinadditiontoasocialinfluenceprocedure.Inthefinalmeetingpriortointerventionimplementation,thesocialinfluenceprocedurewasused,whichincludedreviewingfivespecificpointsdesignedtoenhancetheconnectionbetweentheircommitmenttoimplementtheinterventionandactualimplementation.Theperformancefeedbackprocedureconsistedofmeetingbrieflywiththeteachertoreviewinterventionplanpermanentproducts,graphingstudentbehavior,andgraphinginterventionimplementation.Theperformancefeedbackinthisstudyusedarapidthinningprocedurewhichallowedforanaverageof5.2contactsduringthethreeweekinterventionperiod.

PAGE 29

22Treatmentintegritywasassessedbypermanentproductsoftheinterventions.Consultantsdevisedapermanentproductchecklisttopermitscoringintegrity.Alltheinterventionmaterialswerecollectedattheendofthethree-weekinterventiontrialandscoredbytheconsultantusingthechecklist.Adirectmeasureofstudentsoutcomeswerecollectedpriortoandattheconclusionoftheintervention.AcademicoutcomesweremeasuredbystandardizedconditionssuchasCurriculum-BasedMeasurement.Behavioralconcernswereassessedbya30minutedirectobservationintheclassroom.AStudentBehaviorChangeIndex(SCBI)wasdevelopedtoallowbothtypesofassessmentdatatobesummarizedonacommonmetric.AnAnalysisofVariance(ANOVA)wasconductedtoexaminetheeffectsoftime,treatmentconditions,andtheinteractionofthetwovariables.Resultsindicatedasignificantmaineffectforcondition[F(2,42)=9.0,p=.001]whereinperformancefeedbackwasassociatedwithasignificantlyhigherleveloftreatmentintegritythantheothertwoconditions.Italsowasnotedthattreatmentintegrityalsowassomewhathigherinthefirstweekofplanimplementationacrossallconditions.Withregardtostudentoutcomes,theperformancefeedbackgroupexhibitedsubstantiallygreaterstudentbehaviorchangethaneitheroftheothergroupsasmeasuredbytheSBCI.Thecorrelationbetweentreatmentintegrityandchildoutcomeswasinthemoderaterange.Theauthorsconcludedfromthisstudythatperformancefeedbackwassuperiortotheothertwoconsultationproceduresexamined,thus,reviewofimplementationdataandrelevantfeedbackappearstobeacriticalfactorinmaintaininginterventionimplementation.Resultsofthisstudywereconsistentwithpreviousresearchinthat

PAGE 30

23implementationwasfrequentlypooranddeterioratedovertimeintheabsenceofperformancefeedback.Negativereinforcement.Aspreviouslymentioned,Noellandcolleaguessupporttheuseofperformancefeedbackforimprovingtreatmentintegrity.DiGennaro,Martens,andMcIntyre(2005)studiedtheeffectofcombiningperformancefeedbackwithadirectedrehearsalprocedure(negativereinforcement)ontreatmentintegrityforfourelementaryschoolteacherswhohadsoughtassistancetoaddressproblembehaviorsforasinglestudent.Specifically,teacherswereabletoavoidmeetingwiththeconsultantfordirectedrehearsal(i.e.practiceofthemissedstepsoftheintervention)byimplementingtheprocedurewith100%integrity.Performancefeedbackwasthinnedfromoncedailytoeverytwoweeks.Treatmentintegrityandstudentoff-taskbehaviorwerebothassessedthroughdirectobservationoftheconsulteebyatrainedobserver.Afive-phasemultiple-baselinedesignwasusedacrossteacher-studentdyads.Inthetrainingphase,consulteeswererequiredtohavetwosessionswith100%accuracyofimplementation.Inthenextphase,ImplementationBaseline,theconsultantremovedassistanceandtheobservercontinueddirectobservation.Allfourconsulteesdroppedtobetween20%and30%treatmentintegritywiththeremovalofconsultantsupport.Next,performancefeedback/negativereinforcementwasinitiatedinwhichconsulteesreceiveddailywrittenfeedbackandgraphsoftheirprogress.Implementationaccuracyintheperformancefeedback/negativereinforcementphaserangedfromaccuracyof70%to97%.Treatmentintegrityincreasedforallfourteachers,andthegainsweremaintainedovertimewithaprogressiveschedulethinningprocedure.Threeoffourstudentsdecreasedinoff-taskbehavior.

PAGE 31

24Severallimitationstothisstudythatmaylimitgeneralizabilityoftheseresultswerementioned.First,interventionswereonlyimplementedfortenminutesperday.Also,reactivitytobeingmonitoredmayhavecontributedincreasedratesoftreatmentintegrity.Studentoutcomesforthisstudyshouldbeinterpretedwithcaution,astwoofthefourstudentsbeganpsychotropicmedicationduringthestudy.Also,theinterventionwasnotsufficientlymatchedtothefunctionoftheoff-taskbehaviorsothatitcouldbecontrolledinthestudy.Despitethelimitations,resultsofthisstudydohaveseveralimplicationsforschoolbasedconsultation.Theresultsofthisstudyindicatethatlikestudentbehavior,consulteebehavior(orplanimplementation)issubjecttocontingenciesofreinforcement.Performancefeedbackcanbeeffectivetoenhancetreatmentintegrityandenhancesresearchdemonstratingthatdailymeetingsmaynotbenecessary.Asanextensiontopreviousresearch,highlevelsoftreatmentintegritymaybemaintainedthroughprogressivethinningoftheperformancefeedback/negativereinforcementschedule.Summarizationofconsultantfactorsinfluencingtreatmentintegrity.Insummary,differentaspectsoftheconsultationprocesshavebeenmanipulatedtoinvestigatetheiraffectsontreatmentintegrity.Thestudiespresentedhereindicatethatensuringhighlevelsoftreatmentintegrityshouldbeginwithdirecttrainingoftheteacherbeforebeginningimplementation.Onceinterventionimplementationbegins,theteacherwillneedsupportfromtheconsultanttoreachandmaintainacceptablelevelsoftreatmentintegrity.Performancefeedbackandnegativereinforcementhavebeenshowntoproducehighandstablelevelsoftreatmentintegrity.Progressivethinningandfadingscheduleshavebeenusedtosustaininterventionimplementation.Asnotedabove,severaltraining

PAGE 32

25techniqueshavebeenassociatedwithdesirablechangesinteacherbehavior(Han&Weiss,2005;Rose&Church,1998).Theconsultantischargedwithtrainingtheteachertoacceptandimplementthetechnologywithsufficientfidelitytoensuresuccess(Scott&Nelson,1999).ConsulteeFactorsInfluencingTreatmentIntegrityAsmentionedabove,thereisanemergingbodyofliteraturerelatedtotechniquesthatconsultantscanusetoimprovetreatmentintegrity.Thoughtherearethingsthattheconsultantcanalterintheconsultationrelationship,theconsulteealsoenterstheconsultationrelationshipwithvariablesthatmayaffectinterventionimplementationandtreatmentintegrity.Justasstudentsarriveatthesituationwithcharacteristicsthatwillrequiredifferentdosagesofintervention,itstandstoreasonthatteachersalsomayhavesomecharacteristicsthatwillrequiredifferentiallevelsofconsultantsupporttochangehisorherbehaviorandtoreachanacceptableleveloftreatmentintegrity.In1990,GutkinandHickmanconductedanexploratorystudytoinvestigateconsulteeresistancetoschool-basedconsultation.Datawerecollectedfrom23schoolpsychologyconsultants(universitypracticumstudents).ACaseDescriptionQuestionnaireandanOutcomeQuestionnairewerecompletedbytheconsultantsforeachcasetheyhadhandledduringthe14-weekperiod.Afactoranalysiswasconductedthatloadedtheitemsontothreefactors:consulteecharacteristics,consultantcharacteristics,andorganizationalcharacteristics.Consulteecharacteristicswerehighlycorrelatedwithperceptionsofconsulteeresistanceandperceptionsofoverallconsultationsuccess(bothsignificantatthep<.05level)andaccountedforthelargestproportionofuniquevariance.

PAGE 33

26Thefindingsofthisstudywereconsistentwithpreviousresearchindicatingthatconsultantstypicallyattributeconsultationfailurestotheconsultee.Theconsultantsperceptionofresistanceandconsultationoutcomesuccesswerestronglynegativelycorrelated(r=-.80,p<.01).Theseresultsshouldbeinterpretedwithcautionastheconsultantswereschoolpsychologypracticumstudentsandnotactualpracticingschoolpsychologists.Theresearchersalsosuggestfurtherresearchbeconductedtodiscernifconsultantscandiscernothercausesforunsuccessfulconsultations.Still,thisresearchsupportstheneedforfurtherexaminationofconsulteevariablesinconsultation.Adultbehaviorchange.Althoughconsultationdiscussionheavilyfocusesonthestudentsbehavior,theinitialchallengefacingtheconsultantissupportingbehaviorchangeonthepartoftheteacher.Tochangestudentsbehavior,theadultsintheenvironmentmustusuallychangesomeaspectoftheirbehaviortoimplementtheinterventionasintended(Sterling-Turner,etal.,2001;Witt,1997).Thoughtheconsultantwillfacilitatethedevelopmentoftheinterventionplan,itwillultimatelybeimplementedbytheteacher.Thus,teachersarefrontlineworkersinthebehaviorchangeprocess(Morin&Battalio,2004,p.251).Implementationofaschool-basedinterventionplanmayhavefeaturesincommonwithotherperformancemanagementcontextsinwhichadultsmayengagesuchasdietandexerciseprograms.Theseprogramsmayinvolvebehaviorsthatarenew,effortfulandrequireresourcesthepersonlacks(Noell,etal.,2005).Giventhesechallenges,adultbehaviorchange,inadditiontostudentbehaviorchange,mustbeafocusoftheconsultant(Noell,Duhon,Gatti,&Connell,2002).Ensuringplanimplementationisnotanactivitythatcanbeoverlooked.Inreviewinghisworkintwentyyearsofbehavior

PAGE 34

27analysis,Foxx(1996)estimatedthatinsuccessfulcases,10%ofeffortwasdevelopingthebehaviorsupportplanwhile90%wasfiguringouthowtoensurethatpeopleimplementedtheplan.Teacherbeliefs.Hunzicker(2004)suggeststhatteachersmayresistchangeduetolackofmotivation,lowlevelsofknowledge,orpoormoraloregodevelopment.Shesuggeststhateffortstochangeteacherbehaviorfocusonchangingtheteachersbeliefs.Tochangebeliefs,thenewinformationmustbepresentedrepeatedlyovertime.Fortrainingtobesuccessful,theteachermustdevelopnewwaysofthinkingaboutthebehavior.HanandWeiss(2005)wroteanarticletodiscussteacher-levelfactorsthatinfluenceinterventionimplementationandsustainability.Theyalsonotedthatperceptionsandbeliefsaboutanewprogramandtheteachersacceptanceoftheinterventionareimportant.Thesepreimplementationfactorsareimportantbecausetheteachersperceptionsoftheinterventionwillaffecttheirmotivationandeffortwhenattemptingtoimplementtheintervention,whichwillinturnaffecttreatmentintegrityandsustainability.Teacherknowledgeanduseofbehavioralprinciplesandinterventions.Traininginappliedbehavioranalysisandfunctionalinterventionproceduresisnotcommoninthebackgroundsofmostteachers(Scott&Nelson,1999).Asaresult,theinterventionstypicallyusedbyteacherstodealwithchallengingbehaviorsareineffectivebecausetheytendtobeunsystematic,negative,orboth.WeigleandScotti(2000)surveyed55teachersofstudentswithdevelopmentaldisabilitiesandreportedthatteachersknowledgeofbehavioralprinciplesimpactedthe

PAGE 35

28teachersratingsofacceptabilityandeffectivenessofdifferentinterventions.Theseauthorsadvocatethatteachertrainingprogramscontinuetoteachbehavioralprinciplesandfunctionalinterventionprocedurestobothgeneralandspecialeducationteachers. Wilsonandcolleagues(1998)investigatedtheknowledgebase,interventionuse,andproblem-solvingprocessesthroughsemi-structuredinterviewswith20thirdandfourthgradeteachers.Teacherswerepresentedwithastandardizedvignettetoreadandaskedtoprovidedetaileddescriptionsofinterventionstheyhadusedorcouldusewiththechallengingbehaviors.Then,theteacherswereaskedtothinkaboutastudenttheyhadtaughtoverthepastthreeyearswhowasultimatelyidentifiedasmildlyhandicappedanddescribetheprocessleadingtotheeligibility.Datawerecodedforcontentandspecificity.Forbothscenarios,teacherswereabletogenerateaboutnineinterventions,thoughthelevelofspecificityofinterventiondescriptionwascategorizedaslowformostcases.Fortheactualreferralcase,80%ofteachersindicatedthattheproblememanatedfromthechildren,theirparents,oracombination.Mostteachersusedhaphazarddatacollectionmethodsincludingmentalnotes,occasionaljournalentries,andintuitionwhichfocusedonrecordingthenegativebehaviorswhichwouldconfirmtheneedforspecialeducation.Ninety-fivepercentofteachersdescribednochangeintheirbehaviorafterthechildwasfoundeligibleforspecialeducation.Manyteachersindicatedthattheyhadenteredthepre-referralprocesswiththeintentionofreferringthechildforspecialeducationandthatthedecisionwastypicallymadeinfourweeksorlessbytheteam.Teachersfrequentlymentioneduncertaintyregardingifandwhentocontacttheschoolpsychologistorotherspecialists.

PAGE 36

29Theauthorshypothesizedthatteacherslimitedknowledgeofbehavioralinterventionsmayhaveimpededbrainstormingandleadtolowexpectationsuponenteringthepre-referralprocess.Theysuggestthatschoolpsychologyconsultantsfunctioninaneducativecapacitytofamiliarizeteacherswithboththeinterventionandtheprocessbothpriortoandduringtheconsultation.Insummary,mostteachersarenottrainedinbehavioralprinciples(Scott&Nelson,1999).Knowledgeoftheseprinciplesisrelatedtointerventionacceptabilityandperceptionsofeffectiveness(Weigle&Scotti,2000).Currently,teachersareunabletodescribeinterventionsforstudentswithchallengingbehaviorsindetailandareunfamiliarwiththepre-referralprocess.Datacollectionproceduresusedbyteachersarenotsystematicandareusedmainlyforthepurposeofdocumentingnegativebehaviors(Wilson,etal.,1998).TeacherEfficacyTeacherefficacybeliefsrefertotheteachersjudgmentabouthisorhercapabilitiestobringaboutdesiredoutcomesofstudentengagementandlearning,evenamongthosestudentswhomaybedifficultorunmotivated(Bandura,1977;Tschannen-Moran&Hoy,2001).Teacherefficacyisanimportantconstructbecauseitdealswiththeteachersbeliefsystemregardingstudents.Theteachersbeliefsystem,inturn,influenceshisorherbehavior.Teacherbehaviorthenaffectsthestudentsbehaviorandoutcomes.AccordingtoHanandWeiss(2000),teacherswithastrongsenseofefficacyaremoreopentonewideasandaremorewillingtoexperimentwithnewmethodsduetoincreasedlevelsofeffortandpersistence.

