Socia l Skill Trainin g in an Integrate d Preschoo l Progra m · Socia l Skill Trainin g in an...

School Psychology Quarterly, Vol. 16, No. 2, 2001, pp. 158-175

Social Skill Training in anIntegrated Preschool Program

Hindi M. Guglielmo and Georgiana Shick TryonCity University of New York Graduate School and University Center

This study examined the effectiveness of a commercially available social skills trainingprogram plus classroom reinforcement for use with preschoolers with developmental de-lays. Two groups of 19 participants each received either the combined treatment packageor classroom reinforcement of target behaviors only. An additional 20 participants servedas a control group. The combination of social skills training plus classroom reinforcementof target behaviors resulted in statistically significant increases in sharing behavior overthose obtained by both control participants and participants receiving reinforcement ofclassroom behavior alone. Group behavior was increased over that shown by control par-ticipants through the use of either the social skills plus classroom reinforcement treatmentpackage or classroom reinforcement alone. Social skills interventions were viewed favor-ably by both classroom teachers and preschool participants.

The introduction of P.L. 94-142 in 1975, most recently reauthorized as P.L. 105-17, Individuals with Disabilities Education Act Amendments of 1997 (IDEA1997), mandates a free, appropriate public education in the least restrictive envi-ronment for all children with disabilities. Children with developmental delaysmeet criteria in one of 13 categories, including a noncategorical, developmentaldelay classification for preschool children, Preschool Student with a Disability.According to New York State, where this study was conducted, these children"exhibit a significant delay or disorder in one or more functional areas related tocognitive, language and communication, adaptive, social-emotional, or motordevelopment which adversely affect the students' ability to learn" (New YorkState Education Department, 1993, p. 3).

Since 1975, the continuum of educational options for students with specialneeds has moved from segregated settings to less restrictive placements in inte-grated, mainstream settings with nondisabled peers. In 1992 nationwide, it wasestimated that 34% of students with developmental delays were being educatedin mainstream settings (Bronson, Hauser-Cram, & Warfield, 1995). Studies con-

Address correspondence to Hindi M. Guglielmo, Interdisciplinary Center for Child Development,

35-55 233rd Street, Bayside, New York 11361; E-mail: [email protected].

158

SOCIAL SKILLS TRAINING 159

ducted over the past 15 years have reported no ill effects for students, whether ingeneral or special education, who were educated in integrated settings. A synthe-sis of two meta-analyses showed small to moderate educational and social bene-fits for students with disabilities in integrated settings, compared with students insegregated settings (Baker, Wang, & Walberg, 1995; Madden & Slavin, 1983).

An underlying premise of inclusion is the presumed acquisition of age-appro-priate skills by children with developmental delays through imitation of typicallydeveloping peers. This assumption, however, has come under scrutiny, particu-larly as it applies to the acquisition of social skills for youngsters with develop-mental delays. Although there is some evidence for small gains in the social out-comes for students with disabilities (Baker et al., 1995), Gresham (1982) andothers (Odom & Strain, 1984) have argued that placement alone is inadequate toensure the acquisition of appropriate social skills. Students with developmentaldelays frequently lag behind their nondisabled peers in the development of socialskills and are often deficient in the prerequisite skills needed to model the age-appropriate social behaviors that their nondisabled peers exhibit (Guralnick &Weinhouse, 1984; Odom & McEvoy, 1988).

According to results of a survey by Odom, McConnell, and Chandler (1993),most early childhood special education teachers reported a great need for socialinterventions for their students but indicated that few formal curricula wereavailable and that fewer have been tested empirically in comparison to other in-terventions. They surveyed teachers on the acceptability, feasibility, and currentuse of social competence training and found that most early childhood specialeducation teachers considered, on average, 74% of their students to be in need ofsocial competence intervention. Some of the most commonly endorsed interven-tions considered to be feasible for the classroom included teaching social skillsto children and praising individuals and groups for the use of social skills.

Curriculum-based social skills training can be economical and efficient. It canaddress the social delays of several children simultaneously within the context ofthe classroom, the childrens' natural environment (McAllister, 1991; Odom &Strain, 1984). Social skills training in the classroom is compatible with the inclu-sive model of educating children with developmental delays in the least restric-tive environment. A social skills curriculum can be embedded into the classroomand implemented by either regular or special education teachers. Thus, this re-search examined a short-term, commercially available group intervention basedon lessons from Taking Part, Introducing Social Skills to Children (Cartledge &Kleefeld, 1991) to address the need for a social skills program for children withdevelopmental delays.

Numerous studies have found deficits for social interactions (Bronson et al.,1995; Faught, Balleweg, Crow, & Van Den Pol, 1983; Hanline, 1985) and peer-entry skills (Guralnick, 1992) for preschool students with disabilities. Two targetbehaviors, sharing and being in a group, were selected for study after a review ofthe literature showed that these behaviors are often deficient in children with de-velopmental delays (Bronson et al., 1995; LeBlanc & Matson, 1995; Faught et

160 GUGLIELMO AND TRYON

al., 1983). Further, McConnell (1987) discussed the issue of entrapment effectsthat allow behavior to come under the control of natural consequences in relationto social skills training. He suggested including behaviors for training, such assharing, that would create a natural reinforcement loop. Thus, two lessons fromthe Taking Part "Cooperating with Peers" unit (Cartledge & Kleefeld, 1991) thataddress sharing and being in a group formed the basis for skills training in thisstudy. These skills are prerequisites for more sophisticated social interactions;once they occur, they can be reinforced easily by natural social consequences inthe classroom (Gresham & Elliott, 1990).

