supervision goals ABA
Which of the following three common supervision issues do you think will be most
challenging for you? Why?
Managing supervision with other responsibilities (time management)
Providing effective and individually-tailored supervision to each supervisee
Deciding how to act when an RBT/supervisee is not performing to standards
What would you do to deal with this issue?
Be sure to support your Assignment with specific references to behavior-analytic theory
and research. Search the internet for peer-reviewed articles to support your Assignment.
Use proper APA format and citations, including those in the Learning Resources.
Sample Supervision Contract for
BCBA/BCaBA Trainees
The following sample contract describes a relationship between a supervisor and a trainee
who is acquiring experience in applied behavior analysis toward an eligibility requirement
for either BCBA or BCaBA certification. The sample contract is provided for general
reference and informational purposes only and should not be construed as legal advice. The
BACB does not warrant or guarantee the legal enforceability of any provision contained in
the sample contract. Contracts should be drafted by individuals familiar with local and state
laws and requirements.
The terms identified in the sample contract below are in no way intended to be
comprehensive. We have identified key clauses as they relate to BACB requirements;
however, each party may wish to consult with legal counsel to discuss the following
contract elements: storage of and rights to client documentation; release and
indemnification; mediation and arbitration; consequences for failure to pay/reimburse
supervisor; legal remedies; injunctive relief; cost and attorney’s fees; contract
modifications; severability; waivers; governing law; contacts for receipt of notice; and if the
experience is offered through a university, a statement about whether the supervision
contract is supplemented with a syllabus (as opposed to replacing a syllabus). In addition,
each party should address with legal counsel the comments embedded in the sample
contract to ensure that the contract is individualized for each unique situation.
If the supervisor is not affiliated with the trainee’s experience setting, as might be the case
with an independently contracted supervisor or a university faculty member, it is strongly
recommended that the supervisor and trainee meet with relevant personnel (e.g., program
director, on-site clinical supervisor) at the trainee’s experience setting to (a) ensure that all
provisions in this contract can be faithfully executed in the setting, (b) familiarize the off-
site supervisor with setting-specific rules, regulations, and practices, and (c) identify if and
how the off-site supervisor’s recommendations will be incorporated into the setting. Finally,
trainees who are receiving their supervision from an employer should meet to discuss
potentially challenging dual relationship (i.e., employer and experience supervisor) issues.
======================================================================
============
Sample Contract
This Supervision agreement is made on _______ (date) between _________________(supervisor
name & certification #) and __________________ (trainee name) for supervision of activities
under the experience standards.
Experience Activities and Training Objectives:
During the experience, the trainee will engage in the following types of activities: ___
By the end of the experience, the trainee will: ___
Trainee Agrees to:
Provide proof of the completed to the supervisor prior to the onset of experience.
Begin experience only after starting qualifying coursework.
Compensate the supervisor at an hourly rate of $___. Payment will be made immediately
upon conclusion of each supervision session or by a mutual agreement arranged at the
outset of the supervisory relationship.
Attend additional professional meetings relevant to the experience, such as ___, which might
not count towards the BACB’s experience requirements.
Supervisor Agrees to:
Meet the BACB’s supervision eligibility requirements (see BACB experience standards) and
provide proof of eligibility to the trainee prior to the onset of experience.
Not accept remuneration from the trainee above the negotiated hourly rate or salary
identified above.
Provide supervision in accordance with the , including, but not limited to providing
supervision to the trainee only within the boundaries of his/her competence.
Provide ongoing documented performance feedback to the trainee in accordance with
sections 5.05 and 5.06 of the including, but not limited to, providing feedback on the
trainee’s progress toward stated training objectives (e.g., including specific areas where the
trainee is making adequate or inadequate progress) and client interactions.
Sign the Monthly and Final Experience Verification Forms if the experience hours are
completed satisfactorily.
Withhold signature on the Monthly and Final Experience Verification Form in the following
situations: ___
Terminate the supervision relationship in the following situations:
Experience hours are completed satisfactorily.
Progress is not made after ___ instances of corrective feedback.
Egregious violations of the BACB’s .
Trainee and Supervisor Agree to:
Meet and stay up-to-date on the BACB experience standards (e.g., requirements for
documentation, acceptable activities) for the duration of the experience.
Abide by the (e.g., confidentiality, consent for services, professionalism) and relevant laws.
Work together to facilitate in-depth discussion/understanding of issues affecting practice -
both personally and professionally.
Keep all information relating to current or potential clients including, but not limited to, any
medical or clinical information, confidential in whatever form.
Ensure that clients have consented to the observation of service delivery and sharing of
confidential client information.
Work together to facilitate in-depth discussion and understanding of issues affecting
practice to improve the trainee’s behavior-analytic expertise.
Protect the time and space for supervision by keeping to agreed upon appointments and the
time allotted. Privacy will be respected and interruptions avoided. Any party requiring a
variance in schedule will notify the other party as early as possible if rescheduling of
supervision is needed.
Retain this contract, documentation of experience and supervision, and the Monthly and