PAGE 37

30Teacherefficacyandperceptionsofstudentoutcomes.MorinandBattalio(2004)believethattheinitialconstrualofamisbehaviorisapivotalpoint.Theteachersprofessionalattitudeabouthisorherresponsibilityandhisorherskillinmanagingsuchepisodesarerelatedtoteacherefficacyandareapartoftheprocessofmakingmeaningofthesituation.Ateacherwithhighteacherefficacymightlookbeyondthesurfacelevelbehaviorfortheunderlyingcauseofit(anexternalcircumstancesuchasatraumaticeventorfrustration)asanexplanationforthebehavior.Recognizingthestudentsbehaviorasapartofalargercontext,theteachermightbemorelikelytouseproactiveinterventionswhichleadtomoreenduringoutcomes.Ateacherwithlowteacherefficacymayexplainstudentmisbehaviorasintentionalanddeliberateactsbythestudent.Thisteachermaybemorelikelytousereactiveinterventionstotemporarilycombatthesituationandperpetuatethecycleofmisbehavior.Morin(2001)promoteshelpingteacherstoseehowtheymaybecomeagentsofthemisbehaviortheyaretryingtoreduce.TournakiandPodell(2005)examinedtheeffectsofstudentandteachercharacteristicsontheteachersperceptionsofstudentacademicsocialsuccess.Asampleof384generaleducationteacherswereaskedtopredictstudentsuccessforcasestudiesdesignedtoexperimentallymanipulategender,readingachievement,socialbehavior,andattentivenessinadditiontocompletingateacherefficacyscale.Resultsindicatedthatteachersadjusttheirpredictionsbasedontheirownsenseofteachingefficacy.Teacherswithhighefficacytendedtomakelessnegativepredictionsandtoadjusttheirpredictionsasthecharacteristicschanged.Teacherswithlowefficacyseemedtopayattentiontoasinglecharacteristicwhenmakingtheirpredictions.Forexample,despitethereason,a

PAGE 38

31teacherwithlowefficacyrespondedsimilarlyregardlessofthereasonthatthestudentwasnotongradelevel.Theauthorsconcludedthathighteacherefficacyseemstoinoculateteachersfrommakingnegativepredictionsaboutstudents.Thoughthisstudyinvolvedtwoself-reportmeasuresanddidnotreportonactualteacherbehavior,itsupportstheconclusionthatattitudesandpredictionsofstudentsuccessareaffectedbyteacherefficacy.Teacherefficacyandteacherbehaviors.Justasteacherefficacyhasbeenrelatedtoteacherattitudesasdiscussedabove,italsohasbeenrelatedtoteacherbehaviors.Efficacyhasbeenrelatedtotheeffortteachersputforthinreachingthegoalstheysetandtheirlevelofaspiration.Teacherswithastrongsenseofefficacytendtobemorewillingtotrynewthingstomeettheindividualneedsoftheirstudentsandlesslikelytorefertospecialprograms(Tschannen-Moran&Hoy,2001).MeijerandFoster(2001)hypothesizedthattheteachersratingsofproblemseriousnesscouldbepredictedbyteachersdegreeofpersonalself-efficacy.Atotalof230teachersfromschoolsintheNetherlandsparticipatedinthestudy.Teacherscompletedasetofquestionsaddressingteachingcharacteristicsandselfefficacyandreadcasestudiesidentifyingstudentsintermsoftypeofproblem(learningorbehavior),gender,andsocialbackground.Theteachersthenwereaskedtowriteanumberfrom0to100toindicatewhetherthestudentwouldcauseaproblemandtheirlikelihoodofreferringthestudenttospecialeducation.Higherself-efficacyscoreswererelatedtolowproblemratingsandreferralchance.Theauthorspredictedinthediscussionthatifarelationshipbetweenselfefficacyandreferralvariableswereestablished,intervention

PAGE 39

32strategiescouldalsobedirectedtowardsboostingteacherselfefficacyaswellasinterventionstrategies.Inastudydesignedtoinvestigatetruestudent-teacherinteractions,Jordan,Lindsay,andStanovich(1997)soughttodeterminewhatcharacteristicsofteacherbeliefsandpracticesmightcontributetoeffectiveinstructionininclusivesettings.Participatingteacherswerenineelementaryschoolteachersofthirdgradestudents.Teacherswereinterviewedtodetermineiftheteachersperspectiveregardingexceptionalityaspathognomonic(i.e.,teacherassumesthatadisabilityisinherentintheindividualstudent)orinterventionist(i.e.,teacherholdstheperspectivethatmoststudentscanprofitfromthegeneraleducationclassroom).Thesescoreshavebeenshowntocorrelatewithteacherefficacywiththeinterventionistperspectivebeingassociatedwithhigherteacherefficacy.Eachoftheteachersnominatedsixstudentsfromhisorherclassroom,threewhowereformallyidentifiedasexceptionalorthoughttobeatriskofneedingassistanceandthreewhowerefunctioninginthemid-averagerangeoftheclassroom.Interactionsbetweentheteacherandthetargetedstudentswererecorded,transcribed,andcoded.Teacherswithaninterventionistperspective(associatedwithhigherteacherefficacy)interactedmorefrequentlywiththestudentswhowereexceptionaloratriskandathigherlevelsofcognitiveextension.Teacherswithapathognomonicperspective(associatedwithlowerteacherefficacy)hadlowratesofinteractionswithexceptionaloratriskstudents.Theauthorsspeculatedthatthismaybebecausetheseteacherslackconfidenceordonotseeteachingthesestudentsastheirresponsibility.Theauthorsconcludedfrom

PAGE 40

33thestudythatattitudes,beliefs,andefficacyappearedtoaffecttheteachersbehaviorinattemptstoengagetheexceptionaloratriskstudents.Teacherefficacyandexpectanciesforconsultation.Hughes,Grossman,andBarker(1990)conductedastudytodeterminetherelationshipbetweenteachersself-efficacyandtheirexpectationsforconsultation.Inthisstudy,72elementaryschoolteacherscompletedaquestionnaireofcommonproblemsexperiencedinelementaryschoolclassrooms.Ona5-pointLikert-typescale,theteachersfirstindicatedhowdifficultitwouldbetosolvetheproblemwithoutconsultation.Thisratingrepresentedtheteachersself-efficacy.Theythenwereaskedtoindicateonthesamescale,howdifficultitwouldbetoresolvetheproblemwiththeassistanceofapsychologicalconsultant.Thedifferenceofthetwoscoreswastheteachersoutcomeexpectancyscoretorepresentthedegreetowhichtheythoughtconsultationwouldbehelpful.Astatisticallysignificantrelationshipwasfoundbetweenthetwoscales(r=-0.37,p=.001).Teacherswithhigherself-efficacyhadlowerexpectationsthatconsultationwouldhelptheminresolvingclassroomproblems.Inadditiontothesurveysadministered,55oftheteachersinthestudywereinschoolsinwhichadoctoralstudentconsultantwaspresent.Dataonteachersevaluationofconsultationwereavailablefor27teachers.Ofthoseteacherswhohadusedconsultation,thosewithhigherself-efficacyscoreswerelesslikelytoreportchangingtheirperformanceasaresultofconsultation.Teacherswhohadpositiveexpectationsforconsultationviewedtheconsultationasmorehelpful.Also,teachersweremorelikelytoseekconsultationforbehavioralratherthanacademicproblems.

PAGE 41

34Inthediscussion,theauthorssuggestthatconsultantsemphasizethatconsultationisnotaremedialserviceforweakerteachers.Also,whenconsultingwithteacherswhoexpressahighlevelofconfidence,theconsultantshouldfocusconsiderableenergyonhelpingtheteacherfeelresponsibleforrecommendationsthatrequiretheteachertochangetheirbehaviortorespondtotheproblem.Theresultsofthisstudyareindirectcontrasttoassumptionsmadebyotherresearchers.Hughesandcolleagues(1990)reportthathighefficacyteacherswerelesslikelytoreportchangingtheirbehaviorasaresultofconsultation,whereas,othersindicatethatteacherswithhighefficacyaremorewillingtoexperimentwithnewmethods(Han&Weiss,2005).Theseconflictingresultsindicateaneedforfurtherstudyofthisphenomenon.Furthermore,theauthorsofthisstudysuggestfutureresearchtoassesschangesinefficacyandotherattributionsinteacherswhohaveparticipatedinconsultation.SustainabilityandGeneralizationofInterventionUsageInadditiontoinitialimplementation,sustainabilityoftheinterventionisimportanttothestudentscontinuedsuccess.Itisalsohopedthatteacherswouldgeneralizetheprocessandskillslearnedthroughconsultationtootherstudents.Researchindicatesthatteacherscommonlydonotgeneralizeeithertheproblem-solvingprocessorclassroomskillslearnedwithotherstudentspresentingchallengingbehaviors(Riley-Tillman&Eckert,2001).Inameta-analysisbyRoseandChurch(1998),only8of20studieswerecategorizedascompletemaintenanceindicatingthatperformanceofthetargetskillswassustainedatorabovelevelsattainedduringtraining.Thestudieswhichmetthiscriterionhadseveralthingsincommonincludingtrainingofteachersintheir

PAGE 42

35ownclassroomwithapracticeandfeedbackcomponentandabehavioralanalysisapproachtotraining.Theperiodfrompost-testtofollow-uptendedtobeshortwith9of20studiescollectingfollow-updataonlyfourweeksafterpost-test.Riley-Tillman&Eckert(2001)conductedastudydesignedtoexaminetheextenttowhichteachersgeneralizeabehavioralinterventionwithageneralizationpromptandgeneralizationtraining.Amultiple-baselineacrossthreesubjectswasusedwiththeinterventionofspecificpraise.Resultsweremixedas2of3participantsdemonstratedatleastsomeincreaseingeneralizingpraisetostudentsotherthanthetargetedstudent.Theresultsdo,however,suggestsomepromiseforthesestrategiestopromotegeneralization.Inastudyusingperformancefeedback,Coddingandcolleagues(2005)assessedtheshort-termmaintenanceeffectsofperformancefeedbackonimplementationofantecedentandconsequenceproceduresforteachersinaprivateschoolforstudentswithacquiredbraininjury.Theresearchersadministeredperformancefeedbackeveryotherweekthroughasessionapproximately12minutesinlengththatoccurredonthesamedayastheobservation.Implementationwasobservedtobesustainedathighratesduringthefollow-upphaseforallteachersforperiodsof8to22weeks.Previousfollow-upresultsforperformancefeedbackhavenotproducedsuchgoodresults.Theauthorshypothesizethatthelatencybetweensessions(oneobservationeverytwoweeks)madetheresultsmoresustainable.Tosummarize,studiesonsustainabilityofbehavioralinterventionshaveproducedmixedresults.Ingeneral,follow-updatadoesnotindicatesustainabilityofskills.Resultsongeneralizationpromptsandgeneralizationtrainingaremixed.Performancefeedbackhasbeenindicatedtobeastrategyassociatedwithhighratesofsustainability

PAGE 43

36comparedwithotherstrategies,howeverdirectresearchonsustainabilitywiththisstrategyismixed.PurposeoftheStudyThereisagrowingbodyofliteratureexploringconsultantfactorsthatmayaffectinterventionimplementationandtreatmentintegrity.Thereis,however,lessliteratureexploringteachervariablesthatmayaffecttreatmentintegrity.Thepurposeofthisresearchwastoexaminetherelationshipbetweenparticipationinatertiarylevelbehaviorinterventionprogramandteacherefficacy,confidenceindealingwithchallengingbehavior,andimplementationofbehaviorinterventions.Questionssurroundingbothpreimplementationfactorsandsustainabilitywereexplored.

PAGE 44

37Chapter3MethodThepurposeofthisresearchwastoexaminetherelationshipbetweenparticipationinthePreventTeachReinforce(PTR)behaviorinterventionprogramandteacherefficacy,confidenceindealingwithchallengingbehavior,andimplementationofbehaviorinterventions.Twostudieswereconductedtoexamineaseriesofresearchquestions.Study1usedarchivaldatatoexaminetheinfluenceofteacherefficacy,priortoparticipationinPTR,ontheamountofcoachingneededforateachertoimplementaninterventionwithanacceptablelevelofintegrity.Toexploresustainability,Study2usedasurveyofPTRparticipantsandnon-participantstodetermineiftheydifferedonteacherefficacy,confidenceindealingwithchallengingbehaviors,anduseofrecommendedbehavioralstrategies.Procedures,participants,andinstrumentsforbothstudiesarediscussedinthischapter.ContextoftheStudyPreventTeachReinforce(PTR)overview.ThecontextofthisresearchwasalargerrandomizedcontrolstudydesignedtostudytheefficacyofthePreventTeachReinforce(PTR)modelforstudentswithchallengingbehavior(UniversityofSouthFlorida&UniversityofColoradoatDenver,2006).PTRisateam-basedmodeldesignedtodecreasechallengingbehaviorandincreaseappropriatebehaviorforstudents.ProceduresusedinthePTRprocessweredevelopedusingthetheoreticalframeworkof

PAGE 45

38PositiveBehaviorSupport(PBS)andprinciplesofAppliedBehaviorAnalysis(ABA).Aconsultantguidestheteamthroughastandardizedfive-stepprocessusingatreatmentmanualthatdescribesthestepsoftheprocessindetail.Thefivestepsareteambuilding,goalsetting,functionalassessment,interventiondevelopmentandimplementation,andevaluation.Intheinitialstep,teambuilding,theconsultanthelpstheteamidentifyappropriatemembershipandpracticesforeffectiveteaming.Inthegoalsettingstep,theteamidentifieschallengingbehaviorstobetargetedandappropriatebehaviorstoteachthestudent.Alsointhisstage,theconsultanthelpstheteamtodeveloptheBehaviorRatingScale(BRS)toratetheirperceptionofthestudentsbehaviordaily,thusbeginningbaselinedatacollection.Stepthreeisafunctionalassessmentofthestudentschallengingbehavior.Throughinformationcontributedbytheteammembersanddirectobservationoftheenvironment,theconsultantassiststheteamindeterminingthefunctionofthebehavior.Next(step4)theconsultanthelpstheteamtodevelopabehaviorinterventionplanbasedonthefunctionofthebehavior.AllbehaviorinterventionplansdevelopedthroughthePTRprocessincludethreecomponents:prevention,teaching,andreinforcement.Oncetheplanisdeveloped,theconsultantcoachestheteammembersoncorrectimplementationintheplan.Thefinalstep(step5)isevaluation,inwhichtheteammembersevaluatetheeffectofthePTRprocessonthestudentsbehaviorandplanforcontinueddatacollectionandinterventionimplementation.Consultantprotocolchecklistsandteacherself-evaluationformswereusedateachstepoftheprocesstomonitoradherencetothetreatmentprocess.Datacollection