This study compared behavioral observations of sharing and being in a groupfor three groups of preschoolers with developmental delays. Group A receivedsocial skills intervention plus classroom reinforcement of target behaviors.Group B received classroom reinforcement of target behaviors only. Group Cdid not receive the treatments given to Groups A and B and served as a control.It was hypothesized that students receiving the combination of social skills train-ing and classroom reinforcement of target behaviors would display more targetbehaviors during free play than students receiving classroom reinforcementalone or students in the control group. It was also hypothesized that students re-ceiving classroom reinforcement alone would display more target behaviors thanstudents in the control group.

METHOD

Participants, Setting, and Assignment to Groups

Participants were 58 children with developmental delays attending integratedclasses in a publicly funded, privately operated preschool with locations inQueens and Nassau Counties in New York State. The school is accredited by theNational Association for the Education of Young Children (NAEYC) and fol-lows Developmentally Appropriate Practices (DAP) guidelines formulated byNAEYC. DAP guidelines call for teacher decisions to be based on three criteria:"1) what teachers know about how children develop and learn; 2) what teachersknow about individual children in their group; [and] 3) knowledge of the socialand cultural context in which those children live and learn" (Bredekamp & Cop-pie, 1997, p. vii).

The study was conducted during a 6-week summer session. The school en-rolled a total of 161 students in the 1999 summer session, 69 of these partici-pants were students with developmental delays who were enrolled in integratedclasses in large classrooms with typically developing peers. Parents of the 69students were sent consent forms for participation in the study, 2 refused partici-pation, 8 did not respond, and 1 withdrew after the first week of the study. Thus,the students included in the study represented 84% of the preschoolers with de-velopmental delays in integrated classrooms in the school.


Before entering the preschool, study participants met eligibility for classifica-tion as a Preschool Student with a Disability (PSWD) based on New York Stateguidelines. Eligibility for special education is determined by performance on acomprehensive assessment that includes standardized testing. Preschool-agedchildren qualify as developmentally delayed and are classified as a PSWD ifthey meet criteria in one of several ways. New York State guidelines specify thata PSWD must exhibit a "significant" cognitive, language and communication,social-emotional, or motor development delay. A significant delay is "a 12-month delay in one or more functional area(s), or, a 33 percent delay in one func-tional area, or a 25 percent delay in each of two functional areas" (New YorkState Education Department, 1993, p. 5). On individually administered standard-ized tests, students' scores must be two SD below the mean in one functionalarea or 1.5 .SO below the mean in each of two functional areas (New York StateEducation Department, 1993). Students with developmental delays were identi-fied using the following tests: Stanford-Binet Intelligence Scale, 4th edition(SBIS; Thorndike, Hagen, & Sattler, 1986), Bayley Scales of Infant Develop-ment, 2nd edition (BSID; Bayley, 1993), and the Preschool Language Scale(PLS; Zimmerman, Steiner, & Pond, 1991).

In New York City, preschool students who are eligible for special educationmay be recommended for an integrated classroom. Because the school where thestudy was conducted had both integrated and segregated classrooms, only stu-dents whose parents had accepted a recommendation to place their child or chil-dren in an integrated setting were solicited for the study.

Once a student met eligibility for preschool special education, was recom-mended for an integrated class, and enrolled in the host preschool, school admin-istrators randomly assigned the student to an available seat in one of the inte-grated classes. There were nine integrated classes in the school at the time of thestudy. They contained either 12 or 15 students with one lead teacher and two as-sistant teachers. The ratio of students with developmental delays to typically de-veloping students was 6:6 in classes of 12, and 7:8 in classes of 15.

Three classes each were randomly assigned to Groups A, B, and C, thus en-suring that within each class, all children with developmental delays received thesame treatment. It should be noted that although typically developing studentswere in the same classes as students with developmental delays, they did not par-ticipate in the study and did not receive social skills instruction. Because targetbehaviors were reinforced in the classroom, typically developing students did re-ceive reinforcement for sharing and being in a group, but no data on target be-haviors were collected for these students.

Groups A and B each contained 19 students with developmental delays;Group C had 20 students with developmental delays. A total of 42 boys and 16girls participated in the study (Group A: 14 boys, 4 girls; Group B: 12 boys, 7girls; Group C: 15 boys, 5 girls). There was no statistically significant differencebetween groups regarding gender, x2 (N = 58, 2) = 1.29; p > .05.