Final Experience Verification Forms for seven years.
Termination:
This agreement shall be effective on ___ and shall remain in effect until either party
terminates this agreement by providing ___ day’s advance written notice to the other of the
intention to terminate. In the event of termination, clause ___, shall survive and remain valid.
Miscellaneous Provisions
--------------------------------------------------------------------
Entire Agreement. This contract contains the entire agreement between the Parties hereto
with respect to the subject matter hereof.
All parties agree to the stipulations herein:
Trainee Name: ____________________________________
Trainee Signature: ____________________________________ Date: ___________________
Supervisor Name: ____________________________________ Certification #: ____________
Supervisor Signature: ____________________________________ Date: _________________
RESEARCHARTICLEAnEvaluationoftheImpactofSupervisionIntensity,SupervisorQualificatio
ns,andCaseloadonOutcomesintheTreatmentofAutismSpectrumDisorderDennisR.Dixon1&Er
ikLinstead2&DoreenGranpeesheh1&MarlenaN.Novack1&RyanFrench2&ElizabethStevens2
&LauraStevens2&AlvaPowell1Publishedonline:9June2016#AssociationforBehaviorAnalysis
International2016AbstractAmpleresearchhasshownthebenefitsofintensiveappliedbehavior
analysis(ABA)treatmentforautismspec-
trumdisorder(ASD);researchthatinvestigatestheroleoftreatmentsupervision,however,islimit
ed.Thepresentstudyexaminedtherelationshipbetweenmasteryoflearningobjec-
tivesandsupervisionhours,supervisorcredentials,yearsofexperience,andcaseloadinalargesa
mpleofchildrenwithASD(N=638).Thesedatawereretrievedfromalargearchi-
valdatabaseofchildrenwithASDreceivingcommunity-
basedABAservices.Whenanalyzedtogetherviaamultiplelinearregression,supervisionhoursan
dtreatmenthoursaccountedforonlyslightlymoreoftheobservedvariance(r2=0.34)thantreatm
enthoursalone(r2=0.32),indicatingthatincreasedsupervisionhoursdonotdramaticallyincreas
ethenumberofmasteredlearningobjectives.Inadditionalre-
gressionanalyses,supervisorcredentialswerefoundtohaveasignificantimpactonthenumberof
masteredlearningobjec-
tives,whereinthosereceivingsupervisionfromaBoardCertifiedBehaviorAnalyst(BCBA)master
edsignificantlymorelearningobjectives.Likewise,theyearsofexperienceasaclinicalsupervisors
howedasmallbutsignificantimpactonthemasteryoflearningobjectives.Asupervisor’scaseload,
however,wasnotasignificantpredictorofthenumberoflearningobjectivesmastered.Thesefindi
ngsprovideguid-
anceforbestpracticerecommendations.KeywordsAutismspectrumdisorder.Supervision.Appl
iedbehavioranalysis.TreatmentoutcomesAppliedbehavioranalysis(ABA)isawell-
establishedframe-
workforthetreatmentofautismspectrumdisorder(ASD;Eldeviketal.,2009;Reichow,2012;Reic
howetal.,2012).ABA-basedtreatmentisconductedatahighintensity,typical-
lybetween30and40h/week,formultipleyears,oftenbegin-
ninginearlychildhood(Eldeviketal.,2009;Reichowetal.,2012).Whileastrongconsensusexiststh
atABAisaneffec-
tivetreatmentforASD,evidencealsoindicatesagooddealofvarianceinindividualresponsetotrea
tment(Eldeviketal.,2010;Howlinetal.,2009).Severalfactorshavebeensuggestedtohaveaneffec
tonABAtreatmentoutcomes.Somefactorsarespecifictotheindividualatthestartoftreatment;for
instance,youngerage(Ben-
Itzchak&Zachor,2011;Eldeviketal.,2012;Flanaganetal.,2012;Granpeeshehetal.,2009;Harris&
Handleman,2000;Makrygianni&Reed,2010;Perryetal.,2011;Virués-
Ortegaetal.,2013),higherIQ(Ben-
Itzchak&Zachor,2007;Eikesethetal.,2002,2007;Eldeviketal.,2006;Eldeviketal.,2010;Eldevike
tal.,2012;Harris&Handleman,2000;Haywardetal.,2009;Magiatietal.,2007;Magiatietal.,2011;P
erryetal.,2011;Remingtonetal.,2007;Smithetal.,2010),lowerseverityofASDsymptoms(Ben-
Itzchak&Zachor,2011;Eldeviketal.,2012;Perryetal.,2011;Remingtonetal.,2007;Smithetal.,20
00),greateradaptiveskills(Eldeviketal.,2010;Flanaganetal.,2012;Magiatietal.,2011;Makrygia
nni&Reed,2010;Perryetal.,2011;Remingtonetal.,2007;Sallows&Graupner,2005),strongerlan
guageskills(Ben-
Itzchak&Zachor,2011;Eldeviketal.,2006;Magiatietal.,2007;Magiatietal.,2011;Sallows&Graup
ner,2005),andgreatersocialskills(Ben-
Itzchak&Zachor,2007;Sallows&Graupner,2005)havebeen*DennisR.Dixond.dixon@centerfor
autism.com1CenterforAutismandRelatedDisorders,21600OxnardStreet,Suite1800,Woodlan
dHills,CA91367,USA2SchmidCollegeofScienceandTechnology,ChapmanUniversity,OneUnive
rsityDrive,Orange,CA92866,USABehavAnalysisPractice(2016)9:339–
348DOI10.1007/s40617-016-0132-1
associatedwithsuperioroutcomes.Otherfactorsaretreatmentspecific;forexample,greatertreat
mentintensity(Eldeviketal.,2010;Granpeeshehetal.,2009;Makrygianni&Reed,2010;Remingto
netal.,2007),longertreatmentduration(Luisellietal.,2000;Makrygianni&Reed,2010),andgreat
eroverallinterventiontime(Virués-Ortega,2010;Virués-
Ortegaetal.,2013)havebeenshowntohaveapositiveimpact.Althoughresearchislimited,thereis
evidencetosuggestthatvariablesrelatedtothesupervisionofABA-basedtreat-
mentalsosignificantlycontributetotreatmentoutcome.Forexample,ameta-
analysisconductedbyReichowandWolery(2009)examinedtherelationshipbetweensuperviso
rtrainingmodelsandtreatmentoutcomes.Theirfindingssuggestedthatstudiesthatimplemente
dsupervisor-
trainingprotocolsbasedontheUniversityofCalifornia,LosAngeles(UCLA)modelproducedgreat
ergainsinIQthanstudiesthatemployedothertrainingprocedures.Forthemostpart,evaluationof
theimpactofsupervisionontreatmentoutcomeshasbeenlimitedtotreatmentprogramsthatarep
arentmanaged,meaningparentsareresponsibleformanagingtheimplementationoftheirchild’s
treatmentpro-
gramwhilereceivingsomedegreeofclinicaloversightfromaprofessional.Severalstudieshavebe
enconductedtoevaluateparent-
managedtreatment,combinedwithvaryinglevelsofprofessionalsupervision,asacost-
effectivealternativetoclinic-
basedtreatmentprograms.Forexample,Bibbyetal.(2002)foundparent-
managedABAprograms(describedindetailbyMudfordetal.,2001)toproducerelativelypoortre
atmentoutcomesascomparedtotheclinic-
basedtreatmentoutcomesreportedbyLovaas(1987).Anumberoffactorsweresuggestedbythea
uthorstohavecontributedtothedis-
crepantoutcomes,includingolderageandlowerIQatstartoftreatment,fewertreatmenthours,inf
requentsupervisioncon-ductedaboutonceevery3months,andlesscompetentsuper-
visors(approximately80%ofwhomwerenottrainedtoLovaastreatmentmodelstandards).Ther
efore,thoughtheoutcomeswereclearlypoorerthanthosedocumentedbyLovaas(1987),therole
ofsupervisionintheworkbyBibbyetal.(2002)isdifficulttoevaluate,asitisjustoneofnumer-