PAGE 46

39activitiesaredelineatedintothreemainphases:baseline,interventiontrainingandcoaching,andposttest.AnoverviewoftheactivitiesanddatacollectionspecifictothecurrentstudyisprovidedinTable1.Baselinephaseprocedures.Duringthebaselinephase,teachersand/orteamsparticipatedinthefirstfourstepsofPTRwhichincludedteambuilding,goalsetting,functionalassessment,andinterventiondevelopment.BaselineassessmentsincludingtheTeachersSenseofEfficacywerecompleted.FurtherdescriptionoftheTeachersSenseofEfficacyScaleisdiscussedbelowintheinstrumentssectionofthischapter.Interventiontrainingandcoachingprocedures.Afterthebehaviorinterventionplanwasdeveloped,acoachingtrainingsessionwasheld.Inthissession,theteacherand/orteamwastrainedbytheconsultantonhowtoimplementtheinterventionwithnostudentspresent.Trainingmethodsincludeddiscussion,verbalquestionandanswer,modeling,androleplaywithfeedback.Theinitialcoachingsessionwasheldthefirsttimetheinterventionwasimplementedintheclassroom.Inthissession,theteacherimplementedtheinterventionwiththestudentwhiletheconsultantwasavailabletoprovidecoaching,modeling,orotherassistancetotheteacher,dependentuponteacherpreference.Beginningwiththeinitialcoachingsession,uptotwelvehoursofcoachingortechnicalassistance(e.g.,revisionstointerventionplan)wasprovidedbytheconsultant.Integritychecks(seeinstrumentssectionbelowforfurtherdescription)wereconductedbytheconsultantthroughoutthecoachingphase.Coachingcontinueduntiltheteacherreachedcoachingintegrity(twotreatmentintegrityscoresof80%orgreater)orthetwelvehoursoftechnicalassistancewereexhausted.Thiscriterionisconsistentwith

PAGE 47

40Table1OverviewofPTRActivitiesandDataCollectedbyPhasePhaseActivitiesDataCollected BaselinePhasePTRStages1-4:Team-Building,goalsetting,functionalassessment,interventiondevelopmentTeachersSenseofEfficacyScale InterventionTrainingandCoachingPhase1.CoachingTrainingSession:NoStudentsPresent2.InitialCoachingSession:StudentsPresent3.Coachingand/orTechnicalAssistance:ContinueduntilteachercoachingintegrityorcoachinghourswereexhaustedIntegrityChecksPostTestPTRMeeting5:EvaluationFinalIntegrityCheck literaturewhichindicatesthattreatmentintegrityneedstobeachievedatanacceptabletosuperiorlevelduringtrainingtoensuretheinterventionisimplementedasintended(Han&Weiss,2005;Rose&Church,1998).Atthispoint,theconsultantwithdrewassistancetotheteacher.

PAGE 48

41Post-testphaseprocedures.Toensurethattheinterventionwasstillbeingimplementedcorrectlyafterthewithdrawaloftheconsultant,afinalintegritycheckwasconductedeightschooldaysaftercoachingintegritywasattained.Ifthefinalintegrityscorewas80%orhigher,thepost-testassessments(seeTable1)wereconductedtwodayslater(tendaysaftercoachingintegrity).If80%wasnotattained,theconsultantprovideduptothreeadditionalhoursofcoachingandrescheduledasecondfinalintegritycheck.Study1ThepurposeofStudy1wastoexaminethenatureoftherelationshipbetweenteacherefficacybeforeparticipationinPTRandtheamountofcoachingnecessaryforateachertoimplementabehaviorinterventionwithintegrity.Acorrelationaldesignandarchivaldatawereusedtodetermineifoverallteacherefficacyoranyfacetsofteacherefficacy(studentengagement,instructionalstrategies,andclassroommanagement)couldbeusedtopredicttheamountofcoachingnecessaryforateachertoreachanacceptableleveloftreatmentintegrity.DataCollectionRecruitment.PriortoinitiationofthePTRefficacyrandomizedcontrolstudy,permissionwasobtainedfromtheuniversityInstitutionalReviewBoard.Entrytothesiteswasobtainedbycontactingtheschooldistricts(Hillsborough,Pasco,andPinellasCountiesinFlorida)andgettingpermissiontoconductresearch.DistrictlevelpersonnelchosepotentialschoolswithstudentsgradesK-8forparticipationornotifiedschoolssothattheycouldchoosetoparticipate.Attheindividualschoollevel,abriefoverviewmeetingwasheldwiththeadministrators.Iftheadministratorsagreedtoparticipate,a

PAGE 49

42brieffifteen-minuteoverviewwaspresentedtothefacultytoexplainthestudyandsolicitparticipants.TeachersofstudentsingradesK-8inbothgeneraleducationandspecialeducationwereeligibletoparticipate.TeacherswishingtoparticipatecouldnominateuptothreestudentswithchallengingbehaviorsbycompletinggatesoneandtwooftheSystematicScreeningforBehaviorDisorders(SSBD)(Walker&Severson,1992)andastudentnominationformwhichsoliciteddemographicdataincludingage,cognitionlevel,languagedevelopment,placement(specialeducationorgeneraleducation),anddisabilitycategory,ifany.StudentswithatleastfivecriticaleventsindicatedontheSSBDgatetwowereconsideredeligible.Informedconsentdocumentswerereviewedwithteachers,parents,andstudentsbyamemberofthestudystaffbeforecommencinganystudyrelatedactivities.Thesedocumentsexplainedtheproceduresforthestudy,potentialrisksandbenefits,anddescribedthevoluntarynatureofthestudy.Aftersignedconsentwasobtainedfromallparties,studentswereassignedtothetreatmentorwaitlistcontrolconditionusingablockrandomizationprocedureconsideringage,cognitionlevel,language,andplacement.TeachersofstudentsassignedtothetreatmentconditionparticipatedinthePTRprocessduringthatschoolyear.Teachersofstudentsassignedtothewaitlistcontrolconditioncompletedassessmentsofstudentbehavior,butdidnotparticipateinthePTRprocess.Studentsinthecontrolconditionwereeligibletoreceivetheinterventioninthefollowingschoolyearifthechallengingbehaviorswerestillpresentandthestudentsteacheragreedtoparticipate.Teachersofstudentsrandomlyselectedtothetreatmentgroupwere

PAGE 50

43paida$500stipendandteachersofcontrolstudentswerepaid$100attheconclusionoftheprocess.Datacollectionprocedures.Instrumentscompletedbyteachers(TheTeachersSenseofEfficacyScales)weregiventoandcollectedfromteachersbydatacollectorsfromtheprojectstaff.Instrumentscompletedbyconsultants(integritychecklistsandtheconsultantcostanalysissheets)weregiventothedatacollectorsbytheconsultantforentryintothedatabaseandfiling.Alldocumentsweresecuredattheuniversitytorestrictaccesstoonlythestudystaff.Datawerecodedwithanidentificationnumbertode-identifyeachteacher,student,andschoolandenteredintoacomputerdatabase.Twentypercentoftherecordsforthecurrentstudywereexaminedbyvisualinspectiontocheckfordataentryerrorsandnoerrorswerefound.ParticipantsTheunitofanalysisforthisstudywasdataonteachersassignedtothetreatmentconditionforthePTRrandomizedcontrolstudyweretheparticipants.Atotalof71recordswereinitiallyexaminedforpossibleinclusion.Ofthese,17recordswereexcludedbecausetheteachersdidnotreachanintegrityscoreof.80duringthecoachingphase.Anadditional18recordswereexcludedbecausetherewasnotaccompanyingteacherefficacydatapairedwiththecoachingtimedata.(Note:Teacherefficacywasnotintroduceduntilthesecondyearofthethreeyearstudy).Thus,36recordswereincludedintheanalysis.Table2displaysthecredentialsandexperienceoftheparticipantsinStudy1.Themajorityheldaregularteachingcertificate(86%)andaBachelorsdegree(75%).

PAGE 51

44Themeannumberofyearsofexperiencewas7.96andmeanyearsatthecurrentschoolwas4.69.Additionally,34ofthe36participantswerefemale(94%).Table2Study1ParticipantCredentialsandExperiencePercentageFrequency TeachingLicense Regular86%31Probationary14%5HighestDegreeMasters25%9Bachelors75%27YearsTeaching0-5Years56%206-10Years19%711-15Years11%416-20Years8%321orMoreYears6%2YearsatthisSchool0-5Years75%276-10Years14%511-15Years6%216-20Years6%2 N=36

PAGE 52

45TeachingassignmentdataofparticipantsinStudy1ispresentedinTable3.Twentyoftheparticipants(56%)indicatedthattheywereaspecialeducationteacherandseventeen(47%)indicatedthattheywereageneraleducationteacher.Themajorityofteacherstaughtstudentsintheprimarygrades.Participantswerepermittedtoendorsemultipleitems,thuspercentagesdonottotal100%.Table3Study1ParticipantTeachingAssignmentPercentageFrequency GradeLevel Kindergarten17%6Grade125%9Grade239%14Grade328%10Grade422%8Grade519%7Grade68%3Grade711%4Grade88%3General/SpecialEducationGeneralEducation47%17SpecialEducation56%20 *Note:N=36;Percentagestonottotal100%

PAGE 53

46InstrumentsTeachersSenseofEfficacyScale.TheTeachersSenseofEfficacyScalewasadministeredtoteachersduringbaseline.ItwasdevelopedbyTschannen-MoranandHoy(2001)atOhioStateUniversity.Thescalecontains24(longform)or12(shortform)Likert-typequestions(seeAppendixA).Therespondentisaskedtoindicatehowmuchtheycandoinlightofavarietyofobstaclestoteachingonanine-pointscalewith1=Nothing,5=SomeInfluence,and9=AGreatDeal.Theshortformwasusedinthisstudyasthescalesauthorsindicateitisacceptableforuseexceptwhenbeingusedwithpre-serviceteachers.Scoresarecalculatedbydeterminingthemeanoftheoverallscaleandthesubscales.Thisscalehasbeenusedinpreviousresearch.Tschannen-MoranandHoy(2001)detailedthreepilotstudiesusedinthedevelopmentofthescaledesignedtoestablishconstructvalidityandtoaddressreliability,validity,andfactoranalysis.Tschannen-MoranandHoysdefinitionofteacherefficacywasadoptedastheoperationaldefinitionofteacherefficacyforuseinthecurrentresearch:ateachersefficacybeliefisajudgmentofhisorhercapabilitiestobringaboutdesiredoutcomesofstudentengagementandlearning,evenamongthosestudentswhomaybedifficultorunmotivated(p.783).Positivecorrelationswithothermeasuresofteacherefficacyprovideevidenceofconstructvalidity.Reliabilityfortheoverallscaleis.90.Inpilottestingbythescalesauthorsthemeanscoreforoverallefficacyrangewas7.1.Thepilotstudiesyieldedthreefactors:instructionalstrategies(abilitytodeliverinstructionandassesslearning),classroommanagement(abilitytostructureclassroomexpectationsandroutinesandtorespondtochallengingbehavior),andstudent

PAGE 54

47engagement(abilitytomotivatestudents).Reliabilitiesforthesubscalesrangefrom.81to.86.Meanscoresforthesubscalesrangedfrom6.7to7.3.Integritychecklists.Treatmentintegritywasmeasuredbytheconsultantthroughdirectobservationofinterventionimplementationbytheteacher.Becauseeachinterventionwasindividualized,theconsultantdevelopedindividualintegritymeasuresforeachbehaviorinterventionplan.First,eachstepoftheinterventionwastaskanalyzed.Next,eachstepwasdefinedintermsofadherence(wasthestepimplemented)andquality(wasthestepimplementedcorrectlyasplanned).Twoseparatescoresfortreatmentintegritywerecalculated.Theadherencescoreindicatedthepercentageofstepsimplementedtoasufficientdegreetodemonstrateaneffect(e.g.,praisesthestudentafterhandraise).Thequalityscoreindicatedthepercentageofstepsimplementedtothedegreenecessarytoachieveanoptimaleffect(e.g.,givesspecificpraiseforhand-raisingwithinfivesecondsofthebehavior).Theconsultantobservedtheimplementationoftheplanandrateditforbothadherenceandqualityusingthepossibleresponseofyes,no,ornotapplicable.Beforebeginningthestudy,theprojectstaffdesignatedascoreof80%onthequalitydimensionasthecriterionforacceptabletreatmentintegrity.ConsultantCostAnalysisSheet.Theconsultantrecordedthetime(tothenearestfiveminuteincrement)spentineachactivitybyallpartiesinvolvedonthecostanalysissheet.Consultantcoachingtimeincludedtimespentmodelingtheinterventionproceduresand/orpromptingimplementationbyschoolstaff.Italsoincludedtimespentobservingstaffimplementingtheinterventionandprovidingfeedbackonimplementation.