TABLE 1. Age, SES Scores, IQs, and Speech and Language Scores forParticipant Groups

Descriptor

Age in MonthsSES ScoreCognitive ScoreSpeech and Language

Score

Total Sample

M

50.7133.1671.7168.98

SD

8.2914.2717.4314.89

Group A

M

49.3730.5366.3768.42

SD

9.1514.2817.4612.62

Group B

M

49.8433.5369.8963.89

SD

7.3613.3617.5112.57

Group C

M

52.8037.0578.5074.35

SD

8.2615.0715.8717.55

Table 1 presents ages, socioeconomic status (SES) scores as assessed by theHollingshead Four Factor Index of Social Status (Hollingshead, 1975), cognitivetest scores, and Speech and Language scores by participant groups. Most students'parents were employed within the skilled craftsmen, clerical, and sales workerssector, which is considered to fall in the middle income range (SES scores from 30to 39). There were no statistically significant differences between groups by age,F(2, 57) = .99,p >. 05; SES, F(2,57) = 1.02,/? > .05; cognitive test scores, F(2,57)= 2.66,p > .05; or Speech and Language scores, F(2,57) = 2.55, p > .05. Table 2presents frequency data on participants' race and ethnicity. There were no statis-tically significant group differences for race, x2 (N = 58, 8) — 3.26,p > .05.

Target Behaviors

Sharing and Being in a Group were the two target behaviors used in this study.They were selected for study after a review of the literature showed that thesebehaviors are often deficient in children with developmental delays (Bronson etal., 1995; LeBlanc & Matson, 1995; Faught et al., 1983). Elliott, Racine, andBusse (1995) indicated that the two components of social interactions of most in-terest to investigators of preschool children's social skills are peer entry skillsand skills that maintain social interactions. Being in a Group was chosen forstudy because it is an indicator of successful peer entry skills. Sharing was cho-

TABLE 2. Race or Ethnicity of Participants in Each Croup

Total Sample Group A Group B Group C

Race or Ethnicity

CaucasianAfrican AmericanHispanicAsian AmericanOther

n

21101764

%

36.2117.2429.3110.346.90

n

63622

%

31.5815.7831.5810.5310.53

n

75511

%

36.8426.3226.32

5.265.26

n

82631

%

40.0010.0030.0015.005.00


sen because this behavior would serve to maintain social interactions. Defini-tions of these target behaviors follow.

Sharing. Sharing was defined as the observed student jointly using a toy ormaterials with another student. Either student may have initiated sharing. For ex-ample, a student asked another student being observed for some blocks and theobserved student gave the other student a block. Sharing was coded as (a) stu-dent shared a toy or materials with another student, or (b) student did not sharetoys or materials with another student.

Being in a Croup. Being in a Group was defined as the observed student re-maining in close proximity to at least one other student for the duration of theobservation period. The student must have been observed to either talk to, listento, or look at the other student(s). Being in a Group was coded as (a) student wasin a group, or (b) student was not in a group.

Social Skills Treatment Program

Taking Part, Introducing Social Skills to Children (Cartledge & Kleefeld, 1991)is a commercially available social skills training program for children in generaleducation and mainstream classes. The information taught is considered to be es-sential for the development of social skills. The curriculum defines the behaviorand teaches the skill through exemplars, question answering, and providing feed-back. Practice items are also included in the curriculum to facilitate generaliza-tion.

After a review of the literature, four early childhood social skills curriculawere identified. One program (Davis, 1988) was eliminated because of restrictedage range (4- to 5-year olds), and a second (Dinkmeyer & Dinkmeyer, 1982) wasnot used because of limited coverage of targeted social skills. Another (Walkeret al., 1983) was not developed for preschool-aged students; with the exceptionof Taking Part (Cartledge & Kleefeld, 1991), none were specific for use withstudents with developmental delays. In addition, the use of puppets, which hasbeen found to be effective with preschool-aged children (Hundert & Houghton,1992; Matson, Fee, Coe, & Smith, 1991), and the flexibility of administration ofTaking Part (Cartledge & Kleefeld, 1991) led to the selection of this curriculum.A unique feature of the Taking Part social skills training program is its potentialuse with both regular education and mainstream students and, therefore, it wouldbe appropriate for all students attending an integrated program.

Taking Part includes six units with five to seven skills in each unit. The cur-riculum is flexible and skills may be taught separately or as a part of the entireunit. Two lessons from Unit 4, Cooperating with Peers, were selected for use inthis study: Joining a Group Activity and Sharing Materials. These lessons wereselected because these skills are considered critical in the development of socialskills (Gresham & Elliott, 1990), are prerequisite skills for more complex socialbehaviors, and are developmentally appropriate for preschool-aged children.Several other studies have examined sharing and social initiations as evidence of


social competence (Bronson et al., 1995; LeBlanc & Matson, 1995). Appendix Aincludes directions provided to instructors for use with the lesson for the first dayof social skills training in this study. Appendix B gives the checklist for treat-ment integrity that was used for this lesson.

Design

Table 3 presents a schematic representation of the design used in the study. Thisdesign permitted examination of the comparative effectiveness of (a) the socialskills training plus classroom reinforcement of target behavior versus classroomreinforcement of target behaviors alone; (b) social skills training plus classroomreinforcement of target behaviors versus no intervention; and (c) classroom rein-forcement of target behaviors alone versus no intervention. The design allowedexamination of maintenance of possible gains from the combined treatments byclassroom reinforcement alone and allowed for comparison of maintenance byclassroom reinforcement to the joint effects of social skills training plus class-room reinforcement. The design also enabled evaluation of the effects of re-peated assessment alone of the control group.

