A comparison of the effects of two social skill training approaches on teacher and child behavior.

Abstract. The importance of social skills to achievement of important life outcomes has been long documented, as have the social skill deficits encountered by many children with disabilities. Numerous social skill interventions have been implemented in inclusive settings with young children with disabilities; however, many of the interventions were multi-faceted, making it difficult to identify specific intervention activities that accounted for behavioral change. The purpose of the present study was to compare two social skill interventions to determine whether changes in teacher behavior resulted in differential changes in preschool children's social skill behaviors. The results indicated that teacher behavior could be modified and that praise was a strong component in facilitating change among children. The results also demonstrated that the inclusion of young children with disabilities in small-group art activities did not negatively impact the social behaviors of the group. However, neither intervention resulted in significant changes in the social behaviors of the young children with disabilities. Continued research is needed to identify interventions that teachers can readily implement and that will positively impact the social skill development of preschool-age children, with and without disabilities.

**********

The importance of age-appropriate social skills to the success of children in the classroom has been documented numerous times in the literature (Gresham, Sugai, & Horner, 2001; McFall, 1982; Parker & Asher, 1987). The lack of appropriate social skills has been associated with later difficulties in life, such as failure to complete school, difficulties with the police, unemployment, social isolation (Brown, Odom, & Conroy, 2001; Parker & Asher, 1987; Sheridan, 1998) and as a limiting factor in school success (Fantuzzo, Sekino, & Cohen, 2004). Research that validates strategies for enhancing the social skill and social competence levels of young children is critical to both identifying social skill deficiencies and implementing effective plans for instructional remediation.

Because the development of age-appropriate social skills is critical for all children, all can benefit from curricula that include an emphasis on social skills (Lau, Higgins, Gelfer, Hong, & Miller, 2005). However, as Baumgart, Filler, and Askvig (1991) found in a large survey of parents and teachers, a focus upon social skill development is particularly important for young children with identified disabilities. While the lack of appropriate social skills is a primary barrier to the successful inclusion of children with disabilities in general education settings (McGinnis & Goldstein, 1984), the inclusion of children with disabilities will likely increase as major organizations, such as the National Association for the Education of Young Children, continue advocating for individualizing instruction to meet the needs of all children (Bredekamp & Copple, 1997). The movement toward inclusive preschool settings has significant implications for preschool teachers as well as the children in their charge. Teachers must be provided with training in research-based social skill teaching/intervention strategies that can be used effectively and efficiently in the course of typical activities.

This change in preferred preschool placement for young children with disabilities has been accompanied by numerous studies of the impact of inclusive placements on the social interactions of young children with and without disabilities. Guralnick and Groom (1988) found that young children with delays exhibited twice as many positive social interactions when placed in integrated play groups than when in the segregated play groups. Similarly, Buysse and Bailey (1993) reviewed the literature and concluded that integrated placements for young children were beneficial. Peck, Carlson, and Helmstetter (1992) surveyed parents of typically developing kindergarten children who attended inclusive kindergartens and found that the parents overwhelmingly deemed the inclusive placements in a favorable light. The parents did not believe that their children imitated undesirable behaviors and felt that their children were more accepting and helpful toward other children.

Although placement in integrated settings has been demonstrated to have a positive impact on the social development of young children with disabilities, an inclusive placement alone does not guarantee that the children's social behavior will improve. Simple contact or physical proximity does not ensure that children (with and without disabilities) will begin to play together or that children with disabilities will develop more appropriate social skills through modeling or imitating their peers' age-appropriate skills (Filler & Xu, 2006; Guralnick, 1990). In fact, Guralnick, Connor, Hammond, Gottman, and Kinnish (1995) found that children with disabilities may even exhibit more negative social behavior in an integrated setting than in a specialized setting. The need for specific teacher-mediated and/or peer-mediated procedures designed to support pro-social behavior has long been recognized in the literature (e.g., McConnell, McEvoy, & Odom, 1992).