PAGE 55

48Interobserveragreement(IOA)datawerecollectedduringasampleofsixsessions.Twomembersofthestudystaffattendedacoachingsessiontogetherandeachpersonrecordedthenumberofminutesspentbyeachpartyinthesession.IOAwascalculatedbydividingtheshorterdurationbythelongerdurationandmultiplyingby100todetermineapercentageofagreement(Cooper,Heron,&Heward,1987).Themeanforagreementonconsultantcoachingtimewas94%witharangeof77%to100%.Study2ToexplorethesustainabilityoftheeffectsofparticipationinPTR,afollow-upsurveywasusedtodetermineifteacherswhoparticipatedinPTRdifferedfromteacherswhohadnotparticipatedonlevelofteacherefficacy,confidenceinabilitytodealwithchallengingbehavior,andinthefrequencyofuseofrecommendedstrategies.Thefollowingresearchquestionswereinvestigated:a)IsthereadifferenceinteacherefficacybetweenteacherswhohaveparticipatedinthePTRinterventionprocess,teacherswhohadsomeparticipationwiththePTRprojectbutdidnotimplementinterventions,andteacherswhodidnotparticipateinPTR?b)IsthereadifferenceinconfidenceindealingwithchallengingbehaviorsbetweenteacherswhohaveparticipatedinthePTRinterventionprocess,teacherswhohadsomeparticipationwiththePTRprojectbutdidnotimplementinterventions,andteacherswhodidnotparticipateinPTR?c)IsthereadifferenceinthefrequencyofABAandPBSrecommendedstrategiesusedforchallengingbehaviorsbetweenteacherswhohaveparticipatedinthePTRinterventionprocess,teacherswhohadsome

PAGE 56

49participationwiththePTRprojectbutdidnotimplementinterventions,andteacherswhodidnotparticipateinPTR?DataCollectionPriortoinitiationofthestudy,approvalwasobtainedfromtheuniversityInstitutionalReviewBoard(IRB)andtheparticipatingschooldistrictstosolicitparticipationfromtheschools.Participationinthestudywasvoluntaryandexplainedverballyatafacultymeetingoverviewfornewparticipantsandthroughaninformedconsentstatementatthebeginningofthesurvey.Datawerecollectedthroughanonlinesurveyandparticipantswereadvisedthatclickingonthesurveylinkandansweringthequestionsimpliedconsenttoparticipate.Recruitment.ParticipantsinthisstudywerepreviousPTRparticipantsandotherteachersworkinginschoolswhichparticipatedinPTR.Anaprioripoweranalysiswasconductedtodetermineanappropriatenumberofparticipants(Stevens,1999).Withtheexpectationofasmalleffectsize(atleasttwoofthegroupsdifferingfromeachotherbyatleast.50standarddeviationsonatleastonedependentvariable)asamplesizeof115wouldbeneededtoobtainapowerof.80.TorecruitpreviousPTRparticipantsforthecurrentstudy,alistof159e-mailaddressesofpreviousPTRparticipantswascompliedbycollectingtheinformationfromstudystaffandreviewofrecords.Toselectschoolsfromwhichtosolicitnon-volunteercomparisonteacherstoparticipate,aproportionalstratifiedsamplingprocedurewasused.First,theratioofschooltypes(elementaryschool,middleschool,exceptionalstudentcenter)wasdetermined.Thenschoolswereselectedtomatchthisratiotosolicitnon-PTRteacherparticipantsforthecurrentstudy.Onlyschoolsparticipatingduringboth

PAGE 57

50yearswereconsidered.Thisstipulationwasinplacetomakesurethatteachershadbeenexposedtoafacultyoverviewtorecruitparticipants(2006-2007)andthattheschoolwasstillactivelyparticipatingintheproject(2007-2008).Fourteenelementaryschools,onemiddleschool,andthreeexceptionalstudentcentershadinterventionteamsduringbothschoolyears.Participantsforthisstudywererecruitedfromfourelementaryschools,onemiddleschool,andoneexceptionalstudentcenter.Oncetheschoolswereidentified,theresearchercontactedtheadministrationoftheselectedschoolstoconfirmapprovaltosolicitteachersforparticipationandtoscheduleatimetoannouncethestudytothefaculty.Theresearchermadeabrief(fiveminute)announcementatafacultymeetingsolicitingparticipantsforthestudy.Intheannouncement,theresearcherinformedtheteachersthatparticipationwasvoluntarybutappreciated,andthatthosewhocompletedthesurveywouldbeenteredintoanincentivedrawing.Seventy-twoteachersprovidedtheirnameande-mailaddressonsign-upsheetsprovidedatthefacultyoverviewmeetingsindicatingtheywishedtoparticipate.Oneelementaryschoolwasanexceptiontothisrecruitmentstrategy.Theprincipaloftheschoolrequestedthatinsteadofanannouncementatafacultymeeting,shesendalinktotheteachersintheschoolthroughtheinternale-mailsystemtosolicitparticipants.Anadditionaltenparticipantswererecruitedthroughthiswebcollector.Datacollectionprocedures.DataforthisstudywerecollectedthroughtheSurveymonkey.comcomputerprogram(Finley,n.d)whichsente-mailstotheparticipants,trackedresponsestatus,andrecordedandstoredresponses.Computerizeddatacollectionwasselectedtoeliminatedataentryerrorsintroducedbypaper-based

PAGE 58

51surveysandtoincreasereturnratebyallowingteacherstocompletethesurveyattheirleisureinsteadofaprescribedtime.E-mailaddressesforallpotentialparticipantswereenteredintotheprogramwhichsentane-mailonbehalfoftheresearchertoallpotentialparticipantswithalinktotheonlinesurveymaterials.Thesee-mailmessages(seeAppendixB)informedthepotentialparticipantthatresponsewasvoluntarybutappreciatedandthatthosewhocompletedthesurveyswouldbeenteredintoadrawingforoneoffive$20giftcardsforalocaldepartmentstore(e.g.,Target).Iftheteacherwasrecruitedthroughafacultyoverviewmeeting,theinitiale-mailmessagewassentwithintwodaysofthemeeting.TheresearcherprogrammedSurveymonkey.comtosendreminderstothosewhohadnotresponded.Atotalofthreeremindersweresentatoneweekintervals.Ofthe216e-mailsthatweresenttovalide-mailaddresses,therewere123completeand14partialresponsesfora63%responserate.Requestsweresenttoparticipantswithpartiallycompletedsurveysrequestingthemtocompletethesurveywithsomesuccess.Anadditional10responseswerecollectedthroughthewebcollectorfromtheschoolinwhichtheprincipalsenttheinvitationtoherstaff.Attheconclusion,therewere132completeand16partiallycompleteresponses.Thelastsectionofthesurvey,demographics,wasthemostlikelyportiontobeincomplete.Surveyswhichcontainedallnecessarypiecesofinformationforanalysisandgrouping(missingonlydemographics)totaled136andareincludedinthisanalysis.ParticipantsGroupingofparticipants.ParticipantswereassignedtothreegroupsbasedontheirlevelofparticipationinPTRincludingwhethertheyhadbeenaparticipantor

PAGE 59

52nominatedastudentforPTR.ParticipantswerealsoaskedtoendorsePTRactivitiesinwhichtheyhadparticipatedtofurtherdiscernthelevelofparticipationfordistinctionbetweengroups.TeacherswhoparticipatedinthePTRprocessthroughandincludingimplementinginterventionsandparticipatingincoachingwereassignedtoGroup1(n=49).TeacherswhohadsomeexposuretoPTRwithoutinterventionimplementationwereassignedtoGroup2(n=37).Thisgroupincludedfifteenteacherswhohadvolunteeredtoparticipatebutwereassignedtothewaitlistcontrolgroup.Thoughthisgroupofteachersmayhavehadtheintentiontoparticipateintheintensivebehaviorinterventionprogram,theydidnotparticipateinthestepsofthePTRprocess,werenotexposedtothematerials,didnotworkwithaconsultant,norsawtheeffectsoftheinterventionplansdevelopedthroughthePTRprocess.Group2alsoincludesfiveteacherswhoparticipatedinsomeofthePTRprocessbutdidnotimplementinterventionsforsomereason(e.g.,studentmoved,teacherreassigned,droppedout),andseventeenteacherswhoparticipatedinfollow-updatacollectionforastudentwhohadbeenaparticipantinPTRinthepreviousyear.TeachersinGroup3(n=50)hadnotparticipatedinanyPTRactivities.Thepurposeofthisthirdgroupofteacherswastodetermineifresponsesweredifferentbetweenteacherswhovolunteeredtoparticipateinatimeintensivebehaviorinterventionprojectandnon-volunteers.Thoughtheseteachersvolunteeredtoparticipateina20-minutesurvey,theydidnotvolunteertoparticipateintheintensiveinterventionprocess.FurtherexplanationofgroupingproceduresandquestionsisprovidedinAppendixC.

PAGE 60

53Participantdemographics.Demographicinformationabouttheparticipantsisreportedinthissectionincludingdataoncredentials,experience,andteachingassignment.Thesamplewasratherhomogeneouswithregardstoteachingcredentials(seeTable4).Themajorityofparticipants(89%)heldaregularteachingcertificateandwerenotteachingoutoffield(93%).MostheldaBachelorsdegree(70%).Table4Study2ParticipantCredentialsGroup1(n=48)Group2(n=35)Group3(n=49)Total(n=132) TeachingLicense Regular33%(44)25%(33)30%(40)89%(117)Probationary2%(3)1%(1)5%(7)8%(11)Emergency/None1%(1)1%(1)2%(2)4%(5)HighestDegreeDoctorate1%(1)----1%(1)Specialists1%(1)1%(1)--2%(2)Masters10%(13)6%(8)11%(15)27%(36)Bachelors25%(33)20%(26)26%(34)70%(93)OutofFieldNotOutofField36%(47)24%(32)33%(44)93%(123)OutofField1%(1)2%(3)4%(5)7%(9)

PAGE 61

54ExperiencelevelsofparticipantsinStudy2aredetailedinTable5.Nearlyseventypercenthadtenyearsoflessorteachingexperiencewiththelargestgroupbeingteacherswith0-5yearsofexperience(48%).Approximatelythree-quarters(76%)ofparticipantshadbeenattheircurrentschoolforfiveyearsorless.Table5Study2ParticipantExperienceGroup1(n=49)Group2(n=37)Group3(n=50)Total(n=136) YearsTeaching 0-5Years15%(21)13%(17)20%(27)48%(65)6-10Years8%(11)5%(7)7%(9)20%(27)11-15Years6%(8)1%(2)3%(4)10%(14)16-20Years1%(2)2%(3)3%(4)7%(9)21-25Years3%(4)2%(3)4%(5)9%(12)25orMoreYears2%(3)4%(5)1%(1)7%(9)YearsThisSchool0-5Years27%(37)20%(27)29%(40)76%(104)6-10Years4%(6)4%(6)2%(3)11%(15)11-15Years2%(3)1%(2)2%(3)6%(8)16-20Years1%(1)1%(1)2%(3)4%(5)21orMoreYears1%(2)1%(1)1%(1)3%(4)

PAGE 62

55TeachingassignmentsofparticipantsinStudy2aredetailedinTable6.Mostparticipantstaughtatelementaryschools(67%).Themostcommonteachingassignmentwaselementaryschoolgeneraleducation(41%)followedbyelementaryschoolspecialeducation(21%).Nosignificantdifferenceswerefoundinanyofthedemographiccategoriesbetweenthegroupswithchi-squareanalysis.Table6Study2ParticipantTeachingAssignmentsGroup1(n=48)Group2(n=35)Group3(n=49)Total(n=132) SchoolType Elementary23%(31)20%(26)23%(31)67%(88)Middle8%(10)5%(6)12%(16)24%(32)ExceptionalCenter5%(7)2%(3)2%(2)9%(12)AssignmentPre-KorK(GE)2%(2)2%(2)2%(3)5%(7)Elementary(GE)14%(18)10%(13)17%(23)41%(54)MiddleSchool(GE)2%(2)1%(1)8%(11)11%(14)Pre-KorK(SE)2%(3)1%(1)1%(1)4%(5)Elementary(SE)9%(12)9%(12)3%(4)21%(28)MiddleSchool(SE)5%(7)3%(4)2%(3)11%(14)SpecialtyArea----2%(2)2%(2)Other3%(4)2%(2)2%(2)6%(8) Note:GeneralEducation(GE);SpecialEducation(SE)

PAGE 63

56InstrumentsParticipantswereaskedtocompleteTeachersSenseofEfficacyScale(seeinstrumentssectionStudy1andAppendixA)andselectedportionsoftheQuestionnaireaboutTeachersandChallengingBehaviors(Westling,2006).ParticipantsalsoansweredasetofquestionsregardingpreviousparticipationinthePTRprojectandsomedemographicquestions.ThecompletesurveyispresentedinAppendixD.Researchquestion2ausedthesameoperationaldefinitionofteacherefficacyasStudy1.QuestionnaireaboutTeachersandChallengingBehaviors.TheQuestionnaireaboutTeachersandChallengingBehaviorswasdesignedtogatherinformationonteachersviewsandapproacheswhendealingwithstudentswithchallengingbehaviors(Westling,2006).Teachersprovidedescriptiveinformationaboutthemselvesandtheirstudentsandreporttheirperceptionsonsevendimensionsrelatedtochallengingbehavior.Psychometricdataonthisinstrumentwerereportedinastudyconductedbythequestionnairesauthorinwhich70generalandspecialeducationteacherscompletedthequestionnaire.Contentvaliditywasestablishedthroughreviewbyapanelof15nationalexperts.Test-retestreliabilitieswereabove.70withtheexceptionoffouritems.InternalconsistencyreliabilitieswerereportedforselectedsectionswithCronbachsalphavaluesrangingfrom.76to.92.Dataforresearchquestion2bincludedresponsesfromthesectionofthequestionnairetitledYourProfessionalPreparationinDealingwithChallengingBehaviors(seeAppendixD).Inthissection,theparticipantratedtheirconfidencein

PAGE 64

57applyingskillsinsixareasoftrainingrelatedtoapplyingbehaviorinterventionsonaLikert-typescalewhere0=Unconfidentand3=HighlyConfident.Confidenceindealingwithchallengingbehaviorwasoperationalizedastheparticipantsperceivedabilitytouseskillsin(a)principlesofappliedbehavioranalysis,(b)functionalbehaviorassessment,(c)classroommanagement,(d)individualbehavioralinterventions,(e)datacollectionandassessment,and(f)school-widepositivebehaviorsupports.Scoresonthesixitemsweresummedandpossiblescoresrangedfrom0to18withhighervaluesindicatinghigherconfidencelevels.Inpreviousresearchbytheauthorofthequestionnaire,test-retestreliabilityofthissectionwas.95andinternalconsistencyreportedasCronbachsalphawas.83.Dataforresearchquestion2cincludedresponsesonthesectionofthequestionnairetitledCurrentStrategiesYouUseforDealingwithChallengingBehaviors.ThissectionincludeseighteenrecommendedABAandPBSstrategiesinwhichtheteacherindicatedthefrequencyofuseona5-pointLikert-typescalewhere1=Ineverusethisstrategyand5=Ialwaysusethisstrategy.FrequencyofABAandPBSrecommendedstrategiesusedforchallengingbehaviorswasoperationalizedbysummingthescoresonthe18itemswithscoresrangingfrom18to90withhigherscoresindicatingmoreuseofrecommendedstrategies.Inpreviousresearchbytheauthor,test-retestreliabilityonthissectionwas.99andinternalconsistencyreportedasCronbachsalphawas.81.Fewteachersreportedfrequentlyusinganyofthestrategiesoftenorveryoften.