The social skills instruction plus classroom reinforcement of target behaviorscondition was included because the literature reviewed (Kamps, Ellis, Mancina,& Green, 1995; Storey, Danko, Ashworth, & Strain, 1992) indicated that thiscombination is more effective than social skills training alone for social skills ac-quisition and generalization. The reinforcement of target behaviors-alone groupwas included to determine the effects of reinforcement alone and to isolate theeffects of social skills training. The no treatment group controlled for the pas-sage of time. A social skills-only group was not included because the research

TABLE 3. Schematic Representation of Study Design

Group

A(n=19)

B(n=19)

C (n = 20)

FirstAssessment(3 days)

Observation ofTarget Behaviors



FirstIntervention(8 days)

Social Skills +ClassroomReinforcementof TargetBehaviors

ClassroomReinforcementof TargetBehaviors

NoIntervention

SecondAssessment(3 days)




SecondIntervention(8 days)

ClassroomReinforcementof TargetBehaviors

Social Skills +ClassroomReinforcementof TargetBehaviors

NoIntervention

ThirdAssessment(3 days)

Observationof TargetBehaviors




cited previously has shown that children only use acquired social skills for ashort time in the absence of reinforcement of target behaviors.

Procedure

Assessment was conducted preintervention and at the end of the two treatmentphases. Behavioral observation data were collected during 3 days for each as-sessment period by two observers blind to treatment condition, who entered theclassroom during free play periods (30 minutes) and observed each participant inthe group using a time-sampling procedure. Observations (20 minutes) were per-formed by recording the frequency of target behaviors in 15-second blocks foreach participant in the group, sequentially and in fixed order. Within each 15-second interval, observers recorded the frequency of occurrence of each targetbehavior. The list of participants was randomly ordered for each assessment.Each target behavior was observed for 20 intervals daily, 60 for each assessmentphase, totaling 180 observation intervals for the entire study.

After 3 days of preintervention assessment, Group A participants received 8days of social skills training administered by a trained social skills instructor andassistant plus classroom reinforcement of target behaviors. During social skillsinstruction, the classroom teacher left the room and the typically developing stu-dents went to another part of the classroom where they were supervised in freeplay by the teacher assistant. While their typically developing peers were play-ing in one part of the room, participants with developmental delays were giveninstruction in joining a group and sharing by the social skills instructor and assis-tant. Activities included a motivational activity where the skill was presentedduring a short skit using puppets, followed by a modeling activity where the stu-dents could practice the skill. After the skill presentation, participants were givenspecific instructions and practice-initiating interventions, such as establishingeye contact with at least one group member, tapping another child on the shoul-der to gain attention, and asking to play with others. Participants were given spe-cific activities that required sharing, such as coloring using one box of crayonsand sharing play materials. Other activities included reading a book to partici-pants about playing together, followed by an interactive discussion on how play-ing with others was important in the story. A story focusing on sharing was alsoread and discussed with participants. To ensure integrity of the social skills inter-vention, each social skills training session was tape recorded. Two raters inde-pendently reviewed each tape according to a protocol that evaluated treatment-specific intervention components from the treatment manual.

Group B participants received 8 days of general time with the social skills in-structor and assistant, plus classwide reinforcement of target behaviors. Duringthis time the teacher left the classroom and the typically developing studentswent to another part of the classroom where they were supervised in free play bythe teacher assistant. General time with the social skills instructor was includedfor Group B to control for exposure of the students to the instructor. During this


general time with the social skills instructor, students were allowed free play.Social skill training sessions and general time with the social skills instructoreach lasted approximately 20 to 30 minutes.

During classroom free play in the morning and afternoon of the 8-day inter-vention period, participants from Groups A and B received continuous verbal re-inforcement for target behaviors and intermittent preferred tangible reinforcers.In addition, typically developing children without developmental delays presentin the classroom also received reinforcement for sharing and being in a group,although no data were collected on the effects of this reinforcement on their be-havior and they were not provided with social skills training. Teachers andteacher assistants met with the first author for two, 1-hour training sessionswhere both modeling and role playing were used to teach proper administrationof reinforcement of target behaviors. Definitions of target behaviors were dis-tributed to all teaching staff. A total of 6 teachers and 14 assistant teachers ad-ministered classroom reinforcement.

Participants were individually evaluated for tangible reinforcement prefer-ences. Nine items from the Reinforcement Inventory (Willis, LaVigna, & Don-nellan, 1987) were selected based on participant preference to include ediblesand stickers that were not typically available to participants in school. Partici-pants were asked to state, point to, or select one item they liked best. For three 2-minute intervals during both free play periods, teachers and assistants providedpreferred tangible reinforcers and continuous verbal reinforcement to partici-pants, and their typically developing peers, for target behaviors paired with ver-bal praise. Reinforcement was also provided by teachers and assistants during adaily classroom group activity. This was included to help generalize social skillsfrom free play to a semi structured classroom activity. To ensure integrity of rein-forcement, a trained observer randomly observed teachers and aids. The ob-server recorded the frequency of reinforcement for each target behavior for allparticipants in a group. Observations were conducted twice daily for 5 minutesat random times during the treatment phases. When observations indicated thatreinforcement of target behaviors was less than 80%, the first author retrainedteachers and assistants on implementation of reinforcement.

