ERIC EJ808980: A Social Stories[TM] Intervention Package with Students with Autism in Inclusive Classroom Settings

2008·0.14 MB·English

by  ERIC

JOURNALOFAPPLIEDBEHAVIORANALYSIS 2008, 41, 405–409 NUMBER3 (FALL2008)
A SOCIAL STORIESTM INTERVENTION PACKAGE FOR STUDENTS
WITH AUTISM IN INCLUSIVE CLASSROOM SETTINGS
JEFFREY M. CHAN AND MARK F. O’REILLY
UNIVERSITYOFTEXASATAUSTIN
ASocialStoriesTMinterventionpackagewasusedtoteach2studentswithautismtoreadSocial
Stories,answercomprehensionquestions,andengageinroleplays.Appropriatesocialbehaviors
increased and inappropriate behaviors decreased for both participants, and the effects were
maintained for up to 10 months. This intervention package appears to be useful in inclusive
classroomenvironments anddoes notrequireintensive supervision ofthe child’sbehavior.
DESCRIPTORS: autism,inclusionclassroom,roleplay,socialskillstraining,SocialStories
_______________________________________________________________________________
Social StoriesTM (Gray & Garand, 1993) are general education inclusion settings. Including
afrequentlyusedtreatmentmethodforchildren students with autism in general education
with autism; however, the research literature on classrooms can create meaningful social oppor-
their use includes only a small number of well- tunities with peers without disabilities (Fryxell
controlled experimental studies (e.g., Green et &Kennedy,1995)andisconsideredapreferred
al., 2006). Studies have often combined Social alternativetoself-containededucationalsettings
Storieswithotherinstructionalmethodssuchas (Individuals with Disabilities Education Act,
prompting (e.g., Swaggart et al., 1995), rein- 1997). Therefore, the purpose of the current
forcement (e.g., Swaggart et al.), and self- study was to examine the use of a Social Stories
evaluation (e.g., Thiemann & Goldstein, intervention package on the social communica-
2001), making it difficult to determine the tion behaviors of 2 students with autism
unique or additive effects of intervention enrolled in full-inclusion kindergarten class-
components. However, Reynhout and Carter rooms.
(2006) found little difference in effect sizes
across studies using Social Stories alone and METHOD
thoseusingSocialStoriesinterventionpackages.
Participants, Setting, and Target Behaviors
Social Stories have been experimentally
Matt, a 6-year-old Asian American boy, and
investigated in home settings (Lorimer, Simp-
Ted, a 5-year-old Caucasian boy, attended two
son, Myles, & Ganz, 2002), self-contained
different general education kindergarten class-
special education/resource classrooms (Delano
roomsfull-timeatapublicelementaryschool.A
& Snell, 2006), and campus outdoor areas
teaching assistant was assigned to Matt’s
(Sansosti & Powell-Smith, 2006) but not in
classroom, and Ted had a teaching assistant
whowasassignedtoaccompanyhimduringthe
This study was completed in partial fulfillment of the
entire school day. Both participants had been
requirements of the MA degree from the Department of
diagnosed with autism and had language skills
SpecialEducationoftheUniversityofTexasatAustin.We
thank the parents, administrators, and teachers of the that were appropriate for the Social Stories
participants for their cooperation. The first author also intervention (standardized assessment scores are
thanksBenSmith,MissyOlive,LeeKern,LindaLeBlanc,
available from the first author).
LauraPeroutka, Eric Larsson,andNicole Murillo.
Address correspondence to Jeffrey M. Chan, Depart- Training sessions took place in the morning
mentofSpecialEducation,UniversityofTexasatAustin, before the start of the school day in private
1 University Station, D5300, Austin, Texas 78712 (e-
rooms at the students’ school. Hour-long
mail:[email protected]).
doi:10.1901/jaba.2008.41-405 observation sessions were conducted in the
405
406 JEFFREY M. CHAN and MARK F. O’REILLY
participants’ classrooms during circle time and one to three stories per day). There was one
centers time on the same day as training story per target behavior for each participant. A
sessions. Matt’s target behaviors were inappro- total of six stories were used in this study.
priate social interactions (e.g., standing or Stories were written using Gray’s (1995)
sitting within a few centimeters of peers, guidelines. The text was presented on white
causing peers to lean or move away), appropri- paperwithnopictures,andthefontsizewas22
ate hand raising (e.g., above shoulder and to 26 points (i.e., the largest font that allowed
vertical rather than horizontal extension), and each story to fit on a single page). The
inappropriate vocalizations(e.g., monosyllables, interventionconsistedofthreesteps:(a)reading
noises, comments irrelevant to classroom activ- the story, (b) answering comprehension ques-
ities). Target behaviors for Ted were appropri- tions, and (c) role play.These steps were always
ate hand raising, appropriate social initiations presented in this order.
(e.g., approaching peers and asking to play), First, the participant was instructed to read
andinappropriatevocalizations(e.g.,comments the story aloud or to choose another reading
irrelevant to classroom activities). Data were option, such as having the instructor read the
collected using a frequency count for appropri-
story aloud or to read the story silently. Matt
ate social initiations, inappropriate social inter-
always read the story aloud, and Ted read the
actions, and inappropriate vocalizations. Dur-
storyaloudexceptforthreesessionsinwhichhe
ingeachobservationsession,datawerecollected
chose to read silently.
on one to three behaviors per participant.