PAGE 65

58SummaryThecontextofthecurrentresearchwasthePTRrandomizedcontrolstudy.Study1examinedarchivalrecordsoftheTeachersSenseofEfficacyScaledataandconsultantcoachingtimeforthirty-sixparticipants.Asurveyof136previousPTRparticipantsandnon-participantswasconductedinStudy2.TheparticipantscompletedanonlinesurveyincludingtheTeachersSenseofEfficacyScale,QuestionnaireaboutTeachersandChallenging,questionsaboutlevelofPTRparticipation,anddemographicinformation.

PAGE 66

59Chapter4ResultsThischapterwilldescribethedataanalysisandresultsforthecurrentresearch.RegressionanalyseswereusedinStudy1toexaminethenatureoftherelationshipbetweenteacherefficacyandtheamountofcoachingtimerequiredforateachertoimplementabehavioralinterventionwithintegrity.Study2usedaMANOVAanalysistodetermineifPTRparticipantsdifferedfromnon-participantsinteacherefficacy,confidenceindealingwithchallengingbehaviors,andreporteduseofrecommendedstrategies.Dataanalysis,datascreening,considerationsofassumptions,andresultsforbothstudiesarepresentedinthischapter.Study1ThepurposeofStudy1wastodeterminethenatureoftherelationshipbetweenteacherefficacyandtheamountofcoachingtimerequiredforateachertoreachanacceptableleveloftreatmentintegrityinimplementationofabehavioralintervention.Analyseswereusedtodetermineifteacherefficacyoranyfacetsofitcouldbeusedtopredicttheamountofcoachingneeded.Outcomevariablesweretheamountofconsultantcoachingtime(asdefinedinChapter3)andthenumberofcoachingsessions.DataAnalysisFirst,regressionanalyseswererunwithoverallteacherefficacyasthepredictorvariable.Tofurtherexploretherelationship,multipleregressionanalyseswererunwith

PAGE 67

60theteacherefficacysubscalesasthepredictorvariables.Thenullhypothesiswasthattherewasnorelationshipbetweenteacherefficacyandtheamountofcoachingnecessarytoreach80%treatmentintegrity.Thesignificancelevelwaspresettop<.05.Descriptivestatistics.DescriptivestatisticsforteacherefficacyandcoachingtimevariablesarepresentedinTable7.Teacherefficacyitemswereeachratedonascaleofonetonine.Themeanofoverallefficacywas6.69whilemeanscoresontheefficacysubscalesrangedfrom6.98to7.40.Allscaleswereslightlynegativelyskewed.Allmeasuresofcentraltendencywerewithinonepointofeachother.Therewerefewoutlierswithonlyoneeachontheoverallefficacy,instructionalefficacy,andmanagementefficacyscales.Kurtosisvalueswerenearzerowiththeexceptionofinstructionalefficacywhichwasslightlyleptokurtic,withthemeanpeakingat7.40.Table7Study1DescriptiveStatisticsforTeacherEfficacyandCoachingTimeVariablesMeanSDSkewnessKurtosisMinMax OverallEfficacy6.690.80-0.530.444.628.38 EngagementEfficacy6.981.06-0.16-0.604.759.00InstructionalEfficacy7.400.98-1.504.643.759.00ManagementEfficacy7.101.01-0.740.784.259.00NumberofSessions4.922.45-0.40-0.851.009.00ConsultantTime395.83174.190.08-0.0850.00815.00 NoteN=36Thenumberofsessionsrangedfromonetoninewithnooutliers.Theaveragenumberofsessionswasfive.Thedatawerenegativelyskewedwithamedianof5.5and

PAGE 68

61amodeof6.Consultantcoachingtimewaswideranging(50to815minutes)withafairlynormaldistribution.Beforetheanalyseswereconducted,correlationsamongthevariableswerecalculatedandresultsarepresentedinTable8.Thecorrelationofmanagementefficacyandengagementefficacywassignificantatthep<.001level.Correlationsbetweeninstructionalefficacyandeachoftheothertwoefficacysubscalesweresignificantatthep<.0001level.NeitheroutcomevariablecorrelatedsignificantlycorrelatedwithanypredictorvariablesTable8Means,StandardDeviations,andIntercorrelationsamongTeacherEfficacyandCoachingTimeVariablesVariable12345 1.EngagementEfficacy-2.InstructionalEfficacy0.62**--3.ManagementEfficacy0.52*0.60**--4.NumberofSessions-0.120.040.06--5.ConsultantTime-0.03-0.15-0.140.26--Mean6.987.407.104.92395.83StandardDeviation1.060.981.012.45174.19 NoteN=36,*p<.001,**p<.0001Datascreening.Assumptionsofregressionanalysiswerereviewedpriortorunningtheanalyses.Regressionisrelativelyrobusttotheassumptionthatthepredictor

PAGE 69

62variableisfixed.Thescaleusedtomeasurethepredictorvariableshasadequatereliabilitylevels(seeChapter3).Toavoidintroducingmulticollinearity,analyseswereseparatedbetweenoverallteacherefficacyandtheefficacysubscalesbecausethesubscalescoresarecontainedintheoverallefficacyscores.VarianceInflationFactorswereexaminedandnonewereapproachingorexceedingten.CooksDvaluesoftheresidualswereexaminedandnonewereapproachingorexceedingoneindicatingnoinfluentialoutliers.Graphsofthedatawereexaminedforlinearityandplottedresidualswereexaminedforhomoscedacity.Itappearedacceptabletoproceedwiththeanalysiswhileinterpretingwithcautionduetosmallsamplesizeandpossiblemeasurementerror.Results.Simpleregressionanalyseswereconductedwithoverallefficacyasthepredictorvariable.Whenthenumberofcoachingsessionswasconsideredastheoutcomevariableresultsofthesimpleregressionwerenotsignificant(F(1,34)=0.00,p=0.9525,adjustedR2=-0.0293).Whentheamountofcoachingtimewasconsideredastheoutcomevariableresultsofthesimpleregressionalsowerenotsignificant(F(1,34)=0.51,p=0.6632,adjustedR2=-0.0141).Multipleregressionanalyseswereconductedwiththethreesubscalesofteacherefficacyaspredictorvariables.Theresultsoftheregressionanalysiswithnumberofcoachingsessionsastheoutcomevariablewerenotsignificant(F(3,32)=0.53,p=0.6632,adjustedR2=-0.0417)andarepresentedinTable9.Theresultsoftheregressionanalysiswithconsultanttimeastheoutcomevariablewerenotsignificant(F(3,32)=0.41,p=0.7458,adjustedR2=-0.0531)andarepresentedinTable10.

PAGE 70

63Table9RegressionResultsforNumberofSessionsandEfficacySubscalesVariableRegressionCoefficientStandardErrort-valueStandardizedRegressionCoefficient()SquaredSemi-partialCorrelation Intercept4.648083.553311.31---StudentEngagement-0.63297.052384-1.21-0.274190.04346InstructionalStrategies0.345420.610380.570.137530.00953ClassroomManagement0.299700.539020.560.123730.00920 NoteN=36,R2=0.0476,F(3,32)=0.53,p=0.6632,adjustedR2=-0.0417

PAGE 71

64Table10RegressionResultsforConsultantCoachingTimeandEfficacySubscalesVariableRegressionCoefficientStandardErrort-valueStandardizedRegressionCoefficient()SquaredSemi-partialCorrelation Intercept596.10698253.839572.35---StudentEngagement22.4040537.387050.600.136720.01080InstructionalStrategies-29.6978543.56372-0.68-0.166580.01398ClassroomManagement-19.2547038.47019-0.50-0.111990.00754 NoteN=36,R2=0.0371,F(3,32)=0.41,=p=0.7458,adjustedR2=-0.0531AsummaryoftheresultsofthetworegressionanalysesarepresentedinTable11.Norelationshipsbetweenpredictorvariablesandoutcomevariableswerestatisticallysignificant,thusthenullhypothesiswasnotrejected.Nomodelaccountedformorethan5%ofthesamplesvariabilityintheoutcome.

PAGE 72

65Table11SummaryofResultsofRegressionAnalysesOutcomeVariablePredictorVariable(s)R2,AdjustedR2F-statistic,Probability OverallEfficacyR2=0.0001,Adj.R2=-0.0293F(1,34)=0.00,p=0.9525 NumberofSessionsEfficacySubscalesR2=0.0476,Adj.R2=-0.0417F(3,32)=0.53,p=0.6632OverallEfficacyR2=0.0148,Adj.R2=-0.0141F(1,34)=0.51,p=0.4791CoachingTimeEfficacySubscalesR2=0.0371,Adj.R2=0.0531F(3,32)=0.41,p=0.7458 NoteN=36Study2ToexplorethesustainabilityoftheeffectsofparticipationinPTR,anexpostfactosurveywasconductedtodetermineifteachersdifferedfromoneanotherbasedontheirlevelofparticipationwiththePTRbehaviorinterventionprogram.Group1consistedofteacherswhohadpreviouslyimplementedbehaviorinterventionsinthePTRprogram.Group2includedteacherswhohadsomefamiliaritywithPTR(teachersofstudentsassignedtothecontrolconditionorasateacherofastudentinthefollow-upyear)buthadnotimplementedinterventionsdevelopedthroughPTR.Group3teacherswererecruitedfromPTRschools,buthadnotparticipatedinanyPTRactivities.The

PAGE 73

66dependentvariablesexaminedwerelevelofteacherefficacy,confidenceinabilitytodealwithchallengingbehavior,andinthefrequencyofuseofrecommendedstrategies.DataAnalysisMultivariateanalysisofvariance(MANOVA)wasselectedfordataanalysistodetermineifthegroupsdifferedsignificantlyonthesetofdependentvariables.AnadvantageofMANOVAanalysisisthatitallowstheresearchertogainpowerwhichmaydetectdifferencesthatunivariateanalysesalonemaynotdetect.Thenullhypothesiswasthattherewerenosignificantdifferencesbetweenthegroups.Thesignificancelevelwaspresettop<.05.Descriptivestatistics.Thedependentvariablesexaminedincludethesubscalesofteacherefficacy(studentengagement,instructionalstrategies,andclassroommanagement),confidenceindealingwithchallengingbehavior,andreporteduseofrecommendedstrategies.Meansandstandarddeviationsofeachdependentvariablebygroupandofthewholesample(N=136)arepresentedinTable12.Teachersratedeachteacherefficacyitemonascaleofonetonine.Meanscorespergroupontheefficacyscalesrangedfrom6.90to7.69.Minimumscoresonthescaleswere4.25forstudentengagement,2.75forinstructionalstrategies,and4.50forclassroommanagement.Eachscalehadactualmaximumscoresof9.00.Distributionsofeachefficacysubscalewereroughlynormal,butnotedbyaslightlynegativeskew.Measuresofcentraltendencyforallscalesfellwithinonepointofeachother.Threeoutlierswerenotedbetweenthethreescales.

PAGE 74

67Table12Study2MeansandStandardDeviationsforDependentVariablesbyGroupGroup1 (n=49)Group2 (n=37)Group3 (n=50)Total (n=136) EngagementEfficacy7.19(1.04)7.11(1.08)6.90(1.39)7.06(1.19) InstructionalEfficacy7.67(0.88)7.69(0.79)7.49(1.33)7.61(1.04)ManagementEfficacy7.57(0.86)7.56(1.07)7.51(1.12)7.54(1.01)ConfidenceinBehavior5.33(2.46)4.49(3.17)5.94(2.68)5.32(2.79)RecommendedStrategyUse65.67(8.27)65.38(8.88)62.48(7.92)64.42(8.39) *Note:StandardDeviationsinparenthesesForthevariableconfidenceindealingwithchallengingbehaviors,participantsratedsixitemsonascalefromzerotothreeresultinginapossiblescorerangeof0to18.Actualscoresrangedfrom0to12withgroupmeansfrom4.49to5.94.Datawereroughlynormallydistributedforthisvariable.Thepossiblescorerangeforthevariableuseofrecommendedstrategieswas18to90.Actualscoresrangedfrom33to85withgroupmeansfrom62.48to65.67.ForGroup1,thedistributionwasleptokurtic(kurtosis=4.11).Theothergroupswereroughlynormallydistributed.

PAGE 75

68Datascreening.Beforeproceedingwiththeanalysis,thedatawerescreenedtomakesuretheywereconsistentoftheassumptionsunderlyingMANOVA.Responsesofeachparticipantwereindependentanddistributionsofeachvariablewereroughlynormalwiththeexceptionsnotedabove.ThemultivariateskewnessoftheresidualsfromtheMANOVAwascalculated[b1,p=3.37,2(35,N=136)=78.61,p<.01]andwasstatisticallysignificant.ThemultivariatekurtosisoftheresidualsoftheMANOVAwas[b2,p=38.04,zupper=2.12,zlower=1.04]withtheupperboundbeingslightlyoutsidethecriticalvalueofz+/-1.96.ScreeningformultivariateoutliersproducedamaximumMahalanobisdistanceof24.01[F(5,130)=5.68,p<.01]whichwasstatisticallysignificant.However,thedistanceoftheoutlierwasfoundtobeapossiblevalueandthusincludedintheanalysisduetothenatureoftheresearchquestions.ABoxsMtestrevealednostatisticallysignificantdifferencesinthecovariancematrices[2(30,N=136)=33.29,p=.31].BecauseMANOVAisrelativelyrobusttoviolationsofthenormalityassumptionwhensamplesizesarelarge,theinconsistenciesinthemultivariatenormalitywerenotedandtheanalysiswasconducted.MANOVAresults.TheWilksLambateststatisticwasfoundtonotbestatisticallysignificant[=.93,F(10,258)=.99,p<.45]whenmeasuringteacherefficacyusingthethreesubscalescores.Asmalleffectsizewascalculated(f2=.04).UsingCohensapproach,thestatisticalpoweroftheMANOVAwascalculatedtobe0.54indicatinga46%chanceoffailingtodetectapresenteffectifthepopulationeffectwasf2=.04.Forthesakeofcomparison,theMANOVAwasalsorunusingoverallteacherefficacyinsteadofthethreesubscalestorepresenttheconstructalongwiththedependentvariablesofconfidenceindealingwithchallengingbehaviorand