At the end of the first treatment phase, there were 3 days of assessment (de-scribed above) followed by a reversal of treatments for Groups A and B for 8days. During this time, Group A received classroom reinforcement of target be-haviors only and Group B received social skill instruction plus classroom rein-forcement of target behaviors. Group C continued to receive no treatment. Afterthis, there were 3 final days of assessment.

As a measure of social validity (Wolf, 1978), classroom teachers completed abrief questionnaire at the conclusion of the study. Teachers were asked to com-ment on the following questions: "Are the social skills of sharing and joining agroup valuable to students in your class? Do you think the amount of time thatwas spent in training and administering the reinforcement was worthwhile? Didyou notice changes in student behavior? Would you like to learn more about the


curriculum used in this study to teach students social skills in the future?" Theformat of the questionnaire was open ended and teachers were encouraged to an-swer the questions in a blank space after each question. An additional indicationof social validity was obtained by asking participating students, "Did you havefun playing and learning together?"

RESULTS

Treatment Integrity

Tape-recorded treatment sessions were reviewed by two raters who checkedwhether critical components of social skills training were adhered to using atreatment protocol checklist. All sessions met 100% compliance with the treat-ment manual as scored by both raters.

Observations of frequency of teacher reinforcement of target behaviors weremade. Treatment integrity was computed by dividing the number of target be-haviors paired with reinforcement by the total number of target behaviors andmultiplying by 100 (Alberto & Troutman, 1990). An average treatment integrityof .86 (.90 for Sharing, .81 for Being in a Group) was found. On three occasions(twice for the teachers in one class and once for teachers in another class) whentreatment integrity fell below 80%, teachers and assistants were retrained to ad-minister reinforcement for target behaviors.

Reliability of Behavioral Observations

Interobserver reliability of observations of target behaviors during classroomfree play was computed by dividing the number of agreements by the total num-ber of observations and multiplying by 100 (Alberto & Troutman, 1990). Aver-age interobserver reliability for all observations was .88 (.91 for Sharing and .86for Being in a Group). On the two occasions when reliabilities fell below .80, thefirst author reviewed the target behaviors with the observers and observed thecollection of data during a mock observation or observed during the actual datacollection.

Treatment Results

Table 4 presents means and SD for Sharing and Being in a Group for each of thethree assessment periods. A repeated-measures analysis of variance (ANOVA)found a statistically significant within-subjects effect for Sharing, F(2, 54) =13.61,/? < .001, indicating that participants' sharing scores changed over time. Astatistically significant group x time effect, F(4, 110) = 4.89,/? < .002, showedthat the Sharing behavior of participants in the three groups changed differen-tially across the assessment periods. Table 5 presents the results of follow-up,post-hoc Bonferroni tests. Sharing scores for the groups did not differ at thepreintervention stage but after the initial intervention, Group A (social skills plus


TABLE 4. Means and SD for Sharing and Being in a Group for the ThreeAssessment Periods

Assessment Period

First

Second

Third

Variable

SharingBeing in a GroupSharingBeing in a GroupSharingBeing in a Group

Group A

M

2.684.135.21

10.625.71

10.58

SD

1.712.882.554.193.066.56

Group B

M

3.164.162.769.585.00

11.37

SD

2.373.592.553.223.245.90

Group C

M

2.452.702.682.752.834.15

SD

2.162.422.331.912.302.41

Note. For each assessment period, Sharing and Being in a Group were observed for 60 15-second intervals for eachparticipant.

classroom reinforcement) scores were statistically significantly higher than boththose of Group B (classroom reinforcement only) and Group C (control). Afterthe second intervention when treatments for Groups A and B were reversed, par-ticipants in both Group A and Group B displayed statistically significantly moreSharing behavior than participants in Group C.

A repeated-measures ANOVA found a statistically significant within-subjectseffect for Being in a Group, F(2, 54) = 48.44, p < .001, indicating that partici-pants' Being in a Group scores changed over time. A statistically significantgroup x time effect, F(4, 110) = 8.14,/? < .001, showed that the Being in a Groupbehavior of participants in the three groups changed differentially across the as-sessment periods. The post-hoc Bonferroni results in Table 5 show no group dif-ferences for Being in a Group at preintervention. After both the first and secondinterventions, participants in both Groups A and B displayed more Being in aGroup behavior than control participants from Group C.

Table 6 presents the effect sizes for the two experimental groups (Groups Aand B) relative to the no treatment control group (Group C). The effect size (d)was calculated by subtracting the mean of the control group from the mean of the

TABLE 5. Results of Post-hoc Analyses of Social Skills and Target Behaviors

Dependent Variable

Sharing

Being in a Group

First Assessment(Preintervention)

A = BA = CB = CA = BA = CB = C

Second Assessment(Post Intervention 1)

A>B*A>C*B = CA = BA>C*B>C*

Third Assessment(Post Intervention 2)

A = BA>C*B>C*A = B

A>C*B>C*

Note. The designations A, B, and C indicate Groups A, B, and C, respectively.*p < .05 for Bonferroni post hoc comparison.


TABLE 6. Treatment Effect Sizes for Groups A and B Relative to NoTreatment Controls

experimental group and dividing by the SD of the control group (Borg & Gall,1989, p. 172) for each dependent variable. Seven of the eight effect sizes reportedwere large, exceeding .80 (Cohen, 1992).