Numerous studies have examined the effectiveness of teacher-mediated strategies to enhance children's social skills. By far, the majority have involved compound treatments consisting of simultaneous antecedent environmental manipulations as well as consequent praise for appropriate interactions. For example, Hanline (1993) used the following strategies to facilitate interactions: placing or positioning children with disabilities in areas that would encourage interaction, prompting and reinforcing appropriate social behavior, modeling social interactions, interpreting behaviors of children with disabilities, and answering children's questions regarding their peers with disabilities. Hundert and Houghton (1992) implemented a class-wide approach that targeted children with and without disabilities as well as their teachers. The training components consisted of direct instruction in specific social skills, puppet modeling of the social skill, rehearsal with feedback, teacher prompting and contingent praise, and token contingencies. Odom et al. (1999) compared the effects of four different programs: environmental arrangement, child-specific, peer-mediated, and comprehensive. The environmental arrangement consisted of the teacher dividing children into play groups and providing them with specific activities. The child-specific condition consisted of direct instruction social skill lessons for a child with a disability. The peer-mediated condition consisted of children without disabilities participating in social skill instruction, and the comprehensive program included selected aspects from the other three interventions. Each of these studies consisted of a variety of intervention strategies; however, the contribution of individual intervention strategies to child outcomes was not addressed.

In a review of the research, Vaughn et al. (2003) noted that the social skill intervention procedures were often part of an intervention package that consisted of several different intervention practices. According to their review, common features of the intervention packages include: prompting and rehearsal, play-based activities, free-play activities, reinforcement, modeling, storytelling, and direct instruction. It would seem that these practices fall into two broad categories--those that are antecedent to the occurrence of the specific behavior targeted for improvement (e.g., environmental arrangement, prompting, modeling, story telling, etc.), and those that are delivered as a consequence of the target behaviors (e.g., reward, re-direction, etc.). Clearly, the overall effects of many programs have been evaluated, but the probable differential impact of specific program components has not received sufficient attention. Given the critical nature of social skill development and the need for efficiency of teacher effort, it becomes important to compare the relative value of the individual components of compound interventions.

One purpose of this study was to compare two different approaches to teaching social skills to young students in an inclusive preschool program. To do this, it is necessary to first demonstrate that instruction to teachers regarding the different intervention approaches has, in fact, resulted in a demonstrative change in their behavior. Attention then can be focused upon determining whether or not the changes in teacher behavior are associated with concomitant changes in children's social behavior and whether or not such effects are equally apparent for children with and without disabilities. The first approach, Proactive, consisted of a 4-step direct instruction model based on Skillstreaming in Early Childhood (McGinnis & Goldstein, 1984). The authors of Skillstreaming noted that the program was effective for young children, but research support of the program was not provided. The second approach, Reactive, consisted of the teacher providing verbal praise following child demonstration of a target behavior. Critical elements of these two different strategies have been included in most studies of social skill instruction, but the relative ease with which teachers acquire Proactive versus Reactive strategies, as well as their relative effect upon child behavior, have not been sufficiently examined.

Method

Participants

The study was conducted at a university-sponsored, inclusive preschool located in a major metropolitan area in the southwestern United States. Eight preschool teachers, each with at least an associate of arts degree in child development, were employed at the preschool. From this pool of eight, three were randomly selected to participate in the study and then were randomly assigned to the Proactive, Reactive, or Comparison group.

Letters describing the study and requesting permission for children to participate were sent to parents of all 4- and 5-year-old children enrolled in the preschool (total = 64). Fifty-six (88 percent) of the letters were returned, with 55 (98 percent) of those affirming that the children could participate in the study. The ages of the children ranged from 4 years, 3 months to 5 years, 3 months, with a mean chronological age of 4 years, 9 months.