ousfactorsthatmayhaveimpactedtreatmentoutcomes.Otherstudiesevaluatingtheeffectivenes
sofparent-
managedABAtreatmentprogramswhencombinedwithmorefrequentsupervisionthanreporte
dbyBibbyetal.(2002)haverevealedbetteroutcomes.BothSallowsandGraupner(2005)andHay
wardetal.(2009)comparedparent-managedtreatmenttoclinic-
basedtreatmentandfoundparticipantsinbothgroupstomakemeaningfulgainswithnosignifican
tdifferencesdetectedbetweengroups.SallowsandGraupner(2005)observedsimilartreatment
out-comesbetweengroupsdespitethefactthatlessfrequentsu-pervisionwasgiventotheparent-
managedtreatmentgroup.However,inafurtherexaminationoftheparent-
managedtreatmentgroupdescribedbyHaywardetal.(2009),Eikesethetal.(2009)identifiedastr
ongrelationshipbetweengreatersupervisionintensityandimprovedtreatmentout-
comes.Supervisionintensity,whichrangedfrom2.9to7.8h/month,wassignificantlycorrelated
withimprovementsinIQatfollow-
up.Onaverage,IQincreased0.21pointsforeachhourofsupervisionwithnodetectiblepointofdimi
nishingreturns.Giventhesestudies,therelationshipbe-
tweensupervisionintensityandtreatmentoutcomesisunclear.Whilethereissomeevidencetosu
ggestthatsupervisionin-tensitycorrelateswithtreatmentoutcomesinparent-
managedtreatmentprograms(e.g.,Eikesethetal.,2009),suchresearchhasnotyetbeenconducted
inclinic-
basedtreatmentsettings.InspiteoflimitedresearchontheroleofsupervisioninABAprograms,eff
ortshavebeenmadetopromoteuniformi-
tyintreatmentprovision.TheBehaviorAnalystCertificationBoard(BACB),establishedin1998,is
amongtheleadingorganizationshelpingtosetstandardsinthefield.TheBACBsummarizedbestp
racticesforsupervisionofABA-
basedautismtreatmentinitspracticeguidelinesforfundingagencies(BACB,2014).Whiletheindi
vidualdemandsofeachcasemustbetakenintoaccount,theBACBspecifiessupervisionconducted
ataratioof2haweekperevery10hoftreatmentastherecommendedstandard,withamin-
imumof2hofsupervisionprovidedaweek.Thisreflectsanincreaseintherecommendedsupervisi
onhoursrelativetothepreviouslypublishedBACBguidelines,whichgavearangeof1–
2supervisionhoursforevery10hoftreatment(BACB,2012).TheBACBalsodescribesaveragecas
eloadsizesforsupervisorsoverseeingcomprehensiveABAtreatmentpro-
gramstorangebetween6and16cases,dependingonthetreatmentintensityanddemandsofeachc
ase,competencyandaccessibilityofthesupervisor,andthesupervisor’slevelofsupport.Averagec
aseloadsforsupervisorsoverseeingfo-
cusedtreatmentprogramsarespecifiedbytheBACBasrang-
ingbetween10and24cases.Theserecommendationshavebeensuggestedasbestpracticesinthef
ield;nonetheless,existingresearchdoesnotestablishwhethertheserecommen-
dationsproducesuperiortreatmentoutcomes.TheBACBhasrecommendedstandardsforsuperv
isorqualifications,aswell.TheBACBoffersaBoardCertifiedBehaviorAnalyst(BCBA)certificatio
nforclini-
ciansinthefield.TobecomeaBCBA,applicantsmust,asofJanuary1,2016,holdamaster’sdegreein
behavioranalysis,education,orpsychology(previouslyacceptedmaster’sdegrees,whichmaybe
tterrepresentthecurrentBCBApopulation,includebehavioranalysisorrelatedfieldorothernatu
ralscience,education,humanservices,engineering,ormedicine);satisfyspecificcourseworkre-
quirementsinbehavioranalysis;haveaspecificnumberofworkexperiencehoursdirectlysupervi
sedbyaBCBA;andpassanexam.TheBACBalsooffersadoctoralBCBAcertification(BCBA-
D)forthosewhoholdaqual-
ifyingdoctoraldegreeandsatisfyallotherBCBA340BehavAnalysisPractice(2016)9:339–348
certificationrequirements.Additionally,abachelor’slevelcertification,BoardCertifiedAssistant
BehaviorAnalyst(BCaBA),isoffered;however,theBACBstipulatesthatanysupervisionprovided
byaBCaBAmustbeoverseenbyaBCBAorBCBA-D.Despitetherigorousrequire-
mentstoobtainaBCBA,itshouldbenotedthatABAisabroadfieldnotlimitedonlytothetreatmentof
ASD,andobtainingaBCBAdoesnotnecessarilyindicatecompetencyinthetreatmentofASD(Eikes
eth,2010;Loveetal.,2009).Therefore,trainingandsupervisedworkexperiencesinABAtreatmen
tspecificallyforASDaretypicallyrecommendedinadditiontocertifica-
tion(Eikeseth,2010)andfallwithintheBACB’srequire-
mentthatcertificantspracticewithinthescopeoftheirexperience.Toidentifythepercentageofth
osewithgraduatede-
greeswhoalsoholdaBACBcertification,Loveetal.(2009)surveyedalargegroupofASDtreatment
pro-viders.Fromthesurveyof211supervisors,72%ofre-
spondentsreportedhavingagraduatedegree,and42%reportedhavingaBCBAorBCBA-
D.Thesefindingsmayreflectanefforttosupplementtheinsufficientnum-
berofsupervisorswhopossessBACBcertificationstomeetthehighdemandforABAservices.Addi
tionally,itshouldbenotedthatABAtreatmentservicesforASDhavebeenprovidedforover30year
s,predatingcertifica-tionefforts.Assuch,manywell-
trainedandexperiencedcliniciansarenotcertified,includingindividualswhopioneeredtheappli
cationofABAtothetreatmentofASD.Additionally,BCBAcertificationisonlyoneofmanycredentia
lsrecognizedbycurrentandemergingstateinsurancemandatesthatoftenspecifytheeducation,tr
aining,certification,and/orlicensurerequiredtosuper-
viseABAprograms.Whilerequirementsvaryfromstatetostate,otherrecognizedprofessionalsin
cludelicensedpsychologists,marriageandfamilytherapists,speechandlanguagepathologists,o
ccupationaltherapists,andaudiologistspracticingwithinthescopeoftheirlicensureandcompete
ncy.AlthoughnumerousstatelawsdefinewhomaysuperviseABAprogramsforindividualswithA
SD,researchevaluatingwhethersuchqualificationsac-
tuallyleadtosuperiortreatmentoutcomeshasnotyetbeenconducted.Giventhelackofempiricale
videncetoguidethede-velopmentofbestpracticeguidelinesforsupervisionofABA-
basedASDtreatment,thepurposeofthepresentstudywastoexaminetherelationshipbetweenfac
torsrelatedtosupervisionandABAtreatmentoutcomes.Specifically,thepresentstudytestedthe
hypothesisthatsupervisionhours,supervisorcredentials,yearsofexperi-
ence,andcaseloadwouldbesignificantpredictorsofthenumberofmasteredlearningobjectivesw
ithinalargedatasetcollectedfromacommunity-
basedclinicalsetting.MethodsParticipantsClinicalrecordsweregatheredfromapoolof836child
renbetweentheagesof18monthsand12yearswhowerereceiv-ingABA-
basedservicesfromacommunity-basedautismtreatmentproviderduringa12-
monthperiod(January1,2014throughDecember31,2014).Recordsweresubjecttothefollowing
inclusioncriteria:adiagnosisofASD(AmericanPsychiatricAssociation,2013),autisticdisorder(
AmericanPsychiatricAssociation,2000),pervasivedevelop-mentaldisorder-
nototherwisespecified(PDD-NOS;AmericanPsychiatricAssociation,2000),orAsperger’sdis-
order(AmericanPsychiatricAssociation,2000)byaninde-
pendentlicensedclinician(e.g.,psychologistandpediatri-cian);atleast20hofABA-