Effect sizes do not show whether individual participants' behavior changed ina clinically meaningful way. Jacobson and Truax (1991) suggested that todemonstrate clinically significant improvement, participants' " level of function-ing subsequent to therapy should fall outside the range of the dysfunctional pop-ulation, where range is defined as extending to two standard deviations beyondthe mean for that population" (p. 13). Tables 7 and 8 show how many partici-pants in each group achieved postintervention scores that were two SD beyondthe preintervention sample mean for Sharing and Being in a Group, respectively.These tables show that the combination of social skills plus classroom reinforce-ment of target behaviors resulted in clinically significant improvements in Shar-

TABLE 7. Individual Participants' Changes in Sharing Scores at the Second andThird Assessment Periods

Group

ABC

ABC

n<2SDa

131819

131318


% n = or > 2 SD

68.4 694.7 195.0 1


68.4 668.4 690.0 2

%

31.65.35.0

31.631.610.0

"Columns show the number and percentage of participants whose Sharing scores did not increase or increased lessthan 2 SD (a score of 7) from the group pretest mean and the number and percentage of participants whose Sharingscores increased 2 SD or more.

Dependent measure

Effect sizes Sharing Being in a Group

Group ASecond Assessment 1.08 4.12Third Assessment 1.25 2.66

Group BSecond Assessment .03 3.59Third Assessment .94 2.99


TABLE 8. Individual Participants' Changes in Being in a Group Scores at the Secondand Third Assessment Periods


Group n<2SDa % n = or> 2 SD %

A 10 52.6 9 47.4B 10 52.6 9 47.4C 20 100 0 0.0


ABC

76

20

36.831.6

100

12130

63.268.40.0

"Columns show the number and percentage of participants whose Being in a Group scores did not increase or in-creased less than 2 standard deviations (a score of 10) from the group pretest mean and the number and percentageof participants whose Being in a Group scores increased 2 standard deviations or more.

ing for about one-third of the participants that was maintained, following a 3-dayassessment period of no treatment, with 8 days of classroom reinforcement only.Either social skills plus reinforcement or reinforcement alone resulted in clini-cally significant improvements in Being in a Group for about half of the partici-pants. After the 3-day assessment period followed by a reversal of conditions for8 days, the proportion of participants in both Groups A and B demonstratingclinical improvement in Being in a Group increased to about two thirds.

Social Validity of Treatment

All but 8 of the 58 participating students (86%) responded that they had funplaying and learning together. Seven students did not respond, and one said,"no." Five of the six classroom teachers returned the social validity question-naire. Respondents felt that the skills of joining and sharing were important fortheir students. Four of the teachers commented that most, but not all (particularlythose with severe developmental disorders such as autism), of the children bene-fited from this type of training. The teachers expressed an interest in using partsof the curriculum in their classrooms based on the positive response they had ob-served during the study.

DISCUSSION

The results indicate that social skills lessons from Taking Part (Cartledge &Kleefeld, 1991) incorporated into the curriculum and combined with classroomreinforcement of target behaviors were effective in increasing social behaviorsof children with developmental delays. These gains were maintained by class-


room reinforcement alone. Teachers found the program valuable and felt that itaddressed important classroom behaviors. Participants also generally enjoyedthe program.

Some important differences were found with regard to treatment type and tar-get behavior. After the initial intervention, the combination of social skills train-ing plus classroom reinforcement of target behaviors resulted in significant in-creases in Sharing behavior over those obtained by the no treatment group and agroup of participants receiving reinforcement of classroom behavior alone. There-fore, social skills training plus reinforcement was more effective than reinforce-ment alone for the dependent variable of Sharing. Participants who received socialskills training plus classroom reinforcement and those who received classroom re-inforcement of target behaviors only displayed significantly more Being in aGroup behavior than control participants. Thus, social skills training did not addto the effects of reinforcement only for Being in a Group behavior.

These results may be partially explained by how target behaviors were typi-cally treated in the classroom before the beginning of the study. Teachers com-mented that sharing is a behavior that was often encouraged during classroomactivities, whereas being part of a group was not. Before the study, most partici-pants had been praised for sharing but probably not as consistently as they wereduring this study and the reinforcement alone condition did not result in in-creases in this behavior. Being in a group, which was less often encouraged be-fore the study, increased when either reinforcement alone or the combined treat-ment conditions were administered.

The gains obtained by participants receiving social skill plus classroom rein-forcement for target behaviors were maintained, following a 3-day assessmentperiod of no treatment, by an 8-day period of classroom reinforcement alone.The social skills plus reinforcement treatment package led to clinically notice-able gains in sharing for approximately one-third of participants; both socialskills plus reinforcement and reinforcement alone resulted in clinically signifi-cant gains in being in a group behavior for approximately one-half of the partici-pants.

This study has several limitations that warrant discussion. Participants inGroups A and B were exposed to the social skills instructor throughout the treat-ment phases, regardless of whether they were receiving social skills training.There was, however, no control for this exposure for Group C participants, whowere never exposed to a social skills instructor. Thus, we do not know what ef-fect introduction of a new person into the classroom would have had on the tar-get behaviors of Group C participants.