The children were then categorized as follows: boys without disabilities, girls without disabilities, boys with disabilities, and girls with disabilities. Children were randomly selected from each category and assigned to a group. Groups were balanced by gender, since children may show a gender preference with playmates (Howes & Phillipsen, 1992). Each of the groups consisted of 3 girls without disabilities, 1 girl with a disability, 3 boys without disabilities, and 1 boy with a disability, for a total of 8 children in each of the three groups. Once the children were assigned to a group, the group was then randomly assigned to one of the three conditions (Proactive, Reactive, or Comparison). Descriptive information on the children with disabilities is contained in Table 1.

Group Activities and Study Phases

The study consisted of three phases conducted over 18 consecutive school days: Phase I (5-day Pre-Intervention), Phase II (8-day Intervention), and Phase III (5-day Follow-Up). On each day of the study, each group of children participated in a 10-minute art activity that was videotaped. The art activities were held separately for each group of children and the respective teacher. Examples of the art activities included: painting a dinosaur drawn on a large sheet of butcher paper approximately 6' x 10' in size and painting an empty refrigerator box to resemble a space ship. To increase the likelihood of peer interactions, an environmental structuring tactic (Kohler, Anthony, Steighner, & Hoyson, 2001) was implemented with each activity. Thus, art materials were provided for the group to use rather than providing identical sets of materials for each child. For example, during painting activities, 6 paper plates, each containing a different color of tempera paint, were set out in the middle of the art area, readily available for use by any or all of the eight children in the group.

Phase I (Pre-Intervention). During Phase I (5-day Pre-Intervention), all teachers were directed to interact with the children in their usual manner during the 10-minute art activity. Before beginning the art activity on Day 2 and on all subsequent days, the researcher and each teacher watched the video of the art activity that the teacher had conducted on the previous day. During the Pre-Intervention, teachers received no feedback while watching the video.

Phase II (Intervention). During Phase II (8-day Intervention), the teacher of children in the Proactive group was trained by the researcher to teach 5-minute social skill lessons, using an adaptation of activities and strategies from Skillstreaming in Early Childhood (McGinnis & Goldstein, 1984). Immediately before the art activities, the teacher met with the researcher to review the social skill to be taught that day. The researcher provided the teacher with a written description of the skill, verbally reviewed the skill with her, and provided suggestions for teaching the skill. They then reviewed the video of the previous day's art activity. While watching the video, the researcher verbally rewarded the teacher for teaching the social skill and provided input on missed opportunities. However, the use of specific strategies was not addressed during the review of the video. Social skills taught by the teacher were taken from Skillstreaming in Early Childhood and included: days 1 and 2--"joining in"; days 3 and 4--"sharing"; days 5 and 6--"waiting your turn"; and days 7 and 8--"asking someone to play." The teacher was taught to use the following strategies to teach the social skills to the children during the 5-minute social skill lesson: 1) discussing the importance of the skill, 2) identifying the steps necessary to complete the skill, 3) modeling the skill, and 4) providing feedback to children as they role-played the skill.

During Phase II, the teacher of the Reactive group was instructed to provide children with praise, contingent upon the children "initiating positive interactions with peers" or "responding positively to peers." The importance of positive and reciprocal social interactions has been documented in the literature (Fantuzzo et al., 2004; Vaughn et al., 2003). Immediately preceding each art activity, the teacher and researcher viewed video of the previous day's art activity. During this time, the researcher verbally rewarded the teacher for providing contingent praise to the children and alerted her to missed opportunities for using contingent praise. Some examples of "initiating positive interactions" included children asking peers to pass materials, trade items, work together, and showing or discussing materials. "Responding positively to peers" included such behaviors as passing materials to the peer following a request, trading items following a request, and talking about the art activity. To be verbally rewarded for "responding positively to peers," a positive response had to occur within 3 seconds of an initiation or response behavior from a peer. A behavior that occurred after 3 seconds was counted as an initiation, rather than a response, and the duration is in line with that used by Spohn, Timko, and Sainato (1999).