basedtreatmentpermonth;andatleastonefullmonthofcontinuousservices.Thesecriteriaprodu
cedasamplesizeof638clinicalrecords.Theage,diag-
nosis,andgenderprofilesoftheindividualswhoseclinicalrecordswereusedinthestudywereasfo
llows:528males(agerange2.08–
11.92years,meanage7.42years,317autisticdisorder,166ASD,41PDD-
NOS,4Asperger’sdisorder)and110females(agerange3.17–
11.83years,meanage7.53years,73autisticdisorder,30ASD,6PDD-NOS,1Asperger’sdis-
order).Themeanageoftheindividualswhoserecordsmadeupthissamplewas7.44years(SD=2.3
0).Theaveragenum-
beroftreatmenthoursreceivedpermonthwas71.01(SD=35.26),rangingfrom20.02to197.30h/
month.Anav-
erageof10.98(SD=6.50)supervisionhourswerereceivedpermonth,rangingfrom1.40to67.40.F
urthermore,anaver-
ageratioof1.77(SD=1.14)supervisionhourswereprovidedforevery10hoftreatment,rangingfro
m0.25to9.73.Theaveragenumberofmasteredlearningobjectivespermonthwas31.42(SD=34.4
7),rangingfrom1to245.75permonth.Individualswhoserecordswereincludedinthissampleresi
d-
edandreceivedservicesinthestatesofArizona,California,Colorado,Illinois,Louisiana,NewYork,
Texas,andVirginia.DataCollectionTreatmentdatawerecollectedretrospectivelyfromalargearc
hivaldatabase.Throughouttreatmentdelivery,theSkills™systemwasusedtoidentifydevelopme
ntaldeficits,designindividualizedtreatmentprograms,andtrackongoingprogress.TheSkills™A
ssessmentisaninstrumentthatcom-
prehensivelyevaluatesskillsacrossallareasofchilddevelop-
ment(Dixonetal.,2011).AstudybyPersickeetal.(2014)evaluatedthevalidityoftheSkills™Assess
mentbycontrast-ingparentresponsetotheSkills™itemswithdirectobserva-
tion.Pearsonproduct-
momentcorrelationcoefficientsrangedfrommoderate(r=0.65)tohigh(r=0.95).Treatmentdata
werecombinedwiththebehavioralhealthagency’sBehavAnalysisPractice(2016)9:339–
348341
operationaldata,includingtreatmenthours,supervisionhours,supervisorcredentials,yearsofe
xperience,andcaseload.TreatmentTreatmentprogramswereindividualizedaccordingtoeachp
articipant’sspecificstrengthsanddeficits.Treatmentpro-
gramsaddressedalldevelopmentalareasinwhichthepartic-
ipantdisplayeddeficits,includinglanguage,academics,socialskills,playskills,motorskills,adapti
veskills,executivefunc-
tions,andcognition.Serviceswereprovidedinthehome,school,community,clinic,oracombinati
onofsettings,de-
pendingonfundingagencyrequirementsandothervariables.Alltreatmentprogramsinthisstudy
followedtheCARDmodeloftreatmentdelivery(Granpeeshehetal.,2014)andthereforesharedth
efollowingcommonalities:(a)trainedbe-havioraltherapistsdeliveredone-to-
onetreatment;(b)bothdiscretetrialtrainingandnaturalenvironmenttrainingstrate-
gieswereimplemented;(c)averbalbehaviorapproachwasusedforlanguageintervention;(d)bot
herrorlessandleast-to-
mostpromptingstrategieswereimplemented;(e)empiricallyvalidatedbehavioralprinciplesan
dprocedureswereusedasneeded,includingreinforcement,extinction,stimuluscontrol,generali
zationtraining,chaining,andshaping;(f)afunction-
basedapproachwasimplementedfortheassessmentandtreat-
mentofchallengingbehaviors;(g)parentsreceivedtrainingregularlyandwereincludedinalltrea
tmentdecisions;and(h)directsupervisionwasprovidedonaregularbasis(e.g.,bi-
weekly).Thenumberoftreatmenthoursperparticipantwascollectedfrombillingrecordsandincl
udedalldirecttreatmentservicesprovidedtotheparticipant.Activitiesthatwerenotclient-
specific,suchasattendingtraining,orwerenotdirecttreatmentservices,suchastravelingtopartic
ipant’shome,wereexcluded.Masteryoflearningobjectiveswasusedasthedependentvariablefor
allanalyseswithinthisstudy.Thedefinitionofmasteryofalearningobjectivewassetonanindividu
albasisbythetreatmentsupervisorbutwasrequiredtobewithintheboundsofthefollowingcriteri
a:>70%accuracyofrespondingtothelearningobjectiveforaminimumoftwotreatmentsessionsa
crosstwodifferentdays.Typically,amorestringentmasterycriterionof80%accuracyisrequired,
butsupervisorshavethediscretiontodeviateiftheyfeelitisclinicallyappropriatetodoso.Supervis
ionAllsupervisorsinthepresentstudyreceivedaminimumof6monthsoftraininginABA-
basedtreatmentforASDandearnedacertificationinsupervisionfromtheInstituteforBehavioral
Training.Amultifacetedtrainingapproachwasused,whichincludedacombinationofeLearning(
www.ibehavioraltraining.com),classroom-styletraining,webclas-
ses,andmentorship.Supervisorsreceivedmentorshiponaweeklybasis,whichinvolveddirectob
servation,feedback,andfollow-
uptrainingtoimproveclinicalskills.Examswereadministeredatvariousstagesofthetrainingpro
gram,andtraineeswererequiredtodemonstratefluencyintrainingma-
terialbeforeadvancingtothenextstage.Attheendofthetrainingprogram,supervisorswererequi
redtodemonstrateclinicalcompetencybypassingawrittenpracticumandoralexam.Supervisors
inthepresentstudywereresponsibleforover-
seeingparticipants’treatmentprograms.Thenumberofsuper-
visionhoursperparticipantwascollectedfrombillingrecords.Supervisionhourswererequiredt
obeclient-
specificandwerecomposedofbothdirectandindirectservices,including:(a)makingclinicalreco
mmendationsontreatmentintensityandduration,(b)conductingassessments,(c)developingin
divid-
ualizedtreatmentplans,(d)holdingregularlyscheduledclinicmeetingswithfamiliesandtherapi
sts,(e)observingtreatmentsessions,(f)reviewingdataandadjustingtreatmentplansac-
cordingly,(g)reportingontreatmentprogress,(h)consultingwithteachersandotherserviceprov
iders,(i)conductingther-apistandparenttrainingtoimplementclient-specificproto-
cols,and(j)preservingtreatmentintegrity.Activitiesexcludedfromtheanalysiswereasfollows:(
a)conductingclientin-
takes,(b)conductingtherapistperformanceevaluations,(c)providingstafftrainingsthatwereno
tclient-specific,(d)de-
velopingdischargeplans,and(e)traveltoclienthomes.Ingeneral,supervisionwasprovidedatami
nimumratioof1hofsupervisionperevery10hoftreatmentinaccordancewiththebestpracticesse
tforthatthetimetheseservicesweredelivered(BACB,2012).Foramoredetaileddescriptionofthe
supervisortrainingandresponsibilitiesinvolvedinthepresentstudy,seeGranpeeshehetal.(201
4).Supervisorcaseloadwasdeterminedbycountingthenum-
berofclientsassignedtoeachsupervisorduringa1-
monthinterval(Octoberof2014)withinthelargerperiodoftimethatrecordswerereviewed(Janu
ary1,2014throughDecember31,2014).Giventhateachsupervisordoesnotworkthesamenumbe
rofhourseachweekbutcanrangefromparttimetofulltimeandthattheirworkhoursdirectlyimpa
ctthenumberoftreatmentprogramsthateachcansupervise,caseloadsweredividedbythenumbe
rofhoursthatthesupervisorworkedperweekduringthesame1-
monthinterval.Forexample,apart-
timesupervisorwithacaseloadof15,whoworks25h/week,wouldhaveaweightedcaseloadof0.6.