As indicated above, in the school where the study took place, sharing behaviorwas encouraged and children were praised for this behavior. Although teachersof participants in control Group C were not trained in the use of reinforcement oftarget behaviors for this study, they did praise sharing behavior as part of theirtypical classroom behavior. Because reinforcement frequency data were not col-lected for Group C, it cannot be determined with certainty if Group C teachers'


reinforcement rates, particularly for sharing behavior, differed from those ofGroup A and B during treatment, although it was our clear clinical impressionthat this was the case. Future studies should assess reinforcement rates for bothcontrol and intervention groups.

Demand characteristics may have been operating in the collection of socialvalidity data, particularly for child participants. Children were directly asked bysocial skills instructors if they enjoyed the program. They may have felt pressureto answer affirmatively.

Future researchers should investigate the effectiveness of acquisition andmaintenance of social skills gains using intermittent, rather than continuous, re-inforcement. Future research should also examine maintenance of social skillsgains when reinforcement is faded. The social skills interventions and follow-ups in the current study were brief, limiting information on the effectiveness ofthis curriculum as a whole. Longer-term follow-up assessments of the presentstudents would have been informative to determine if the gains found duringtraining were maintained. Most of the students graduated at the end of the 6-week summer session and follow-up could not be accomplished with these par-ticipants. In addition, this study included only students with developmental de-lays in the social skills training sessions in an effort to keep groups small and toprovide evidence concerning the appropriateness of this curriculum for this pop-ulation. Future studies should include both typically developing peers and stu-dents with developmental delays in the social skills training lessons.

This study examined a short-term intervention for two of the social skills fromthe Taking Part (Cartledge & Kleefeld, 1991) curriculum, which covers 33 so-cial skills that may be taught to children from preschool through third grade. Thesocial skills covered here were relevant for preschool students. Longer interven-tions using more of the skills are needed to determine the validity of this curricu-lum for students in other age groups.

Inferences from the results can only be applied to students with similar char-acteristics. Participants were predominantly students within the mildly delayedrange of cognitive functioning, with obtained speech and language scores com-mensurate with mild delays.

Social skill deficits broadly impact educational outcome and teachers report agreat need for readily available, empirically tested social skills curricula (Odomet al., 1993). Results of this research indicate that the Taking Part program canbe used to teach social skills to preschool children with developmental delays.This program may be used to teach children with developmental delays the so-cial skills they need to interact more effectively with both delayed and typicallydeveloping peers.


REFERENCES

Alberto, P. A., & Troutman, A. C. (1990). Applied behavior analysis for teachers. Englewood Cliffs,NJ: Prentice-Hall.

Baker, E. E., Wang, M. C , & Walberg, H. J. (1995). The effects of inclusion on learning. Educa-tional Leadership, 53, 33-35.

Bayley, N. (1993). Bayley scales of infant development (2nd ed.). San Antonio, TX: The Psychologi-cal Corporation.

Borg, W. R., & Gall, M. D. (1989). Education research: An introduction (5th ed.). White Plains, NY:Longman.

Bredekamp, S., & Copple, C. (1997). Developmentally appropriate practice in early childhood pro-grams. Washington, DC: National Association for the Education of Young Children.

Bronson, M. B., Hauser-Cram, P., & Warfield, M. J. (1995). Classroom behaviors of preschool chil-dren with and without developmental disabilities. Journal of Applied Developmental Psy-chology, 16, 371-390.

Cartledge, G., & Kleefeld, J. (1991). Taking part, introducing social skills to children. Circle Pines,MN: American Guidance Service.

Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155-159.Davis, D. E. (1988). My friends and me. Circle Pines, MN: American Guidance Service.Dinkmeyer, D., Sr., & Dinkmeyer, D., Jr. (1982). Development of understanding of self and others

(DUSO). Circle Pines, MN: American Guidance Service.Elliott, S. N., Racine, C. N., & Busse, R. T. (1995). Best practices in preschool social skills training.

In A. Thomas & J. Grimes (Eds.), Best practices in school psychology-Ill (pp. 1009-1020).Washington, DC: National Association of School Psychologists.

Faught, K. K., Balleweg, B. J., Crow, R. E., & Van Den Pol, R. A. (1983). An analysis of social be-haviors among developmentally disabled and nondisabled preschool children. Education andTraining of the Mentally Retarded, 18, 210-214.

Gresham, F. M. (1982). Misguided mainstreaming: The case of social skills training with disabledchildren. Exceptional Children, 48, 422^433.

Gresham, F. W., & Elliott, S. N. (1990). Social skills rating system manual. Circle Pines, MN: Amer-ican Guidance Service.

Guralnick, M. J. (1992). A hierarchical model for understanding children's peer-related social com-petence. In S. L. Odom, S. R. McConnell, & M. A. McEvoy (Eds.), Social competence ofyoung children with disabilities: Issues and strategies for intervention (pp. 37-64). Balti-more: Brooks.

Guralnick, M. J., & Weinhouse, E. M. (1984). Peer-related social interactions of developmentally de-layed young children: Development and characteristics. Developmental Psychology, 20,815-827.