Next, the instructor asked three questions
Measurement of hand raising was based on
about the story to test for comprehension (e.g.,
the number of opportunities presented during
‘‘What can I try to do when my teacher asks
circle time (i.e., questions posed by the teacher)
questions?’’). Three questions were written for
and is presented as the percentage of opportu-
each story and were printed out as a reference
nities resulting in appropriate hand raising.
for the instructor, but the instructor was free to
alter the order of the questions and general
Design and Procedure
wording. If the student was unable to answer
A multiple probe design across behaviors was
the question correctly or did not provide an
used.
answer, they were prompted to reread the
Baseline.Duringbaseline,data werecollected
portion of the story that pertained to the
on the target behaviors during hour-long
question.
observation sessions described above. Estab-
The final step was a role play of the social
lishedclassroom rules and discipline procedures
situation described in the story. First, the
remained in effect.
instructor introduced the role play by verbally
Social Stories intervention package. Partici-
describing the situation and target behavior.
pants received intervention on an individual
Next, the instructor, participant, and another
basis.Interventionsessionsoccurredonetofour
adult acted out the situation, which ended with
times per week. Matt’s intervention lasted for 5
the performance of the target behavior. For
weeksandconsisted of13intervention sessions.
example, for hand raising, the instructor said,
Ted received intervention for 18 sessions over
10weeks. Sessions occurred priorto the start of ‘‘Let’s pretend we’re in circle time. I’ll be the
the school day and lasted 10 to 20 min, teacher and will read this story to you. When I
depending on the number of stories read and askaquestion,don’tforgettoraiseyourhand.’’
the number of behaviors practiced. Participants The instructor then read a short passage from a
read only stories that pertained to the behaviors bookandaskedaquestion(i.e.,teacherrole).At
targeted for intervention on a given day (i.e., this point, the participant could raise his hand
SOCIAL STORIESTM INTERVENTION 407
and wait to be called on to give an answer. The play. Treatment integrity data were collected
other adult sometimes acted as a student, during 39% of treatment sessions, and the
listened to the story, and occasionally raised percentage of intervention steps implemented
his or her hand in response to the teacher’s correctly was 96% (range, 83% to 100%).
question.Inotherscenarios,theadultactedasa Finally, classroom teachers, teaching assistants,
peer with whom the participant interacted. and parents were asked to rate statements
Verbalpromptsweredeliverediftheparticipant regarding the social validity (i.e., importance
haddifficultycompletinganyofthethreesteps, of skills taught, perceived effectiveness, appro-
and praise was delivered after successful com- priateness, future use) of the intervention
pletion of each of the three steps. packageonaLikert-typescalefrom1(negative)
Follow-up. Follow-up probes were conducted to 5 (positive). Mean rating of social validity
with Matt at 1, 3, 5, and 10 months after statements was 4.3 (range, 3 to 5 across
completion of the intervention. Follow-up questions).
probes were conducted with Ted at 2 and 7
months.
RESULTS AND DISCUSSION
InterobserverAgreement,TreatmentIntegrity,and Figure 1 shows the frequency and percent-
Social Validity age of opportunities for the three target
All data collectors were trained graduate behaviors for Matt and Ted during baseline,
students. Two data collectors independently intervention, and follow-up phases. For Matt,
collected data for 32% of sessions. Agreement introduction of the Social Stories intervention
was calculated for frequency measures by package resulted in (a) an immediate decrease
dividing the lower frequency of a target in inappropriate social interaction behavior,
behavior by the higher frequency and convert- (b) a progressive increase in hand raising, and
ing this ratio to a percentage. Interobserver (c) a reduction of his inappropriate vocaliza-
agreement for the percentage of opportunities tions. For Ted, implementation of the inter-
measure was calculated by dividing the lower vention package led to (a) an increase in hand
percentage-of-opportunities observation score raising to higher and variable levels and (b) an
by the higher percentage of opportunities score increase in appropriate social initiations. Ted’s
and converting this ratio to a percentage. inappropriate vocalizations decreased steadily
Agreements for Matt’s target behaviors were throughout baseline, so the intervention was
95%(handraising),100%(inappropriatesocial never introduced. Follow-up data for both
interaction), and 88% (inappropriate vocaliza- participants indicated that positive behavior
tions). One day’s data were removed from changes were maintained over time, with the
Matt’s data set due to low interobserver final maintenance session conducted during a
agreement. For Ted, agreements were 94% new school year in which both participants
(hand raising), 100% (social initiations), and had been promoted to the first grade and had
92% (inappropriate vocalizations). To measure new teachers.
treatment integrity, a checklist was created that The results of the current study extend
listed the steps of the intervention. The previous studies on Social Stories by extending
checklist consisted of (a) reading the story, (b) the findings to inclusive classrooms and target-
asking three comprehension questions, (c) ing appropriate classroom behaviors. The
engaging in a role play of the target behavior, intervention did not require intensive supervi-
and (d) delivering prompts or positive rein- sion of the child’s behavior or substantial time
forcement following each comprehension ques- for implementation, making it suitable for
tion and at appropriate times during the role regular education classroom settings. Although
408 JEFFREY M. CHAN and MARK F. O’REILLY
Figure1. Frequencyandpercentageofopportunitiesoftargetbehaviorsduringbaselineandinterventionphasesfor
MattandTed during 60-min observation sessions.
no comparative data of classmates’ behavior are For example, most students raised their hand
available, anecdotal evidence suggests that the regularly whentheteacheraskeda questionand
participants performed similarly to their peers made a few social initiations during each circle
following the introduction of the intervention. time or center time.