Similarly,afull-
timesupervisorwithacaseloadof25,whoworks43h/week,wouldhaveaweightedcaseloadof0.6
3.Thisadjustmentmadethecaseloadscomparabledespitethenumberofhoursthesupervisorwo
rkedperweek.Asapartofmaintainingtheirpersonnelrecord,supervisorshadpreviouslyreporte
dthedatethattheyfirstbegantosuper-viseABA-
basedtreatmentforASD(includingtimesspentas342BehavAnalysisPractice(2016)9:339–348
asupervisoratothertreatmentagencies).Asupervisor’syearsofexperiencewascalculatedasthe
differencebetweentheirsupervisionstartdateandOctober1,2014.Supervisorcreden-
tialswerealsoobtainedthroughareviewofpersonnelfiles.Therewere130supervisorsrepresent
edinthedataset.Atotalof37supervisorswereexcludedbecausetheyheldanalternativecredentia
l(e.g.,licensedpsychologistsandli-censedclinicalsocialworker)orprovidedincompleteinfor-
mation,leaving93supervisorstoanalyze.Ofthe93supervi-
sors,67hadaBCBAcredentialand26didnot.Ofthe26supervisorswithoutaBCBAcredential,4rep
ortedabache-
lor’sdegreeastheirhighestlevelofeducationand22reportedamaster’sdegreeastheirhighestlev
elofeducation.The93supervisorshadanaverageof8.87(SD=4.71)yearsofexpe-
rienceinthefield,rangingfrom0.76to25.35andameancaseloadof11.18(SD=4.06),rangingfrom
2to23,asshowninFigs.1and2.ThesupervisorvariablesaresummarizedinTable1.DataAnalysisT
oexploretheroleofsupervisioninthemasteryoflearningobjectives,severallinearregressionanal
yseswerecarriedout.Linearregressionisastatisticaltechniqueformodelingthemathematicalrel
ationshipbetweenindependentvariablesanddependentvariables.Inthesimplecase,thisrelatio
nshipconsistsononlyoneindependentvariable,x,andonedepen-
dentvariable,y.Linearregressionisalsonamedbecausetheunderlyingassumptionofthemodelis
thatgivenavalueforx,thepredictedvalueofthedependentvariable,ŷcanbeex-
plainedwithasimpleline:^y¼mxþbIntheequationabove,theslopeoftheline,m,andtheintercept
oftheline,b,representtheregressionparameterstobelearnedgiventhesampledata.Whilemores
ophisticatedapproachesexist,themostbasictechniquefordeterminingthevalueoftheregressio
nparametersisthemethodofleastsquares.Thiscorrespondstominimizingthesumofsquareddiff
erencesbetweentheobservedvalueofyanditspredictedvalue,ŷ.Mathematicallythiscorrespond
stominimizingerror,E,whereEisdefinedas:E¼X^yiyi2forallobservationsdatapointsðÞ;iThesi
mplecaseofasingleindependentvariablecanbegeneralizedtoseveralindependentvariables,in
whichcasetheresultingmodelisreferredtoasamultiplelinearregres-
sionmodel.Foramorethoroughmathematicaltreatmentofregression,readersmayrefertoRoss(
2010).Becausethenumberoftreatmenthoursandmasteredlearn-
ingobjectivesnaturallyspanordersofmagnitude,alogarith-
mictransformwasappliedbeforefittingthelinearregressionmodel.Duringtheprocessofdataana
lysis,itisoftenthecasethatthevaluesofboththeindependentanddependentvari-
ablesspanoverseveralordersofmagnitude.Forexample,oneparticipantmayhaveonlymastered
fiveobjectivesinagivenperiodoftime,whileanotherparticipantmayhavemastered100.Whenitc
anbeverifiedthatthislargevarianceisalegit-
imatefacetofthedata,andnotdrivenbyoutliers,standardmathematicaltransformscanbeapplied
tothedatatoreducetheskewcausedbythisvariance,aswellasimprovethevisualandmathematic
alinterpretabilityofmodelsappliedtothedata.Acommondatatransformforthispurposeisthelog
a-rithmictransform,whichsimplyappliesthelogarithmfunc-
tiontovariablevalues.Thelogarithmfunctionisorderpre-
serving.Thisisimportanttonotebecauseorder-
preservingTable1SummaryofsupervisorvariablesVariableMean(SD)RangeYearsofexperienc
e8.87(4.71)0.76to25.35Caseload11.18(4.06)2to23CredentialBCBA=67Nocredential=26Year
s of
Experience0510152025051015Fig.1DistributionofsupervisoryearsofexperienceFrequency0
5101520051015Fig.2DistributionofsupervisorcaseloadsBehavAnalysisPractice(2016)9:339
–348343
transformsguaranteethatthenumericalrelationshipofvari-
ablesismaintained,whichmakesitanappropriatechoiceforthistask.ResultsAlinearregressiona
nalysisontreatmenthourswasfoundtobesignificantanddemonstratedastronglinearrelation-
ship(seeFig.3).TheresultingR-
squaredvalueof0.32,basedontheregressionmodel,suggeststhatover32%ofthevarianceinmast
eryoflearningobjectivesisaccountedforbythenumberoftreatmenthours.Thatistosaythat32%
ofthevarianceinthenumberofmasteredlearningobjectivescanbeaccountedforbytreatmentho
ursalone.Toaugmentthisanalysistoincludesupervisionhours,thesamelinearregressionwasre
peatedonlog-
transformeddata,thistime,capturingtherelationshipbetweensolelysupervisionhoursandmast
eredlearningobjectives.Figure4providesthescatterplotofthedata,alongwiththebest-
fitline.Inthiscase,theR-
squaredvalueofthemodeldropsto0.26,accountingforsubstantiallylessvar-
ianceinlearningobjectivesthantreatmenthours.Thissim-
pleexplorationofsupervisionhours,however,failstoac-
countforthefactthatbestpracticerecommendationssug-
gestadirectratioofsupervisionhourstotreatmenthours(e.g.,1–
2supervisionhoursforevery10hoftreatment),andthussupervisionandtreatmentintensitiesar
ehighlycorrelated.Toidentifythefullextentoftheimpactofsupervisiononmasteredlearningobje
ctives,amultiplelinearregressionwasperformed(againonlog-
transformeddata),usingbothtreat-
mentandsupervisionhoursastheindependentvariables,withmasteredlearningobjectivesasthe
dependentvariable.Table2showstheregressionparametersforthismodel,inadditiontothepara
metersforthesinglevariablemodel.TheR-
squaredforthemultipleregressionconsideringbothsupervisionandtreatmentincreasedto0.34,
accountingforlessthan2%morevariancethantreatmenthoursalone.Thisisperhapsbestdem-
onstratedvisuallyinFig.5,whichprovidesathree-
dimensionalscatterplotofthecorrespondingregressionmod-el.Here,theslopeofthetreatment-
masteredlearningobjec-tivelineissubstantiallyhigherthantheslopeofthesupervision-
masteredlearningobjectiveline.TofurtherunderstandtheroleofsupervisionintheefficacyofAB
A-
basedtreatment,thefollowingthreeattributesoftheABAsupervisorsrepresentedbythedataset
werestudied:whethertheyheldaBCBAcertification,theirnumberofyearsofexperience,andtheir
caseload.Theanalysiswasconductedusingstandardregressionmodels.Thenumberofyearsofex
periencewasfoundtobestatisticallysignificantwithapvalueof0.05.Additionally,whetherthesu
pervisorheldaBCBAcertificationprovedtobestatisticallysignificant,resultinginanFvalueof9.7
7forα=0.05.Table3providesregressioncoefficientsforthethreesupervisionattributesTable2Li
nearregressioncoefficientsforsupervisionandtreatmenthoursSupervisionTreatmentSupervi