Hanline, M. F. (1985). Integrating disabled children. Young Children, 40, 45^8 .Hollingshead, A. B. (1975). Four factor index of social status. New Haven, CT: Yale University

Press.Hundert, J., & Houghton, A. (1992). Promoting social interaction of children with disabilities in inte-

grated preschools: A failure to generalize. Exceptional Children, 58, 311-320.Jacobson, N. S., & Truax, P. (1991). Clinical significance: A statistical approach to defining meaningful

change in psychotherapy research. Journal of Consulting and Clinical Psychology, 59, 12-19.Ramps, D. M., Ellis, C, Mancina, C , & Greene, L. (1995). Peer-inclusive social groups for young

children with behavioral risks. Preventing School Failure, 39, 10-15.LeBlanc, L. A., & Matson, J. L. (1995). A social skills training program for preschoolers with devel-

opmental delays. Behavior Modification, 19, 234-246.Madden, N. A., & Slavin, R. E. (1983). Mainstreaming students with mild handicaps: Academic and

social outcomes. Review of Educational Research, 53, 519-569.Matson, J. L., Fee, V. E., Coe, D. A., & Smith, D. (1991). A social skills program for developmen-

tally delayed preschoolers. Journal of Clinical Child Psychology, 20, 428^433.


McAllister, J. R. (1991). Curriculum-based behavioral interventions for preschool children withhandicaps. Topics in Early Childhood Special Education, 58, 48—58.

McConnell, S. R. (1987). Entrapment effects and the generalization and maintenance of social skillstraining for elementary school students with behavior disorders. Behavioral Disorders, 12,252-263.

New York State Education Department. (1993). Update Part 200 Regulations of the Commissionerof Education. Albany, NY: Author.

Odom, S. L., & McEvoy, M. A. (1988). Integration of young children with handicaps and normallydeveloping children. In S. Odom & M. Karnes (Eds.), Early intervention for infants and chil-dren with handicaps: An empirical base (pp. 241—267). Baltimore: Brookes.

Odom, S. L., McConnell, S. R., & Chandler, L. K. (1993). Acceptability and feasibility of class-room-based social interaction interventions for young children with disabilities. ExceptionalChild, 60, 226-236.

Odom, S. L., & Strain, P. S. (1984). Peer-mediated approaches to promoting children's social inter-action: A review. American Journal of Orthopsychiatry, 54, 544-557.

Storey, K., Danko, C. D., Ashworth, R., & Strain, P. S. (1994). Generalization of social skills inter-vention for preschoolers with social delays. Education and Treatment of Children, 17, 29-51.

Thorndike, R. L., Hagen, E. P., & Sattler, J. M. (1986). Stanford-Binet Intelligence Scale (4th ed.).Chicago: Riverside.

Walker, H. M., McConnell, S., Holmes, D., Todis, B., Walker, J., & Golden, N. (1983). The Walkersocial skills curriculum—the ACCEPTS program: A curriculum for children's effective andteacher skills. Austin, TX: Pro-ed.

Willis, T. J., LaVigna, G. W., & Donnellan, A. (1987). Behavior assessment guide. Los Angeles: In-stitute for Applied Behavior Analysis.

Wolf, M. M. (1978). Social validity: The case for subjective measurement or how applied behavioranalysis is finding its heart. Journal of Applied Behavior Analysis, 11, 203-214.

Zimmerman, I. L., Steiner, V. G., & Pond, R. E. (1991). Preschool language scales-3. San Antonio,TX: The Psychological Corporation.

Action Editor: Terry B. GutkinAcceptance Date: February 9, 2001

APPENDIX A

Sample Directions for Social Skills Instructor

The social skills trainer and assistant followed the procedures outlined belowduring two of the social skills training programs.

Materials List

Taking Part manual; squirrel, hawk, turtle, rabbit, and bear puppets from theTaking Part kit; a book made of construction paper that can be torn; a large chil-dren's book; large blocks; and stickers from Taking Part kit.


First Lesson, Day 1

Trainer Reviews Procedure

"We are going to be playing together everyday this week. Some days we willread a story together. The stories will be about sharing toys and playing with ourfriends. After we read the story, you will all have a turn to hold one of the pup-pets and we can all tell the story together again. Then we will talk about thestory, and I will give out stickers (show stickers) to everyone who is listening.You can stop any time you want. Do you want to play?"

Trainer and Assistant Enact the Story Sharing Materials using thePuppets Provided

Trainer engages students in an enactment of the story using the puppets. Trainerassigns three students to be the squirrel, hawk, and turtle. The trainer reviewswith each student the animal puppet they are holding. The trainer reads the storyand prompts students to enact the story with the puppets and repeat the trainer (ifpossible). Trainer allows each student to select a Taking Part sticker.

Trainer: "Today we learned how we can share with our friends. When you arein class today, I want each of you to share with your friends the way Hank Hawkand Shelli Squirrel learned to. Your teachers will be giving everyone specialstickers like the ones I just gave you when you share with your friends."

APPENDIX B

Treatment Integrity Checklist for Day 1 Lesson

Day 1

Trainer was prepared with needed materials.Trainer reviewed the procedures.Trainer followed the story.Trainer engaged the students in the reenactment of the story.Trainer provides stickers.

Date post:	19-Sep-2018
Category:	Documents
Upload:	trinhanh
View:	215 times
Download:	0 times

Socia l Skill Trainin g in an Integrate d Preschoo l Progra m · Socia l Skill Trainin g in an...

Documents