PAGE 76

69recommendedstrategyuseandtheresultswerestillnotsignificant[=.93,F(10,258)=.99,p<.45].ANOVAresults.Giventhenon-significantMANOVA,itwouldbeexpectedthattheANOVAresultsofindividualvariableswouldalsobenon-significant.However,theordertoanswertheoriginalresearchquestionsposed,ANOVAresultsarereportedinTable13.Thefollowingresearchquestionswereinvestigated:a)IsthereadifferenceinteacherefficacybetweenteacherswhohaveparticipatedinthePTRinterventionprocess,teacherswhohadsomeparticipationwiththePTRprojectbutdidnotimplementinterventions,andteacherswhodidnotparticipateinPTR?b)IsthereadifferenceinconfidenceindealingwithchallengingbehaviorsbetweenteacherswhohaveparticipatedinthePTRinterventionprocess,teacherswhohadsomeparticipationwiththePTRprojectbutdidnotimplementinterventions,andteacherswhodidnotparticipateinPTR?c)IsthereadifferenceinthefrequencyofABAandPBSrecommendedstrategiesusedforchallengingbehaviorsbetweenteacherswhohaveparticipatedinthePTRinterventionprocess,teacherswhohadsomeparticipationwiththePTRprojectbutdidnotimplementinterventions,andteacherswhodidnotparticipateinPTR?Thesignificancevaluefortheseanalyseswassettop<.0167tomaintainaconservativeestimateofstatisticalsignificancewiththreegroups.NoneoftheANOVAanalyseswerefoundtobestatisticallysignificant.Thisindicatesthatthegroupsdidnot

PAGE 77

70differsignificantlyonanyofthevariablesexploredbytheresearchquestions.Foreachresearchquestion,thenullhypothesiswasnotrejected.Table13ResultsofANOVAforEachDependentVariableSourceSSMSF-valuep Model2.26701.13350.800.4499 Error187.60741.4106EngagementEfficacyCorrectedTotal189.6845Model1.17860.58930.540.5859Error146.02541.0979InstructionalEfficacyCorrectedTotal147.2040Model0.12340.06170.060.9424Error138.23691.0393ManagementEfficacyCorrectedTotal138.3603Model1.2480.62402.980.0542Error27.85660.2094ConfidenceinBehaviorCorrectedTotal29.1046Model299.1521149.57602.160.1191Error9199.958269.1726RecommendedStrategyUseCorrectedTotal9499.1103 Note:df=(2,133)

PAGE 78

71SummaryThischapterdescribedthedataanalysis,screening,andresultsforbothstudies.Thenullhypothesiswasnotrejectedineitherstudy.IntheregressionanalysesinStudy1,neitherteacherefficacynortheteacherefficacysubscales,werefoundtobesignificantlyrelatedtotheamountofcoachingasmeasuredbytheconsultantcoachingtimeorthenumberofcoachingsessions.InStudy2,previousPTRparticipantsdidnotdiffersignificantlyfromnon-participantsinteacherefficacy,confidenceindealingwithchallengingbehavior,oruseofrecommendedstrategies.

PAGE 79

72Chapter5DiscussionSummarySchoolpsychologistsareincreasinglyusingconsultationasaservicedeliverymodelinschoolstoday.Becauseconsultationisanindirectservicedeliverymodel,interventionsdevelopedthroughconsultationareimplementedbyteachersorotherschoolstaff.Oneofthechallengesofconsultationisthefailureofschoolstafftoimplementandsustaininterventionsdevelopedinconsultation.Inorderforconsultationtobeeffectiveinachievingdesiredstudentoutcomes,theteachermustchangehisorherbehaviorasaresultoftheconsultation.Assuringcorrectplanimplementationshouldnotbeoverlookedbytheconsultant.Therefore,consultantsneedtodirectlytrainschoolstaffoncorrectimplementationoftheintervention,monitorimplementation,andprovidefeedback(Han&Weiss,2005;Noell,etal.,2005;Sterling-Turner,etal.,2001).Legislationandparadigmshiftsareincreasingtheneedtomonitortreatmentintegrity.DuetoNCLB,teachersarebeingheldaccountablefortheirpracticenowmorethanever.WiththeintroductionoftheResponsetoIntervention(RtI)modelforspecialeducationeligibility,treatmentintegritymustbeassessedandmonitoredtoensurethattheinterventionisbeingdeliveredasplanned.Informationregardingtreatmentintegrityiscrucialindetermininginterventioneffectiveness.Ironically,treatmentintegritydataarefrequentlyabsentfromliteratureonschool-basedbehavioralinterventions(Gresham,

PAGE 80

73etal.,1993;McIntyre,etal.,2007)andarerarelymonitoredinpractice(Cochrane&Laux,2008).Uponenteringtheconsultativerelationship,teachersbringwiththemdifferentbeliefs,skillsets,efficacylevels,andothermotivations.Thesefactorsmaypotentiallyaffecttheirinitiallevelofimplementationorsustainabilityofanacceptabletreatmentintegritylevel.Theimplementationofabehaviorinterventionrequiresabehaviorchangebytheteacherwhichmaybeskillsthatarenewandrequireefforttoputintopractice.Mostteachersarenottrainedinbehavioralprinciplesandthelackofknowledgemayaffecttreatmentacceptabilityandperceivedeffectiveness(Scott&Nelson,1999;Weigle&Scotti,2000).Teacherefficacyisrelatedtotheteachersbeliefsystemandhasbeenlinkedtoteacherperceptionsofstudentbehavior,teacherbehavior,andstudentoutcomes(Meijer&Foster,2001;Morin&Battalio,2004;Tschannen-Moran&Hoy,2001).Theliteratureregardingteacherefficacyandinterventionimplementationismixed.Tschannen-MoranandHoy(2001)positthatteacherswithastrongsenseofefficacyaremorewillingtotrynewthingstomeettheneedsofstudents.Conversely,Hughesandcolleagues(1990)foundthatteacherswithhighefficacywerelesslikelytochangetheirbehaviorasaresultofconsultation.Thiscontradictionwarrantedfurtherstudyoftherelationshipbetweenteacherefficacyandtreatmentintegrity.ThepurposeofthisresearchwastoexaminetherelationshipbetweenparticipationinthePreventTeachReinforce(PTR)behaviorinterventionprogramandteacherefficacy,confidenceindealingwithchallengingbehavior,andimplementationofrecommendedbehaviorinterventions.Study1exploredthenatureoftherelationship

PAGE 81

74betweenteacherefficacyandtheamountofcoachingneededtoimplementaninterventionwithanacceptablelevelofintegrity.Study2exploredsustainabilityofeffectsofparticipationinPTRbydeterminingifteacherswhohadparticipatedinPTRdifferedfromnon-participantsinteacherefficacy,confidenceindealingwithchallengingbehaviors,anduseofrecommendedbehavioralstrategies.ResultsStudy1usedarchivaldatacollectedduringalargerrandomizedcontrolstudydesignedtodeterminetheefficacyofPTR.Baselinedatacollectedduringthisstudyincludedameasureofteacherefficacy.Teachersworkedwithabehavioralconsultanttodevelopabehaviorinterventionforastudentintheirclassroom.Dataforthirty-sixteacherswereincludedintheanalysis.Consultantstrainedtheteachersandprovidedin-classcoachingonimplementationtotheteachersuntiltheyreached80%treatmentintegrity.Thenumberofcoachingsessionsaveragedfiveandtheamountofcoachingtimeaveraged6.6hours.Regressionanalyseswereconductedtodetermineifoverallteacherefficacyortheteacherefficacysubscalescouldbeusedtopredicttheamountofcoachingtimeneededtoimplementtheinterventionwithanacceptableleveloftreatmentintegrity.Neitherteacherefficacynorthesubscaleswerefoundtobepredictorsofthenumberofcoachingsessionsortheamountofcoachingtimerequired.Nomodelaccountedformorethan5%ofthevariabilityintheoutcomevariable.InStudy2,atotalof136teacherscompletedaweb-basedsurveyincludingquestionsonteacherefficacy,confidenceindealingwithchallengingbehaviors,andfrequencyofuseofrecommendedstrategiestoexplorethesustainabilityoftheeffectsof

PAGE 82

75PTR.ParticipantsweredividedintothreegroupsbasedontheirlevelofparticipationwiththePTRproject.Group1includedteacherswhohadparticipatedinthePTRprocessthroughandincludinginterventionimplementationandcoaching.Group2includedteacherswhohadsomeexposuretoPTRwithoutinterventionimplementation.TeachersinGroup3hadnotparticipatedinPTR.AMANOVAanalysiswasconductedtodetermineifthethreegroupsdifferedonthesetofpredictorvariables(teacherefficacy,confidenceindealingwithchallengingbehaviors,andfrequencyofuseofrecommendedstrategies).Thegroupsdidnotdiffersignificantlyonthesetofvariablescombinedoranyoftheindividualvariables[=.93,F(10,258)=.99,p<.45].ComparisonofStudy1toPreviousResearchPreviousresearchonteacherefficacyandconsultationhasyieldedconflictingresults.Inthecurrentwork,teacherefficacywasnotfoundtobeasignificantpredictoroftheamountofcoachingneededforateachertoimplementaninterventionwithanacceptablelevelofintegrity.Asaresult,therelationshipbetweenteacherefficacyandconsultationremainsunclear.Inthecurrentresearch,ratingsontheTeachersSenseofEfficacyScalewereneartheceilingoftheassessmenttoolwhichmaysuggestthattheteacherswererespondingtothemeasureinasociallydesirablemanner.Astheratingsclusteredatthehighendofthescale,therewaslittlevariabilityinresponseswhichmayhavemadeitmoredifficulttodetectdifferences.Nonetheless,thedistributionofscoreswassimilartothatfoundbytheauthorsinpilottesting.Incontrast,Hughesandcolleagues(1990)askedteacherstoratehowdifficultitwouldbetosolveabehaviorproblemwithouttheassistanceofaconsultant.Thisvaluerepresentedteacherefficacy.Thus,itispossible

PAGE 83

76thattheassessmentsusedtomeasuretheconstructofteacherefficacywereresponsibleforthemixedresults.ThisstudywassimilartotheworkofHughesandcolleaguesinthatitmeasuredtheactualbehaviorofteachersinconsultation.Intheirdiscussion,Hughesandcolleaguesrefertoresearchindicatingthatlessexperiencedteachersaremorelikelytoseekconsultationandchangebehaviorasaresult.Duetothehomogeneityofthesample,differencesinlevelofexperiencecouldnotbetestedinthisstudy.Themajority(56%)oftheteachersinthesamplehadfiveyearsorlessofteachingexperiencewith75%havingtenorlessyearsofexperience.Study1exclusivelyconsideredteacherefficacyasapredictoroftheamountofcoachingneededtoimplementabehaviorinterventionwithintegrity.Inactuality,othervariablesmaybesignificantcontributorstothemodel.Researchhasshownthatteacherswhodonothavesufficientknowledgeofbehavioralprincipleshavedifficultyinbrainstormingappropriateinterventionsandindescribingtheinterventionswithsufficientspecificitytoensureaccurateimplementation(Wilson,etal.,1998).Similarly,teachersmaynothavesufficientprerequisiteskillinimplementationoftheinterventionstoensuresuccess.Otherfactorsrelatedtotheteacherstrainingandexperiencemayalsobesignificantsuchasthetypeofcredentialsheldortheyearsofteachingexperience.Schoolpsychologistsnoteadministrativesupportasabarriertothemeasurementoftreatmentintegrity(Cochrane&Laux,2008).Inmoreauthenticsituations,schoolclimatefactorsmayalsoaffectteacherswillingnesstoimplementinterventions.Treatmentintegrityratesmayincreasewhenschooldistrictsandadministratorsvalueintervention,monitorimplementation,andholdteachersaccountable.

PAGE 84

77Influencesmoredirectlyrelatedtothepresentingproblemmayincludetheseverityofthestudentschallengingbehavior,theleveloftreatmentacceptability,orothermotivatingfactorstheteacherisfacing.Asdiscussedintheliteraturereview,manyteachersmaynothaveknowledgeofbehavioralprinciples(Scott&Nelson,1999).InStudy1,behavioralconsultantswalkedteachersthroughasystematicprocessincludingafunctionalbehavioralassessmentandlinkingtheinterventiontotheassessment.Consistentwithpreviousresearch(Wilsonetal.,1998),mostoftheteachershadattemptedsometypeofunmonitoredorunsystematicinterventionintheclassroombeforebeginningPTR,usuallywithoutsuccess.Usingafunction-basedinterventionwasnewtomanyteachersunfamiliarwiththebehavioraltheoreticalorientation.ThoughitwasnottheintentofPTRtoteachthefunction-basedinterventiondevelopmentprocess,itwasclearthatthisactivitywasforeigntomostparticipants.Likewise,manyoftheteachersimplementedinterventionsthatwerenotintheirexistingskillset.Thenoveltyoftheprocessandinterventionimplementationproceduresmayhaveextendedthetimenecessaryfortheteacherstoestablishanacceptableleveloftreatmentintegrity.Becauseteachersweresoactiveintheprocessandhighlevelsofsocialvaliditywerereported,treatmentacceptabilityisnotbelievedtohavebeenamajorissueinthisstudy.Thoughguidedbytheconsultant,thePTRprocesswasteacher-driveninthattheteacherultimatelyselectedthetargetbehaviorsandinterventions.Whenadisagreementarosebetweentheconsultantsandtheteachersselectionofintervention,theteachersselectionwasusedaslongastheinterventionmatchedwiththefunctionofthebehavior.