sion+treatmentIntercept0.39−0.54−0.52Supervision0.95–0.38Treatment–
1.000.74R20.260.320.34Ftestp<0.000p<0.000p<0.0002.02.53.00.51.01.52.02.53.0Mastered
Learning
ObjectivesFig.3Treatmenthoursvs.masteredlearningobjectives(logtransformed)1.01.52.02.5
0.51.01.52.02.53.0Mastered Learning
ObjectivesFig.4Supervisionhoursvs.masteredlearningobjectives(logtransformed)344Behav
AnalysisPractice(2016)9:339–348
usingmasteredlearningobjectivesasthedependentvariable,withthepvalueforcredentialandex
perienceprovidingtheonlystatisticalsignificance.DiscussionThepresentstudyevaluatedtherel
ationshipbetweenmasteryoflearningobjectivesandelementsofABAsupervision.Whenevaluat
edalone,asignificantrelationshipwasidenti-
fiedbetweenthenumberofmasteredlearningobjectivesandthenumberofsupervisionhours.Ho
wever,supervisionhoursweregenerallyprovidedinadirectratiototreatmenthours(e.g.,1–
2hofsupervisionforevery10hoftreatment).Thus,participantswhoreceivedgreatersupervision
hoursinthepresentstudytypicallyreceivedhighertreatmenthoursaswell,whichhaspreviouslyb
eendemonstratedtohaveastrongimpactontreatmentoutcomes.Tomitigatetheimpactofthisco
nfound,amultipleregressionanalysiswasperformed.Whentreatmenthoursandsupervisionho
urswereanalyzedtogether,theadditionofsupervisionhoursimprovedthemodel’sabilitytoacco
untfortheobservedvariancebylessthan2%.Tobeclear,thisdoesnotimplythatsupervisionhours
havealowimpactonmasteryoflearningobjectives.Instead,theimplicationisthatthevariancewit
hintheboundsoftypicalsupervisionintensity(e.g.,2hofsupervisionforevery10hoftreatment;B
ACB,2014)resultsinarelativelysmallimprovementinmasteredlearningobjectives.Therelation
shipbetweenmasteryoflearningobjectivesandsupervisorcredentialswasexaminedintheprese
ntstudy.Asignificantcorrelationwasfound,revealingthatsupervisorswithBCBAcertificationsp
roduce73.7%greatermasteryoflearningobjectivesperhourascomparedtosupervisorswith-
outaBCBA.WhilethenumberofsupervisorswithBCBAcertificationshasgrownsinceLoveetal.(2
009)reportedthatlessthanhalfofthesurveyedsupervisorstohaveBCBAcer-
tifications,ascarcityofBCBAsremains.Althoughrequire-
mentsvarystatetostate,inmanystates,otherprofessionalsactingwithinthescopeoftheirlicensu
reareincludedamongthosewhomaysuperviseABAprograms.Inthepresentstudy,samplesofsup
ervisorsinthesegroupswereeitherabsentornotlargeenoughtoevaluateasstandalonegroups.Gi
venthelimited—albeitgrowing—numberofBCBAsandtherecog-
nitionbysomestatesandfundingsourcesofotherlicensedprofessionals,futureresearchshoulde
xaminetheeffective-
nessofsupervisorswithBCBAcertificationsascomparedtootherlicensedprofessionalspracticin
ginthefield.Asupervisor’syearsofexperienceoverseeingASDcaseswereshowntohaveasignifica
nteffectonthemasteryoflearningobjectives.Specifically,theanalysisindicatedthatforeveryyear
ofexperiencethatasupervisorhad,thenumberofmasteredlearningobjectivesincreasedby4%.T
hismaybetrivialwhenconsideringtheimpactofasingleyearbutwouldindicatethatcasesthatares
upervisedbypractitionerswith10yearsofsupervisoryexperiencearemastering40%moreperh
our.Thesedataindicatethatexperiencedpractitionersshouldconsiderwaystosharetheirknowl
edgeandskillsetwithless-experiencedcliniciansthroughmentorshipandcon-
sultation.Therelativelyweakstatisticalsignificanceoftheresultindicatesthoughthattherearesti
llnumerousotherfac-
torsthatimpactthenumberofmasteredlearningobjectives.Simplyhavingworkedasasupervisor
foralongperiodoftimedoesnotguaranteethatperformancewillbebetter.Itmaybethecasethatcli
niciansimproveovertimeduetoexperienc-
ingavarietyofdifferentcasesaswellascontinuingtotrainandhonetheirskills.Unfortunately,dete
rminingwhytherewasacorrelationbetweenyearsofexperienceandincreasednumberofmaster
edlearningobjectiveswasbeyondthescopeofthecurrentdata.Supervisorcaseloadswerenotfou
ndtohaveasignificantrelationshiptothenumberofmasteredlearningobjectives.Thisfindingwas
unexpected,butitisconsistentwiththerelativelyweakimpactthatanincreaseinsupervisionhour
shasonmasteredlearningobjectives.Itislikelythecasethat,onceasufficientlevelofsupervisionh
asbeenprovided,Table3RegressioncoefficientsforsupervisorattributesCredentialExperience
CasesIntercept0.3150.3530.512BCBA0.232––Experience–0.015–Weightedcases––
−0.053Adj.R20.0870.031−0.010Ftestp<0.002p<0.050p<0.7641.52.02.53.03.50.00.51.01.52.
02.53.03.50.51.01.52.02.53.0Treatment HoursMastered Learning
ObjectivesFig.5Jointeffectoftreatmentandsupervisionhoursonmasteredlearningobjectives(lo
gtransformed)BehavAnalysisPractice(2016)9:339–348345
increasingsupervisionofacasedoesnotimprovethenumberofmasteredlearningobjectives.Simi
larly,asupervisorwithasmallercaseloadwouldlikelybeabletoprovidemoresuper-
visiontoeachcase;asreflectedintheanalysisofsupervisionhours,however,theincreasedsupervi
sionhoursdidnotresultinameaningfulimprovement.Inthecurrentstudy,treatmentandsupervi
sionhourswerenotrandomlyassigned.Hourswerebasedonclin-
icalrecommendationsandsubjecttoauthorizationbydi-
versefundingagencies.Thus,thetreatmentandsupervi-
sionhoursthatanindividualreceivedmayhavebeenlessthanwhatwasconsideredmedicallynece
ssarybythetreatingclinician.Furthermore,theparticipantswerenotrandomlyassignedtosuper
visors.Caseassignmentwasbasedonavarietyoffactors,includingavailability,clinicalknowledge,
andfundingagencycredentialre-
quirements.Itstandstoreasonthatmorechallengingcasesmayhavebeenassignedtosupervisors
withgreaterexperience.Furthermore,supervisorsworkingonchal-
lengingcasesmayhavehadlightercaseloadsthansuper-
visorswithlessdemandingcases.Futureresearchshouldtakeintoaccountthesecase-
specificfactors.ASDisamultifaceteddisorderwitheachindividualdisplayinguniquesymptompr
esentationandtreatmentresponse.Thepresentstudydidnotaccountforindividualdiffer-
encesthatareknowntoaffecttreatmentresponse,in-
cludingage,symptomseverity,andskilllevel.Inaddi-
tion,anoteworthylimitationisthatwhileinterobserveragreement(IOA)iscollectedasapartofda
y-to-
dayclinicalpractice,thesedatawerenotstoredinsuchawayastobeaccessiblefortheseanalysis.Fu
turere-
searcherswoulddowelltobuildIOAintotheirdatatrackingsystems.Thepresentstudyexamineds
upervisionhoursoverall.Futureresearchonsupervisionintensityshouldinvestigatethefacetsof
supervisionthathavethegreatestimpactonmas-
teredlearningobjectivestohelpguideclinicalstandards.Forexample,supervisioninvolvesalarge
varietyoftasks,includ-