As with the teachers of the two experimental groups, the teacher of the Comparison group met with the researcher immediately before the art activity and watched a video of the previous day's art activity. However, she was not provided with any feedback; rather, she was simply told to continue interacting with the children as she deemed appropriate.

Phase III (Follow-Up). During Phase III (5-day Follow-Up), all three teachers received the same treatment as during Phase I. Before beginning the art activity, each teacher met with the researcher and watched the video of the previous day's art activity and was told to interact with the children as she deemed appropriate.

Data Collection

The videotapes from each art activity were reviewed and frequency counts were taken daily for the following teacher behaviors: 1) discussing the importance of the skill, 2) identifying the steps necessary to complete the skill, 3) modeling the skill, 4) providing feedback to children as they role-played the skill, 5) providing verbal praise to a child following a positive initiation with a peer, and 6) providing verbal praise to a child following a positive response to a peer.

Frequency counts also were collected daily on the following child behaviors: 1) positive initiations with a peer and 2) positive responses to a peer. While those were the target behaviors for the intervention, frequency counts also were taken for negative initiations to a peer and negative responses to a peer. Inter-rater reliability was computed by comparing the ratings of the researcher and a trained observer, ultimately creating a random sample of 25 percent of the video recordings from each phase of the study for both teacher and child behavior. Percent agreement ranged from 81 percent to 100 percent, with a mean of 86 percent for child behavior and 99 percent for teacher behavior.

Results

The data were analyzed to determine first whether the teachers had changed their behaviors and, if so, whether these changes were associated with changes in child behaviors among groups. Child behaviors of interest included: 1) positive initiations to a peer, 2) positive responses to a peer, 3) negative initiations to a peer, and 4) negative responses to a peer. Separate statistical tests were not conducted with the behaviors of children with disabilities, but their performance is reviewed in the discussion section.

Analyses of Teacher Behaviors

Phase I (Pre-Intervention). A visual analysis of the data indicated that none of the teachers exhibited any of the following target behaviors: 1) discussing the importance of the skill, 2) identifying the steps necessary to complete the skill, 3) modeling the skill, and 4) providing feedback to children as they role-played the skill. Results from two one-way ANOVAs on each of the two remaining target behaviors (providing verbal praise following a positive initiation with a peer and providing verbal praise following a positive response to a peer) indicated that there were no significant differences among teachers on their use of praise. Thus, there were no significant differences among teachers on any of the six target behaviors prior to Intervention.

Phase II (Intervention). Data were then analyzed to determine whether there were significant differences among teachers on the six target behaviors during Intervention. Results from a two-way ANOVA (3 Groups x 6 Behaviors) revealed a significant Group x Behavior interaction, F(4,63) = 16.904, p < .05, with the source of the interaction attributed to an increased occurrence of three of the six target behaviors: identifying the steps necessary to complete the skill, providing praise following a positive initiation to a peer, and providing praise following a positive response to a peer.

Since the data showed that the only time "identifying the steps necessary to complete the skill" was used was during Phase II (Intervention), a one-way ANOVA on this behavior was conducted only for Phase II. The results were significant, F(2,21) = 11.636, p < .05, with the source traced to the fact that only the Proactive teacher used this skill and then only during the Intervention phase. Results from one-way ANOVAs (Teachers x Behavior) also indicated a significant difference for the following two target behaviors during Phase II: "providing praise following a positive initiation to a peer," F(2,21) = 27.892, p < .05, and "providing praise following a positive response to a peer," F(2,21) = 18.903, p < .05. The source of these differences was attributed to the significant increase from Phase I (Pre-Intervention) to Phase II (Intervention) in the use of those two behaviors by the Reactive teacher. An inspection of Figure 1 reveals that training provided by the researcher did affect teacher behavior, with the greatest impact appearing to be with the Reactive teacher, who exhibited significant increases on both behaviors she was taught (providing praise following a positive initiation and providing praise following a positive response); the Proactive teacher, on the other hand, exhibited an increase on only one of four behaviors taught to her (identifying the steps necessary to complete the skill). Meanwhile, the performance of the Comparison teacher on these three behaviors remained unchanged from Phase I (Pre-Intervention).