SOCIAL STORIESTM INTERVENTION 409
The design of this study and the multicom- REFERENCES
ponentnatureoftheinterventiondonotpermit
Delano, M., & Snell, M. E. (2006). The effects of social
conclusions about the behavioral mechanisms stories on the social engagement of children with
that account for the effects of the intervention autism. Journal of Positive Behavior Interventions, 8,
29–42.
package or the importance of the stories
DiGennaro, F. D., Martens, B. K., & Kleinmann, A. E.
themselves in changing behavior. It is possible (2007). A comparison of performance feedback
thatmodelingandroleplaycouldhaveachieved procedures on teachers’ treatment implementation
integrity and students’ inappropriate behavior in
the same effects without the stories. The
special education classrooms. Journal of Applied
advanced language capabilities of the partici-
Behavior Analysis,40,447–461.
pants in this study leave open the possibility Fryxell, D., & Kennedy, C. H. (1995). Placement along
that rule governance may account for the thecontinuumofservicesanditsimpactonstudents’
social relationships. Journal of the Association for
treatment effects (Hayes, Barnes-Holmes, &
Persons withSevere Handicaps,20(4), 259–269.
Roche, 2001). If so, students without such Gray,C.A.(1995).Teachingchildrenwithautismtoread
verbal abilities may not benefit as readily from social situations. In K. A. Quill (Ed.), Teaching
childrenwithautism(pp.219–241).NewYork:Delmar.
Social Stories or might require additional
Gray, C. A., & Garand, J. D. (1993). Social stories:
intervention components to achieve effects via
Improving responses of students with autism with
different behavioral mechanisms (e.g., direct accurate social information. Focus on Autistic Behav-
contingencies in the target setting). ior,8,1–10.
Green,V.A.,Pituch,K.A.,Itchon,J.,Choi,A.,O’Reilly,
Several future studies are needed to fully
M., & Sigafoos, J. (2006). Internet survey of
understand the impact of this popular educa- treatments used by parents of children with autism.
tional intervention. Additional studies should Research in Developmental Disabilities,27, 70–84.
Hayes, S. C., Barnes-Holmes, D., & Roche, B. (Eds.).
be conducted in inclusive settings and should
(2001). Relational frame theory: A post-Skinnerian
attempt to analyze the effects of the different account of human language and cognition. New York:
intervention components by initially introduc- Kluwer Academic/Plenum.
ing Social Stories in isolation and introducing Individuals with Disabilities Education Act. (1997). 20
U.S.C. 1401et seq.
other components if needed. In addition,
Lorimer,P.A.,Simpson,R.L.,Myles,B.S.,&Ganz,J.
group-delivered Social Stories could target all B. (2002). The use of social stories as a preventative
students without singling out an individual behavioralinterventioninahomesettingwithachild
withautism.JournalofPositiveBehaviorInterventions,
child for special treatment. Increased teacher
4,53–60.
involvement is another possible target for Reynhout,G.,&Carter,M.(2006).SocialStoriesTMfor
research, with classroom teachers playing a role children with disabilities. Journal of Autism and
Developmental Disorders, 36,445–469.
in the design and implementation of Social
Sansosti,F.J.,&Powell-Smith,K.A.(2006).Usingsocial
Stories studies (DiGennaro, Martens, & Klein-
storiestoimprovethesocialbehaviorofchildrenwith
mann,2007).Futurestudiesshouldalsoaddress Asperger syndrome. Journal of Positive Behavior
some of themeasurement limitationsevident in Interventions, 8,43–57.
Swaggart, B., Gangon, E., Bock, S. J., Earles, T. L.,
this experiment. For example, rate measures are
Quinn, C., Myles, B. S., et al. (1995). Using social
preferable to the simple frequency counts storiestoteachsocialandbehavioralskillstochildren
reported here, and the ratio methods for with autism.Focus onAutisticBehavior,10, 1–16.
Thiemann, K. S., Goldstein, H. (2001). Social stories,
measurement of interobserver agreement pro-
writtentextcues,andvideofeedback:Effectsonsocial
vide a liberal estimate of agreement compared
communication of children with autism. Journal of
to point-by-point methods. These issues are Applied BehaviorAnalysis,34,425–446.
especially pertinent in light of the fact that data
Received March 21,2007
were discarded from this study because of low
Final acceptanceAugust 10,2007
agreement. Action Editor,Linda LeBlanc