ingtreatmentplanning,parenttraining,directobservation,andtherapistmentoring,amongother
s.Itispossiblethatparticulartasksmayimproveoutcomesmorethanothers.Forinstance,prelimi
naryevidenceshowsthatsupportfromsupervisorspositivelyimpactstherapistself-
ratingsofperformanceandefficacy(Gibsonetal.,2009).Moreover,itispossiblethatgreatersuper
visionintensitymaybenefitsometreatmentdo-
mainsmorethanothers.Similarfindingshavebeenrevealedwithrespecttotreatmentintensity(V
irués-
Ortega,2010).Thepresentstudymeasuredtreatmentoutcomeintermsofmasteredskills.Whiles
tandardizedscalesaremorecommonlyusedtomeasureoutcomewithinASDtreatmentliterature
,themeasurementofacquiredtargetedskillsmaybettershowindividualizedprogressthatiscomp
arableacrossgroups(Matson&Goldin,2014).However,thismeasureofoutcomesisnotwithoutli
mita-
tions;thatis,degreeofdifficultyvariesforeachtargetedskill,andtargetedskillsdonotnecessarilya
ddresscoredeficitsofASD(Fava&Strauss,2014).Assuch,futureresearchemployingtargetmaste
ryasaprimaryoutcomemeasurecouldconsiderincludingonlythosetargetsdi-
rectlyassociatedwithdiagnosticcriteriaofASD,includ-
ingsocialcommunication,socialinteraction,andrestrict-
ed,repetitivebehaviors.Thefindingsreportedinthepresentstudyhavelargeimplications.Bestpr
acticerecommendationsforsupervi-
sionhavebeenmadebytheBACBtoprovideneededguidancetofundingagenciesandtofacilitatetr
eatmentintegrityandeffectiveness.Untilnow,littleresearchhasbeenconductedtosubstantiatet
hoserecommendations.Whiletheresultsofthecurrentstudyshouldbereplicatedinothersample
sandexploredfurther,theyindicatethatthe1–
2hperevery10hoftreatmentdescribedinthe2012versionoftheBACBguidelinesmaybemoreap-
propriatethantherevisedrecommendationsin2014.Further,giventherelationshipbetweentre
atmentresponseandthesupervisorcredential,itseemsevidentthatBACBstandardsforbehavior
analysishaveproducedameaning-fulcertification.Thesestandards,alongwiththeafore-
mentionedadditionaltraininginASDtreatment,maybethefactorsthatenabledsupervisorsinthe
presentstudytotakeongreatercaseloadsandwhy,withhoursofsupervi-
sionpercasethatreflectedthereduced2012BACBguidelines,supervisorswereabletomaintainst
rongclin-icaloutcomes.Giventhatthecurrentstudyfoundnore-
lationshipbetweenmasteredlearningobjectivesandsu-
pervisorcaseload,theoptimalcaseloadshouldbereconsidered.Potentially,supervisorcaseloads
maybecarefullyandincrementallyincreasedovertimetoexpandtreatmentcapacitywhileensuri
ngtreatmentqualityandintegrity.Anotherimplicationofthecurrentstudyisinregardtohowfund
ingresourcesareallocated.Inreal-worldset-
tings,treatmentresourcesarealwayslimited.Typically,consumersandprovidersalikeoftenmak
eharddecisionstotradeonetreatmentcomponentinfavorofanotherinanefforttoyieldthegreate
stimprovementforeachindivid-
ualwithASD.Giventherelationshipbetweentreatmenthoursandmasteredlearningobjectivesin
contrasttotherelationshipbetweensupervisionhoursandmasteredlearningobjectives,itseems
likelythatreallocatingfundingresourcesfromsupervisionhourstotreatmenthourswouldyieldb
etteroutcomesoverall(seeFig.5).Thatistosay,a10%increaseinsupervisionhourswouldyieldon
lya3.6%increaseinmasteredlearningobjec-
tives.Rather,ifthosesamehourswereallocatedtotreat-
ment,masteredlearningobjectiveswouldimproveby7.3%.Thiseffectisfurthermultipliedbythe
observation346BehavAnalysisPractice(2016)9:339–348
thatreimbursementratesareoftensignificantlyhigherforsupervisionthanfortreatment,meani
ngthatfundingfor1hofsupervisioncouldpotentiallyfund2–3hoftreat-
ment,whicharehoursthatresearchconsistentlydemon-
stratestoproducebetteroutcomesforeachchild.Whilesupervisionisrequiredtoensureprogress
andtreatmentintegrity,exactlyhowmuchsupervisionisrequiredisanempiricalquestion.Ackno
wledgmentsWewouldliketothankJulieKornackandJamesE.Carr,Ph.D.fortheirthoughtfulcom
mentsonearlydraftsofthismanu-
script.CompliancewithEthicalStandardsConflictofInterestTheauthorsdeclarethattheyhaveno
conflictofinterest.EthicalApprovalForthistypeofretrospectiveanalysis,formalcon-
sentwasnotrequired.ReferencesAmericanPsychiatricAssociation.(2000).Diagnosticandstatis
ticalmanualofmentaldisorders(4thed.).Washington:AmericanPsychiatricAssociation.Americ
anPsychiatricAssociation.(2013).Diagnosticandstatisticalmanualofmentaldisorders(5thed.).
Arlington:AmericanPsychiatricAssociation.BehaviorAnalystCertificationBoard,Inc.(2012).G
uidelines:healthplancoverageofappliedbehavioranalysistreatmentforautismspectrumdisord
er.Tallahassee:BACB.BehaviorAnalystCertificationBoard,Inc.(2014).Appliedbehavioranalysi
streatmentofautismspectrumdisorder:practiceguidelinesforhealthcarefundersandmanagers
(2nded.).Retrievedfromhttp://bacb.com/wp-
content/uploads/2015/07/ABA_Guidelines_for_ASD.pdf.Ben-
Itzchak,E.,&Zachor,D.A.(2007).Theeffectsofintellectualfunc-
tioningandautismseverityonoutcomeofearlybehavioralinterven-
tionforchildrenwithautism.ResearchinDevelopmentalDisabilities,28(3),287–303.Ben-
Itzchak,E.,&Zachor,D.A.(2011).Whobenefitsfromearlyinter-
ventioninautismspectrumdisorders?ResearchinAutismSpectrumDisorders,5(1),345–
350.Bibby,P.,Eikeseth,S.,Martin,N.T.,Mudford,O.C.,&Reeves,D.(2002).Progressandoutcomesf
orchildrenwithautismreceivingparent-
managedintensiveinterventions.ResearchinDevelopmentalDisabilities,23(1),81–
104.Dixon,D.R.,Tarbox,J.,Najdowski,A.C.,Wilke,A.E.,&Granpeesheh,D.(2011).Acomprehensive
evaluationoflanguageforearlybehav-
ioralinterventionprograms:thereliabilityoftheSKILLSLanguageIndex.ResearchinAutismSpec
trumDisorders,5(1),506–
511.Eikeseth,S.(2010).ExaminationofqualificationsrequiredofanEIBIprofessional.EuropeanJ
ournalofBehaviorAnalysis,11(2),239–
246.Eikeseth,S.,Hayward,D.,Gale,C.,Gitlesen,J.P.,&Eldevik,S.(2009).Intensityofsupervisionand
outcomeforpreschoolagedchildrenreceivingearlyandintensivebehavioralinterventions:aprel
iminarystudy.ResearchinAutismSpectrumDisorders,3(1),67–
73.Eikeseth,S.,Smith,T.,Jahr,E.,&Eldevik,S.(2002).Intensivebehav-
ioraltreatmentatschoolfor4-to7-year-oldchildrenwithautism:a1-
yearcomparisoncontrolledstudy.BehaviorModification,26(1),49–
68.Eikeseth,S.,Smith,T.,Jahr,E.,&Eldevik,S.(2007).Outcomeforchildrenwithautismwhobegani
ntensivebehavioraltreatmentbe-
tweenages4and7:acomparisoncontrolledstudy.BehaviorModification,31(3),264–
278.Eldevik,S.,Eikeseth,S.,Jahr,E.,&Smith,T.(2006).Effectsoflow-
intensitybehavioraltreatmentforchildrenwithautismandmentalretardation.JournalofAutism
andDevelopmentalDisorders,36(2),211–