Analyses of Child Behavior

Phase I (Pre-Intervention). Prior to beginning Phase II (Intervention), analyses were conducted to determine whether differences existed in the three groups of children during Phase I (Pre-Intervention) in the frequency of the following behaviors toward peers: positive initiations, negative initiations, positive responses, and negative responses. Results from one-way ANOVAs from Phase I (Pre-Intervention) for each of the four behaviors indicated that there were no significant differences among groups (Proactive, Reactive, and Comparison) on any of these behaviors.

Phase II (Intervention). To determine whether there were differences among the groups during Intervention, a one-way ANOVA (Group x Behavior) was conducted for each target behavior. The results of a one-way ANOVA (Group x Behavior) for "positive initiations" was significant, F(2,162) = 4.958, p < .05, with the Reactive group exhibiting significantly more "positive initiations" (M = 2.7) than the Comparison group (M = 1.45). Similarly, the results of a one-way ANOVA (Group x Behavior) for "positive responses" was also significant, F(2, 163) = 3.786, p < .05, with the Reactive group exhibiting significantly more "positive responses" (M = 1.89) than the Comparison group (M = 1.05). The third behavior in which a significant difference was found was "negative initiations," F(2,163) = 4.779, p < .05. The Comparison group exhibited significantly more "negative initiations" (M = .68) than either the Reactive group (M = .16) or the Proactive group (M = .25). These analyses indicated significant differences between the Reactive and Comparison groups on three of four behaviors (positive initiations, positive responses, and negative initiations) and between the Proactive and Comparison groups on one behavior (negative initiations).

[FIGURE 1 OMITTED]

Phase III (Follow-Up). To further investigate whether there were treatment effects that may not have appeared during the short Intervention phase, a series of one-way ANOVAs (Group x Behavior) were conducted for each of the behaviors, using data from the Follow-Up Phase. The only behavior in which a significant difference was found among groups was "positive initiations," F(2,103) = 5.147, p < .05. The source of this difference was attributed to the Reactive group, which exhibited significantly more "positive initiations" (M = 3.87) than did the Comparison group (M = 2.13). A graphic portrayal of group scores is contained in Figure 2.

Discussion

The purpose of this study was to compare the impact of two different social skill instructional procedures on teacher behavior and to determine whether any changes in teacher behaviors resulted in differential changes in child behaviors.

The Proactive teacher was taught four different target behaviors (discussing the importance of the skill, identifying the steps necessary to complete the skill, modeling the skill, and providing feedback to children during role-play) to teach social skills during small-group instruction immediately preceding the art activity. Clearly, the literature is replete with examples of the value of pre-planning, step-by-step task or skill analysis, and modeling (e.g., Filler, 1976; Mastropieri & Scruggs, 2007; McConnell et al., 1992; McEvoy, Odom, & McConnell, 1992). However, the only target behavior the teacher incorporated into the art activities during Intervention, at least to any significant degree, was "identifying the steps necessary to complete the skill." Several factors may have resulted in the teacher not using all four target behaviors during the art activities. First, the teacher was not specifically directed by the researcher to incorporate those behaviors into the art activity and, without specific training, it may be difficult to incorporate those behaviors into an ongoing activity. It is also possible that the teacher did not consider stopping an activity in order to have a child role-play a skill to be an effective use of time, and/or she may have thought that to do so would cause a disruption of the natural flow of the activity. Additionally, it is possible that the teacher did not see the value of using all four target skills during an art activity, but instead viewed them as skills to be used during a direct instruction lesson.