PAGE 85

78Theseverityofthebehaviormayhavemotivatedsometeacherstoworkhardertoimplementtheintervention.Challengingbehaviorthatwasfrequent,highinintensity,orverydisruptivemayhavecausedsometeacherstowanttoworkhardtoimplementtheinterventionstoalleviateassociatedproblemsintheclassroom.Itisalsopossiblethatteachers,particularlygeneraleducation,mayhavebeenlesslikelytouseeffortinimplementingbehaviorinterventionsbecausetheirgoalisforthestudenttobeplacedintoanotherclass(Wood,Umbreit,Liaupsin,&Gresham,2007).Inthissituation,improvedstudentoutcomeswouldnotbemotivatingtotheteacherastheywillnothelptheteachertoreachtheirultimategoalremovalofthestudent.Anotherpotentialfactorinfluencingthemotivationwasthestipend.Teacherswerepaidastipendfortheirparticipation,notfortheirsuccessfulimplementationoftheintervention.Theconsultantshypothesizedthattherewereafewteacherswhoparticipatedinthestudyforthestipendastheymademinimalprogresstowardreachinganacceptableleveloftreatmentintegrity.LimitationsofStudy1Potentialforbiaswasintroducedintothestudythroughsamplingbecausetheparticipantsself-selectedtoparticipate,werepaidastipend,andrepresentedalimitedgeographicalarea.AsmallnumberofparticipantsfromthePTRefficacystudymettheinclusioncriteriaforthisstudyresultinginasmallsamplesize.Theseparticipantsmaynotberepresentativeofthesampleofallteachersacrossthepopulationandcautionshouldbeusedingeneralizing.Participantsinthisstudyvolunteeredforaresearchstudy.Therefore,resultsmayhavebeendifferentiftheywereactuallyseekingconsultationfromaschoolpsychologistduetotheirmotivationsandschoolclimate

PAGE 86

79issues.Validityalsowascompromisedduetomortalityasstudentsmoved,orteacherswerere-assignedandorwithdrewthemselvesfromthestudy.Inordertoreduceapossibleproblemwithtreatmentacceptability,teachersweredirectlyinvolvedinthedevelopmentofeachplanandtheinterventionsselectedweredesignedtofitwiththeclassroomenvironmentandcurrentroutines.However,becauseeachinterventionplanwasindividualized,thelevelofdifficultywasnotcontrolled.Thecomplexityofeachinterventionplanvariedintermsofnumberofstepsanddifficultyofimplementation.Theteacherspreviousskilllevelwiththeselectedinterventionsalsowasnotcontrolled.TheresearchprotocolforthePTRstudydidnotspecifythelengthoftimebetweencoachingsessions,thustheintervalvaried.Previousresearchinvestigatingtreatmentintegrityhasemployedastandardizedscheduleforperformancefeedbackrangingfromdailywithprogressivethinningtoeveryotherweek(Coddingetal,2005;Noelletal.,2002).Therefore,itispossiblethatthelengthoftimebetweencoachingsessionscontributedtovariabilityintheamountofcoachingtimerequiredfortheteachertoimplementtheinterventionwithintegrity.ComparisonofStudy2toPreviousResearchStudy2continuedtheexaminationofteacherefficacyconsideringitspossiblerelationshiptosustainabilityofeffectsofparticipationinPTR.Themodelalsoincludedthevariablesofconfidenceindealingwithchallengingbehaviorsandfrequencyofuseofrecommendedstrategies.Thethreegroupsofteachersdidnotdiffersystematicallyonthissetofpredictorvariablesnoronanyoftheindividualvariables.

PAGE 87

80Thefirstresearchquestionofthisstudyfocusedonteacherefficacy.AsnotedaboveinthediscussionofStudy1,teacherefficacyratingsapproachedtheceilingoftheinstrumentandmayhavebeeninfluencedbysocialdesirability.Authorsofthescalealsoindicatethatteacherefficacyisarelativelystableconstruct(Tschannen-Moran&Hoy,2001).Hunzicker(2004)suggeststhatinformationberepeatedovertimetogetteacherstochangetheirbeliefs.Thenextquestionfocusedontheteachersconfidenceindealingwithchallengingbehaviors.Forthisvariablethepossiblescorerangewas0to18.Actualscoresrangedfrom0to12withgroupmeansrangingfrom4.49to5.94.Thoughtheresultingadjustedsignificancevaluewasnotsignificant,itisinterestingtonotethatthehighestmeanscoreonthisvariablewasforteachersinGroup3(PTRnon-participants)andthelowestwasforteachersinGroup2(limitedexposuretoPTR).Itmaybethatteacherswhodidnothaveexperienceinsystematicallyimplementingabehavioralinterventionoverestimatedtheircompetencybecausetheysimplylackedtheknowledgenecessaryforanaccurateself-assessment.Anecdotalaccountssupportthishypothesis.Forexample,oneparticipantinthePTRefficacystudyexpressedconfidenceasshetoldtheconsultantthatshewasknowledgeableaboutbehaviorinterventions.However,thisparticipantexhaustedhercoachingtimewithouteverreachinganacceptablelevelofintegrity.Thelastquestionfocusedontheteachersfrequencyofuseofrecommendedstrategies.Measurementerrormayalsohaveaffectedthisarea.Teachersmayhaveusedtheinterventionsandratedthemselveshighlyonuseofthestrategy,thoughtheywereimplementingitincorrectlyorinconsistently.Thisquestionrepresentsaninformalattempttoinvestigatesustainabilityofinterventionimplementation.However,itis

PAGE 88

81difficulttodiscernifinterventionsdevelopedbythePTRparticipantswerestillbeingimplementedthroughthisgenericquestion.SimilarlytoStudy1,itispossiblethatothervariablesoutsideofthissetofpredictorsareresponsibleformoreofthevariance.Participantfactorssuchasteachingexperience,teachingassignment,andtypeofstudentstaughtwerenotconsidered.Othertheoreticalfactorsmayalsobeinfluencingresponsessuchastheoreticalorientationandpreviousexperiencewithconsultants.LimitationsofStudy2Thisstudyislimitedbyfactorsrelatedtotheparticipantsrecruitedandtheinstrumentsused.Becausetheparticipantsvolunteeredtoparticipate,theirresponsesmaynotreflectthepopulationofteachersexactly.Thesamplewasselectedfromalimitedgeographicalareawhichmayalsolimitgeneralization.Somelimitationswithgroupmembershipmayhavealsobeenintroducedbecauseitispossiblethatsometeachersinthenon-volunteercomparisongroupdidnotvolunteertoparticipateinPTRbecausetheydidnothaveanystudentsinneedofanintensivebehaviorinterventionduringthatschoolyear,thoughtheymayhavedonesoifsuchastudentwereintheirclass.TeacherswhopreviouslyparticipatedinthePTRstudyhadalreadycompletedtheTeachersSenseofEfficacyScaleandQuestionnaireaboutTeachersandChallengingBehaviorsandduringtheirparticipationinthestudyandthushadaleveloffamiliaritywiththeinstruments.ImplicationsandFutureResearchBecausetheresultsofthisstudywerenotstatisticallysignificant,wearenotatthistimeabletouseteacherefficacytopredictwhichteachersaremorelikelytoquickly

PAGE 89

82implementinterventionswithahighleveloftreatmentintegrity.Nonetheless,teacherefficacyinconjunctionwithotherpre-implementationfactorsmaybeusefulindeterminingwhichteachersarereadytoimplementinterventions,thusstreamliningtheconsultationprocess.Withregardtosustainability,anecdotalaccountssuggestthatteachersoverestimatetheirknowledgeofbehavioralprinciples.Teachersmaybenefitfromknowledgeofthescientificfoundationoftheinterventionsbeforeimplementing.Usingasystemsapproachtoprovidesuchtrainingtogroupsofteacherswouldbecosteffectiveasitwouldreducetheamountoftimeneededforthisactivityinindividualconsultation.Schoolpsychologistsshouldcontinuetoseekwaystoreducethetimeneededtogetacorrectlyimplementedinterventioninplaceforthetargetstudent.Decreasingthetimeneededpercasewouldalsoallowtheschoolpsychologisttimetodomoreconsultationandimproveoutcomesforadditionalstudents.ThePTRprocesswasteacher-driven,thus,protocolsusedinthecurrentstudyweredesignedtoalleviatepossibleproblemsinconsultationassociatedwithtreatmentacceptabilityandsocialvalidity.Futureresearchmayusemethodologytofurtherisolateteacherefficacytostudyitseffectontreatmentintegrity.Researchcouldbeconductedwithastandardizedinterventiontoreducetheeffectsoftimerequiredforimplementationandcomplexityoftheintervention.Aspecifiedmonitoringschedulecouldbeusedtoreducetheeffectsassociatedwithvariationintimeinterval.Duetothesmallvariationinteacherefficacyscoresandtheirclusterneartheceiling,additionalinstrumentscouldbeusedtomeasuretheconstructandperhapsdetectdifferences.Theresearchcouldberepeatedwithalargersamplesize.

PAGE 90

83Becauseteacherefficacyislikelynotworkinginisolation,othermodelsshouldbeinvestigatedtodeterminetheinfluenceofadditionalvariables.Futureresearchcouldbeconductedtoevaluatetheeffectsofknowledgeofbehavioralprinciplesandlevelofexpertiseinsystematicallyimplementingbehavioralinterventions.Thisconstructcouldbeexaminedbyaddingitintothecurrentstatisticalmodeltodetermineitsproportionofvarianceexplained.Anotherwaytoinvestigatewouldbetoexperimentallymanipulatethisvariablebyprovidingtraininginbehavioralprinciplesbeforeinitiatingtheintervention.Sucharesearchdesignwouldalsoallowaccuratemeasureofthisvariableinsteadofrelyingoninflatedself-appraisals.Furtheranalysescouldbeconductedtoexaminedifferencesrelatedtotheamountofteachingexperienceandtheteachingassignmentortypeofstudentstaught.

PAGE 91

84ReferencesBandura,A.(1977).Self-efficacy:Towardaunifyingtheoryofbehavioralchange.PsychologicalReview,84,191-215.Cochrane,W.S.&Laux,J.M.(2008).Asurveyinvestigatingschoolpsychologistsmeasurementoftreatmentintegrityinschool-basedinterventionsandtheirbeliefsaboutitsimportance.PsychologyintheSchools,46(6),499-507.Codding,R.S.,Feinburg,A.B.,Dunn,E.K.,&Pace,G.M.(2005).Effectsofimmediateperformancefeedbackonimplementationofbehaviorsupportplans.JournalofAppliedBehaviorAnalysis,38,205-219.Cooper,J.O.,Heron,T.E.,&Heward,W.L.(1987).Appliedbehavioranalysis.UpperSaddleRiver,NJ:PrenticeHall.DiGennaro,F.D.,Martens,B.K.,&McIntyre,L.L.(2005).Increasingtreatmentintegritythroughnegativereinforcement:Effectsonteacherandstudentbehavior.SchoolPsychologyReview,34,220-231.Finley,R.(n.d.)Surveymonkey.com[Computersoftware].Portland,OR:Surveymonkey.com.Foxx,R.M.,(1996).Twentyyearsofappliedbehavioranalysisintreatingthemostsevereproblembehavior:Lessonslearned.TheBehaviorAnalyst,19,225-235.

PAGE 92

85Gable,R.A.,Hendrickson,J.M.,&VanAcker,R.(2001).MaintainingtheintegrityofFBA-basedinterventionsinschools.Education&TreatmentofChildren,24,248-260.Gansle,K.A.(2005).Theeffectivenessofschool-basedangerinterventionsandprograms:Ameta-analysis.JournalofSchoolPsychology,43,321-341.Gansle,K.A.,&McMahon,C.M.(1997).Componentintegrityofteacherinterventionmanagementbehaviorusingastudentself-monitoringtreatment:Anexperimentalanalysis.JournalofBehavioralEducation,7,405-419.Gresham,F.M.(1989).Assessmentoftreatmentintegrityinschoolconsultationandprereferralintervention.SchoolPsychologyReview,18,37-50.Gresham.F.M.&Elliott,S.N.(1990)SocialSkillsRatingSystemManual.CedarPines,MN:AmericanGuidanceService.Gresham,F.M.,Gansle,K.A.,Noell,G.H.,Cohen,S.,&Rosenblum,S.(1993).Treatmentintegrityofschool-basedbehavioralinterventionstudies:1980-1990.SchoolPsychologyReview,22,254-272.Gutkin,T.B.,&Hickman,J.A.(1990).Therelationshipofconsultant,consultee,andorganizationalcharacteristicstoconsulteeresistancetoschool-basedconsultation:Anempiricalanalysis.JournalofEducational&PsychologicalConsultation,1(2),111-122.Han,S.S.,&Weiss,B.(2005).Sustainabilityofteacherimplementationofschool-basedmentalhealthprograms.JournalofAbnormalChildPsychology,33,665-679.

PAGE 93

86Hughes,J.N.,Grossman,P.,&Barker,D.(1990).Teachers'expectancies,participationinconsultation,andperceptionsofconsultanthelpfulness.SchoolPsychologyQuarterly,5,167-179.Hunzicker,J.(2004).Thebeliefs-behaviorconnection:Leadingteacherstowardchange.Principal,84,p.44-46.Jordan,A.,Lindsay,L.,&Stanovich,P.J.(1997).Classroomteachers'instructionalinteractionswithstudentswhoareexceptional,atrisk,andtypicallyachieving.RemedialandSpecialEducation,18,82-93.Kratochwill,T.R.&Bergan,J.R.(1990)Behavioralconsultationinappliedsettings:Anindividualguide.NewYork:PlenumPress.Lane,K.L.,Bocian,K.M.,MacMillan,D.L.,&Gresham,F.M.(2004).Treatmentintegrity:Anessentialbutoftenforgottencomponentofschool-basedinterventions.PreventingSchoolFailure,48,36-43.Lewis,T.J.,&Newcomer,L.L.(2002).Examiningtheefficacyofschool-basedconsultation:Recommendationsforimprovingoutcomes.Child&FamilyBehaviorTherapy,24,165-181.Martens,B.K.,&Ardoin,S.P.(2002).Trainingschoolpsychologistsinbehaviorsupportconsultation.Child&FamilyBehaviorTherapy,24,147-163.McIntyre,L.L.,Gresham,F.M.,DiGennaro,F.D.,&Reed,D.D.(2007).Treatmentintegrityofschool-basedinterventionswithchildreninthejournalofappliedbehavioranalysis1991-2005.JournalofAppliedBehaviorAnalysis,40,659-672.Meijer,C.J.,&Foster,S.F.(1988).Theeffectofteacherself-efficacyonreferralchance.JournalofSpecialEducation,22,378-385.

PAGE 94

87Morin,J.E.(2001)Winningovertheresistantteacher.JournalofPositiveBehaviorInterventions,3,62-64.Morin,J.,&Battalio,R.(2004).Construingmisbehavior:Theefficacyconnectioninrespondingtomisbehavior.JournalofPositiveBehaviorInterventions,6,251-254.Noell,G.H.,Duhon,G.J.,Gatti,S.L.,&Connell,J.E.(2002).Consultation,follow-up,andimplementationofbehaviormanagementinterventionsingeneraleducation.SchoolPsychologyReview,31,217-234.Noell,G.H.,Gresham,F.M.,&Duhon,G.(1998).Fundamentalagreementsandepistemologicaldifferencesindifferentiatingwhatwassaidfromwhatwasdoneinbehavioralconsultation.SchoolPsychologyQuarterly,13,81-88.Noell,G.H.,&Witt,J.C.(1999).Whendoesconsultationleadtointerventionimplementation?Criticalissuesforresearchandpractice.JournalofSpecialEducation,33,29-35.Noell,G.H.,Witt,J.C.,LaFleur,L.H.,Mortenson,B.P.,Rainer,D.D.,&LeVelle,J.(2000).Increasinginterventionimplementationingeneraleducationfollowingconsultation:Acomparisonoftwofollow-upstrategies.JournalofAppliedBehaviorAnalysis,33,271-284.Noell,G.H.,Witt,J.C.,Slider,N.J.,Connell,J.E.,Gatti,S.L.,Williams,K.L.,etal.(2005).Treatmentimplementationfollowingbehavioralconsultationinschools:Acomparisonofthreefollow-upstrategies.SchoolPsychologyReview,34,87-106.