224.Eldevik,S.,Hastings,R.P.,Hughes,J.C.,Jahr,E.,Eikeseth,S.,&Cross,S.(2009).Meta-
analysisofearlyintensivebehavioralinterventionforchildrenwithautism.JournalofClinicalChil
dandAdolescentPsychology,38(3),439–
450.Eldevik,S.,Hastings,R.P.,Hughes,J.C.,Jahr,E.,Eikeseth,S.,&Cross,S.(2010).Usingparticipant
datatoextendtheevidencebaseforintensivebehavioralinterventionforchildrenwithautism.Am
ericanJournalonIntellectualandDevelopmentalDisabilities,115(5),381–
405.Eldevik,S.,Hastings,R.P.,Jahr,E.,&Hughes,J.C.(2012).Outcomesofbehavioralinterventionfo
rchildrenwithautisminmainstreampre-
schoolsettings.JournalofAutismandDevelopmentalDisorders,42(2),210–
220.Fava,L.,&Strauss,K.(2014).Responsetoearlyintensivebehavioralinterventionforautism—
anumbrellaapproachtoissuescriticaltotreatmentindividualization.InternationalJournalofDev
elopmentalNeuroscience,39,49–
58.Flanagan,H.E.,Perry,A.,&Freeman,N.L.(2012).Effectivenessoflarge-scalecommunity-
basedintensivebehavioralintervention:awaitlistcomparisonstudyexploringoutcomesandpre
dictors.ResearchinAutismSpectrumDisorders,6(2),673–
682.Gibson,J.A.,Grey,I.M.,&Hastings,R.P.(2009).Supervisorsupportasapredictorofburnoutan
dtherapeuticself-
efficacyintherapistsworkinginABAschools.JournalofAutismandDevelopmentalDisorders,39(
7),1024–
1030.Granpeesheh,D.,Dixon,D.R.,Tarbox,J.,Kaplan,A.M.,&Wilke,A.E.(2009).Theeffectsofagean
dtreatmentintensityonbehavioralinterventionoutcomesforchildrenwithautismspectrumdiso
rders.ResearchinAutismSpectrumDisorders,3(4),1014–
1022.Granpeesheh,D.,Tarbox,J.,Najdowski,A.C.,&Kornack,J.(Eds.).(2014).Evidence-
basedtreatmentforchildrenwithautism:theCARDmodel.Waltham:Elsevier.Harris,S.L.,&Handl
eman,J.S.(2000).AgeandIQatintakeaspredic-
torsofplacementforyoungchildrenwithautism:afour-tosix-yearfollow-
up.JournalofAutismandDevelopmentalDisorders,30(2),137–
142.Hayward,D.,Eikeseth,S.,Gale,C.,&Morgan,S.(2009).Assessingprogressduringtreatmentfor
youngchildrenwithautismreceivingintensivebehavioralinterventions.Autism,13(6),613–
633.Howlin,P.,Magiati,I.,&Charman,T.(2009).Systematicreviewofearlyintensivebehavioralint
erventionsforchildrenwithautism.AmericanJournalonIntellectualandDevelopmentalDisabilit
ies,114(1),23–
41.Lovaas,O.I.(1987).Behavioraltreatmentandnormaleducationalandintellectualfunctioningi
nyoungautisticchildren.JournalofConsultingandClinicalPsychology,55(1),3–
9.Love,J.R.,Carr,J.E.,Almason,S.M.,&Petursdottir,A.I.(2009).Earlyandintensivebehavioralinter
ventionforautism:asurveyofclinicalpractices.ResearchinAutismSpectrumDisorders,3(2),421
–428.Luiselli,J.K.,O’malleyCannon,B.,Ellis,J.T.,&Sisson,R.W.(2000).Home-
basedbehavioralinterventionforyoungchildrenwithautism/pervasivedevelopmentaldisorde
r:apreliminaryevaluationofoutcomeinrelationtochildageandintensityofservicedelivery.Autis
m,4(4),426–438.BehavAnalysisPractice(2016)9:339–348347
Magiati,I.,Charman,T.,&Howlin,P.(2007).Atwo-yearprospectivefollow-upstudyofcommunity-
basedearlyintensivebehaviouralinterventionandspecialistnurseryprovisionforchildrenwith
au-tismspectrumdisorders.JournalofChildPsychologyandPsychiatry,48(8),803–
812.Magiati,I.,Moss,J.,Charman,T.,&Howlin,P.(2011).Patternsofchangeinchildrenwithautism
spectrumdisorderswhoreceivedcommunitybasedcomprehensiveinterventionsintheirpre-
schoolyears:asevenyearfollow-upstudy.ResearchinAutismSpectrumDisorders,5(3),1016–
1027.Makrygianni,M.K.,&Reed,P.(2010).Ameta-
analyticreviewoftheeffectivenessofbehaviouralearlyinterventionprogramsforchil-
drenwithautismspectrumdisorders.ResearchinAutismSpectrumDisorders,4(4),577–
593.Matson,J.L.,&Goldin,R.L.(2014).Earlyintensivebehavioralinterven-
tions:selectingbehaviorsfortreatmentandassessingtreatmenteffec-
tiveness.ResearchinAutismSpectrumDisorders,8(2),138–
142.Mudford,O.C.,Martin,N.T.,Eikeseth,S.,&Bibby,P.(2001).Parent-
managedbehavioraltreatmentforpreschoolchildrenwithautism:somecharacteristicsofUKpro
grams.ResearchinDevelopmentalDisabilities,22(3),173–
182.Perry,A.,Cummings,A.,Geier,J.D.,Freeman,N.L.,Hughes,S.,Managhan,T.,etal.(2011).Predict
orsofoutcomeforchildrenreceivingintensivebehavioralinterventioninalarge,community-
basedprogram.ResearchinAutismSpectrumDisorders,5(1),592–
603.Persicke,A.,Bishop,M.R.,Coffman,C.M.,Najdowski,A.C.,Tarbox,J.,Chi,K.,etal.(2014).Evaluat
ionoftheconcurrentvalidityofaskillsassessmentforautismtreatment.ResearchinAutismSpectr
umDisorders,8(3),281–285.Reichow,B.(2012).Overviewofmeta-
analysesonearlyintensivebe-havioralinterventionforyoungchildrenwithautismspectrumdis-
orders.JournalofAutismandDevelopmentalDisorders,42(4),512–
520.Reichow,B.,Barton,E.E.,Boyd,B.A.,&Hume,K.(2012).Earlyinten-
sivebehavioralintervention(EIBI)foryoungchildrenwithautismspectrumdisorders(ASD).Coc
hraneDatabaseSystematicReviews,10.Reichow,B.,&Wolery,M.(2009).Comprehensivesynthe
sisofearlyintensivebehavioralinterventionsforyoungchildrenwithautismbasedontheUCLAyo
ungautismprojectmodel.JournalofAutismandDevelopmentalDisorders,39(1),23–
41.Remington,B.,Hastings,R.P.,Kovshoff,H.,degliEspinosa,F.,Jahr,E.,Brown,T.,etal.(2007).Earl
yintensivebehavioralintervention:outcomesforchildrenwithautismandtheirparentsaftertwo
years.AmericanJournalonMentalRetardation,112(6),418–
438.Ross,S.(2010).Introductorystatistics(3rded.).Burlington:Academic.Sallows,G.O.,&Graup
ner,T.D.(2005).Intensivebehavioraltreatmentforchildrenwithautism:four-
yearoutcomeandpredictors.AmericanJournalonMentalRetardation,110(6),417–
438.Smith,T.,Groen,A.D.,&Wynn,J.W.(2000).Randomizedtrialofintensiveearlyinterventionfor
childrenwithpervasivedevelopmen-
taldisorder.AmericanJournalonMentalRetardation,105(4),269–
285.Smith,I.M.,Koegel,R.L.,Koegel,L.K.,Openden,D.A.,Fossum,K.L.,&Bryson,S.E.(2010).Effectiv
enessofanovelcommunity-
basedearlyinterventionmodelforchildrenwithautisticspectrumdisorder.AmericanJournalonI
ntellectualandDevelopmentalDisabilities,115(6),504–523.Virués-
Ortega,J.(2010).Appliedbehavioranalyticinterventionforau-tisminearlychildhood:meta-
analysis,meta-regressionanddose–responsemeta-
analysisofmultipleoutcomes.ClinicalPsychologyReview,30(4),387–399.Virués-
Ortega,J.,Rodríguez,V.,&Yu,C.T.(2013).Predictionoftreat-
mentoutcomesandlongitudinalanalysisinchildrenwithautismundergoingintensivebehavioral
intervention.InternationalJournalofClinicalandHealthPsychology,13(2),91–
100.348BehavAnalysisPractice(2016)9:339–348

supervision goals ABA.pdf