The Reactive teacher readily incorporated the use of verbal praise for "positive initiations" and "positive responses" during Intervention, whereas she had used these behaviors very minimally or not at all prior to Intervention. It is, therefore, apparent that the researcher's training resulted in the teacher using new skills during the art activity. However, it was also possible that rather than it being a case of the teacher acquiring new skills from the training, it was instead true that the process heightened the teacher's awareness, resulting in an increase in the use of verbal praise. This explanation assumes that the teacher knew praise to be effective but that it had slipped from her repertoire. In any event, the fact remains that the Reactive teacher used praise at a significantly higher rate than either the Proactive or Comparison teachers.

Unlike the Proactive and Reactive teachers, the Comparison teacher exhibited no significant changes in any of the targeted behaviors during the entire study. Simply watching a video of the previous day's art activity was not sufficient to cause change in the teacher's behavior.

Taken together, these results suggest that teachers can easily be instructed to use verbal praise during group activities and can readily incorporate the use of praise without interfering in the ongoing activity, a conclusion that is supported by other, earlier studies by these and other investigators (Filler, Hecimovic, & Blue, 1978; Hunt, Staub, Alwell, & Goetz, 1994; Vaughn et al., 2003). Given the Reactive teacher's minimal use of praise during Pre-Intervention, it may be that teachers who fail to use verbal praise on a regular basis could benefit from "refresher" instruction on the importance of contingent praise. The results also suggest that, as Filler and Xu (2006) have noted, teachers may not automatically incorporate social skill instructional activities, such as the structured activities taught to the Proactive teacher, into ongoing activities.

[FIGURE 2 OMITTED]

For children, the results suggest that as more time is spent together in semi-structured small-group art activities, the frequency of both "positive initiations" and "positive responses" naturally increase, regardless of teacher behavior (see Figure 2). However, the frequency of "positive initiations" and "positive responses" for the children in the Reactive group was significantly higher than that of children in the Comparison group during Intervention, which suggests that the verbal praise provided by the Reactive teacher was effective. This difference between Reactive and Comparison groups for "positive initiations" continued into Follow-Up. During Intervention, the children in the Comparison group exhibited significantly more "negative initiations" than those in the Reactive Group; however, the frequency of behavior was so small that it is of little or no practical significance.

The children in the Proactive group received instruction in the following skills: "joining in," "sharing," "waiting your turn," and "asking someone to play." Yet, the training did not result in measurable differences in the target behaviors of "positive initiations" or "positive responses." It was possible that the 5-minute training sessions were too short or that two training sessions per skill were simply not enough training to result in behavior change during the art activity. Prior studies in which the same or similar skills were taught incorporated longer training sessions and intervention phases (Odom et al., 1999; Spohn et al., 1999). It was also possible that the training was too abstract and that the children would have been more successful in using the skills had they been taught in the actual setting. For example, one step taught to children to help them learn the skill of "waiting your turn" was the use of self-talk, in which the child would say to him- or herself, "It's hard to wait, but I can do it." In at least one instance, a child walked up to a peer and repeated that phrase out loud to the peer, which may have indicated that the child really did not understand the intended use of self-talk.

As noted above, the training did not result in measurable differences in target behavior; however, it is of interest to note that while the children in the Proactive group did not differ significantly from those in the Comparison group, neither did they differ significantly from those in the Reactive group, whose behavior did differ from those in the Comparison group. It is possible that the Reactive training did have an effect, but that the Intervention phase was too brief for those results to become apparent.

Statistical analyses of the behaviors of children with disabilities were not conducted for the following reasons: the small group size and the absence of one student from 5 of 8 days of intervention; in addition, a visual inspection of the data revealed that the differences noted in frequency of behavior could generally be attributed to one child in the group exhibiting an unusually high number of behaviors on only one day of the phase. For example, during Phase II (Intervention), one of the two students with disabilities in the Proactive group exhibited 11 positive initiations on one day, while on the other days of intervention, the student exhibited 1, 2, or 0 positive initiations. The visual inspection of the data did not suggest that there were any significant changes attributable to teacher behavior.