PAGE 95

88Podell,D.M.,&Soodak,L.C.(1993).Teacherefficacyandbiasinspecialeducationreferrals.JournalofEducationalResearch,86,247-253.Riley-Tillman,T.C.,&Eckert,T.L.(2001).Generalizationprogrammingandschool-basedconsultation:Anexaminationofconsultees'generalizationofconsultation-relatedskills.JournalofEducational&PsychologicalConsultation,12,217-241.Robbins,J.R.,&Gutkin,T.B.(1994).Consulteeandclientremedialandpreventiveoutcomesfollowingconsultation:Somemixedempiricalresultsanddirectionsforfutureresearchers.JournalofEducational&PsychologicalConsultation,5,149-167.Rose,D.J.,&Church,R.J.(1998).Learningtoteach:Theacquisitionandmaintenanceofteachingskills.JournalofBehavioralEducation,8,5-35.Scott,T.M.,&Nelson,C.M.(1999).Usingfunctionalbehavioralassessmenttodevelopeffectiveinterventionplans:Practicalclassroomapplications.JournalofPositiveBehaviorInterventions,1,242-251.Sheridan,S.M.,Welch,M.,&Orme,S.F.(1996).Isconsultationeffective?Areviewofoutcomeresearch.RemedialandSpecialEducation,17,341-354.Sterling-Turner,H.E.,Watson,T.S.,Wildmon,M.,Watkins,C.,&Little,E.(2001).Investigatingtherelationshipbetweentrainingtypeandtreatmentintegrity.SchoolPsychologyQuarterly,16,56-67.Stevens,J.P.(1999).Intermediatestatistics:Amodernapproach(2nded.).Manwah,NJ:LawrenceErlbaum.

PAGE 96

89Tournaki,N.,&Podell,D.M.(2005).Theimpactofstudentcharacteristicsandteacherefficacyonteachers'predictionsofstudentsuccess.TeachingandTeacherEducation,21,299-314.Tschannen-Moran,M.,&Hoy,A.W.(2001).Teacherefficacy:Capturinganelusiveconstruct.TeachingandTeacherEducation,17,783-805.UniversityofSouthFlorida&UniversityofColoradoatDenver(2006).PreventTeachReinforceModelTeamManual.Tampa,FL:Author.Upah,K.R.F&Tilly,W.D.(2002).Bestpracticesindesigning,implementing,andevaluatingqualityinterventions.In(Eds.Thomas,A.&Grimes,J.).BestPracticesinSchoolPsychologyIV(pp.483-501).NASP:Bethesda,MD.U.S.DepartmentofEducation.(2002).NoChildLeftBehindActof2001.PublicLaw1-7-110.RetrievedDecember30,2006,fromhttp://www.ed.gov/legislation/ESEA02/Walker,H.M.&Severson,H.H.(1992).SystematicScreeningforBehaviorDisorders:UsersGuideandAdministrationManual.Longmont,CO:SoprisWest.Weigle,K.L.,&Scotti,J.R.(2000).Effectsoffunctionalanalysisinformationonratingsofinterventioneffectivenessandacceptability.JournaloftheAssociationforPersonswithSevereHandicaps,25(4),217-228.Westling,D.L.(2006).TeachersandChallengingBehavior:Knowledge,Views,andPractices.Unpublishedmanuscript,WesternCarolinaUniversity.Wickstrom,K.F.,Jones,K.M.,LaFleur,L.H.,&Witt,J.C.(1998).Ananalysisoftreatmentintegrityinschool-basedbehavioralconsultation.SchoolPsychologyQuarterly,13,141-154.

PAGE 97

90Wilkinson,L.A.(2006).Monitoringtreatmentintegrity:Analternativetothe'consultandhope'strategyinschool-basedbehaviouralconsultation.SchoolPsychologyInternational,27,426-438.Wilson,C.P.,Gutkin,T.B.,Hagen,K.M.,&Oats,R.G.(1998).Generaleducationteachers'knowledgeandself-reporteduseofclassroominterventionsforworkingwithdifficult-to-teachstudents:Implicationsforconsultation,prereferralinterventionandinclusiveservices.SchoolPsychologyQuarterly,13(1),45-62.Witt,J.C.(1997).Talkisnotcheap.SchoolPsychologyQuarterly,12,281-292.Witt,J.C.,Gresham,F.M.,&Noell,G.H.(1996).What'sbehavioralaboutbehavioralconsultation?JournalofEducational&PsychologicalConsultation,7,327-344.Witt,J.C.,Noell,G.H.,LaFleur,L.H.,&Mortenson,B.P.(1997).Teacheruseofinterventionsingeneraleducationsettings:Measurementandanalysisoftheindependentvariable.JournalofAppliedBehaviorAnalysis,30,693-696.Wood,B.K.,Umbreit,J.,Liaupsin,C.J.,&Gresham,F.M.(2007).Atreatmentintegrityanalysisoffunction-basedintervention.EducationandTreatmentofChildren,30105-120.

PAGE 98

91Appendices

PAGE 99

92AppendixA:TeachersSenseofEfficacyScale

PAGE 100

93AppendixB:RecruitmentE-mailforStudy2TargetAudience:PreviousPTRTeachers--InitialSubject:USFPreventTeachReinforceFollow-upSurveyDear[FirstName],MynameisBrandiTannerandIamconsultantforthePreventTeachReinforce(PTR)projectattheUniversityofSouthFloridaandadoctoralcandidateintheSchoolPsychologyprogram.YouarebeingcontactedbecauseyouwerepreviouslyaparticipantinthePTRresearchstudy.Weareconductingasurveytocollectfollowupinformationaboutteacherefficacyindealingwithchallengingbehaviors,useofrecommendedbehavioralstrategies,andparticipationinthePTRprogram.Yourresponsewouldbeappreciated.Thisisaninternet-basedsurveywhichtakesapproximately15minutestocomplete.Iunderstandthatyourtimeisvaluable,andyourparticipationistrulyappreciated.Ifyoucompletethesurvey,yournamewillbeenteredintoadrawingforoneoffiveTargetgiftcardsintheamountof$20.*PleaseNote:IfyouareanemployeeofPinellasCountySchools,monetaryrewardsmaynotbepaidtoschoolboardstaff.Ifyournameisselectedfromthedrawing,youwillbenotifiedandtheawardwillbegiventoyourschool.*Hereisalinktothesurvey:[SurveyLink]Thislinkisuniquelytiedtothissurveyandyouremailaddress.Pleasedonotforwardthismessage.Thanksforyourparticipation!BrandiTanner,Ed.S.UniversityofSouthFlorida btanner2@mail.usf.edu 813-974-3461Pleasenote:Ifyoudonotwishtoreceivefurtheremailsfromus,pleaseclickthelinkbelow,andyouwillbeautomaticallyremovedfromourmailinglist.[RemoveLink]

PAGE 101

94AppendixB:(Continued)TargetAudience:Non-PTRTeachers--InitialSubject:USFChallengingBehaviorSurveyDear[FirstName],MynameisBrandiTannerandIamconsultantforthePreventTeachReinforce(PTR)projectattheUniversityofSouthFloridaandadoctoralcandidateintheSchoolPsychologyprogram.Youarebeingcontactedbecauseyourecentlyprovidedyournameande-mailaddresstoparticipateinasurveyaboutyourconfidenceindealingwithchallengingbehaviorsanduseofrecommendedbehavioralstrategies.Thisisaninternet-basedsurveywhichtakesapproximately15minutestocomplete.Iunderstandthatyourtimeisvaluable,andyourparticipationistrulyappreciated.Ifyoucompletethesurvey,yournamewillbeenteredintoadrawingforoneoffiveTargetgiftcardsintheamountof$20.*PleaseNote:IfyouareanemployeeofPinellasCountySchools,monetaryrewardsmaynotbepaidtoschoolboardstaff.Ifyournameisselectedfromthedrawing,youwillbenotifiedandtheawardwillbegiventoyourschool.*Hereisalinktothesurvey:[SurveyLink]Thislinkisuniquelytiedtothissurveyandyouremailaddress.Pleasedonotforwardthismessage.Thanksforyourparticipation!BrandiTanner,Ed.S.UniversityofSouthFlorida btanner2@mail.usf.edu 813-974-3461Pleasenote:Ifyoudonotwishtoreceivefurtheremailsfromus,pleaseclickthelinkbelow,andyouwillbeautomaticallyremovedfromourmailinglist.[RemoveLink]

PAGE 102

95AppendixC:Study2GroupAssignmentGuidelinesStudy2GroupAssignmentGuidelinesGroup1:PTRInterventionTeachersIncludesteacherswhoparticipatedinthePTRtothepointofdevelopingand/orimplementingabehaviorsupportplanInclusioncriteria:ParticipantswhoindicateYesonitem3(wereyouaparticipant)andendorseletterEonitem4Group2:PTRVolunteerTeachersIncludesteacherswhovolunteeredtoparticipateinPTR,butwereunabletoattainparentalconsentforthestudenttheynominated,assignedtothewaitlist/controlcondition,startedthePTRinterventionprocessbut,forsomereason,didnotmakeittointerventiondevelopment,ortheironlyinvolvementwasdatacollectionforastudentduringthefollow-upphase.Inclusioncriteria:oParticipantswhoindicateYesonitem2(didyounominate)andYesonitem3(wereyouaparticipant)andendorseanyoralloflettersA-Donitem4butNOTitemEoParticipantswhoindicateNoonitem2(didyounominate)andYesonitem3(wereyouaparticipant)andendorseanyoralloflettersA-Donitem4butNOTitemETheseteachersmayhavebecomeparticipantsinPTRbecauseastudentinthestudymovedintotheirclass.Whenapproached,theydidagreetoparticipate.oParticipantswhoindicateYesonitem2(didyounominate)andNoonitem3(wereyouaparticipant)Theseteachersmayhavenominatedastudentindicatingwillingnesstoparticipatebutwereunabletoattainconsent,studentmoved,etc.Exclusioncriteria:oParticipantswhoindicateYesonitem3(wereyouaparticipant)andendorseitemE(participatedinintervention)-->Group1oParticipantswhoindicateNoonitem2(didyounominate)anditem3(wereyouaparticipant)-->Group3Group3:Non-PTRTeachersTeacherswhodidnotvolunteertoparticipateinthePTRinterventionstudy.oInclusioncriteria:ParticipantswhoindicateNoonitem2(didyounominate)anditem3(wereyouaparticipant)-->Group3NominatedParticipantGroup Group1ifendorsingEonitem4 YesYesGroup2ifNOTendorsingEonitem4 YesNoGroup2Theseteachersmayhavenominatedastudentindicatingwillingnesstoparticipatebutwereunabletoattainconsent,studentmoved,etc. NoYesGroup2TheseteachersmayhavebecomeparticipantsinPTRbecauseastudentinthestudymovedintotheirclass.Whenapproached,theydidagreetoparticipate NoNoGroup3

PAGE 103

96AppendixD:Study2Survey

PAGE 104

97AppendixD:(Continued)

PAGE 105

98AppendixD:(Continued)

PAGE 106

99AppendixD:(Continued)

PAGE 107

100AppendixD:(Continued)

PAGE 108

101AppendixD:(Continued)

PAGE 109

102AppendixD:(Continued)

PAGE 110

103AppendixD:(Continued)

PAGE 111

104AppendixD:(Continued)

PAGE 112

105AppendixD:(Continued)

PAGE 113

AbouttheAuthorBrandiTannerearnedherBachelorsDegreeinElementaryEducationandMastersDegreeinSpecialEducationfromWestVirginiaUniversity.Afterfouryearsasaspecialeducationteacher,shereturnedtotheuniversitysettingtopursuetraininginschoolpsychology.Ms.TannerholdsaMastersDegreefromGeorgiaStateUniversityandaSpecialistsDegreefromtheUniversityofSouthFloridabothinSchoolPsychology.SheisaNationallyCertifiedSchoolPsychologist.Duringdoctoraltraining,Ms.Tannerhastaughtgraduateandundergraduatelevelcourses.Shehasalsoworkedinavarietyofsettingsincludingpublicschools,privatepractice,andinresearch.Ms.Tannersresearchinterestsincludebehavioralinterventions,consultation,andtreatmentintegrity.


xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200397Ka 4500
controlfield tag 001 002028697
005 20090911144117.0
007 cr mnu|||uuuuu
008 090911s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002805
035
(OCoLC)436297169
040
FHM
c FHM
049
FHMM
090
LB1051 (Online)
1 100
Tanner, Brandi L.
0 245
Teachers' sense of efficacy and use of behavioral interventions :
b consultation effects and sustainability
h [electronic resource] /
by Brandi L. Tanner.
260
[Tampa, Fla] :
University of South Florida,
2009.
500
Title from PDF of title page.
Document formatted into pages; contains 105 pages.
Includes vita.
502
Dissertation (Ph.D.)--University of South Florida, 2009.
504
Includes bibliographical references.
516
Text (Electronic dissertation) in PDF format.
3 520
ABSTRACT: The amount of services delivered by school psychologists through consultation is increasing as is the number of students with challenging behaviors in the classroom. In this type of delivery model, the school psychologist works as a consultant to the teacher who will actually deliver the intervention to the student. The purpose of this research was to examine the relationship between participation in a tertiary level behavior intervention program and teacher efficacy, confidence in dealing with challenging behavior, and implementation of behavior interventions. Two studies were conducted to examine a series of research questions. Study 1 used archival data to examine the influence of teacher efficacy before participation in a tertiary level behavior intervention on the amount of coaching necessary for a teacher to implement an intervention with an acceptable level of integrity. To explore sustainability, Study 2 used a survey of teachers who had participated in a tertiary behavior intervention as well as teachers who had not, to determine if they differed on teacher efficacy, confidence in dealing with challenging behaviors, and use of recommended behavioral strategies. Teacher efficacy was not found to be a statistically significant predictor of the amount of coaching time necessary for the teacher to implement the intervention with integrity. It is hypothesized that other factors such as readiness to change may be contributing to the model. PTR participants did not significantly differ from non-participants in any of the proposed areas. It is possible that non-participants over-estimated their knowledge and abilities. Future research should continue to explore the effects of consultation and its sustainability and while considering these additional factors.
538
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
590
Advisor: Harold Keller, Ph.D.
653
Student behavior
Teacher behavior
Treatment integrity
Teacher beliefs
Intervention training
690
Dissertations, Academic
z USF
x Psychological and Social Foundations
Doctoral.
773
t USF Electronic Theses and Dissertations.
4 856
u http://digital.lib.usf.edu/?e14.2805