It is interesting that the children with disabilities tended to exhibit more "positive initiations" and "positive responses" than "negative initiations" or "negative responses" across all phases of the study. It appeared that, at least for these children, social skill interventions would need to be more intensive or of a greater duration to result in increased skill performance. Still, data do suggest that the inclusion of children with disabilities in the groups did not have a negative impact on the social skill performance of children without disabilities.

Taken together, these findings suggest that children with disabilities can be included in the typical early childhood classroom without negatively impacting the behavior of children without disabilities, but they also require more intensive social skill intervention to change behavior than do their typically developing peers. The data also suggest that social skill intervention may be beneficial for children without disabilities, as previously noted by Lau et al. (2005). Further research is needed to determine whether there are components of social skill training programs that will be readily implemented by preschool teachers and result in positive outcomes for young children with disabilities.

References

Baumgart, D., Filler, J., & Askvig, B. (1991). Perceived importance of social skills: A survey of teachers, parents and other professionals. Journal of Special Education, 25, 236-252.

Bredekamp, S., & Copple, C. (Eds.). (1997). Developmentally appropriate practice in early childhood programs (Rev. ed.). Washington, DC: National Association for the Education of Young Children.

Brown, W. H., Odom, S. L., & Conroy, M. A. (2001). An intervention hierarchy for promoting young children's peer interactions in natural environments. Topics in Early Childhood Special Education, 21, 162-175.

Buysse, V., & Bailey, D. B. (1993). Behavioral and developmental outcomes in young children with disabilities in integrated and segregated settings: A review of comparative studies. The Journal of Special Education, 26, 434-461.

Fantuzzo, J., Sekino, Y., & Cohen, H. L. (2004). An examination of the contributions of interactive peer play to salient classroom competencies for urban Head Start children. Psychology in the Schools, 41, 323-336.

Filler, J., & Xu, Y. (2006). Including children with disabilities in early childhood education programs: Individualizing developmentally appropriate practices. Childhood Education, 83, 92-98.

Filler, J. W., Jr. (1976). Modifying maternal teaching styles: Effects of task arrangement on the Match-to-Sample performance of retarded preschool age children. American Journal of Mental Deficiency, 80, 602-612.

Filler, J. W., Jr., Hecimovic, A., & Blue, S. (1978). An analysis of the effectiveness of a preservice workshop for educators of severely handicapped young students. Journal of the Association for the Severely Handicapped, 3, 173-178.

Gresham, F. M., Sugai, G., & Horner, R. H. (2001). Interpreting outcomes of social skills training for students with high-incidence disabilities. Exceptional Children, 67, 331-344.

Guralnick, M. J. (1990). Social competence and early intervention. Journal of Early Intervention, 14, 3-14.

Guralnick, M. J., & Groom, J. M. (1988). Peer interactions in mainstreamed and specialized classrooms: A comparative analysis. Exceptional Children, 54, 415-425.

Guralnick, M. J., Connor, R. T., Hammond, M., Gottman, J. M., & Kinnish, K. (1995). Immediate effects of mainstreamed settings on the social interactions and social integration of preschool children. American Journal on Mental Retardation, 100, 359-377.

Hanline, M. F. (1993). Inclusion of preschoolers with profound disabilities: An analysis of children's interactions. The Journal of the Association for Persons with Severe Handicaps, 18, 28-35.

Howes, C., & Phillipsen, L. (1992). Gender and friendship: Relationships within peer groups of young children. Social Development, 1, 230-242.

Hundert, J., & Houghton, A. (1992). Promoting social interaction of children with disabilities in integrated preschools: A failure to generalize. Exceptional Children, 58, 311-320.

Hunt, P., Staub, D., Alwell, M., & Goetz, L. (1994). Achievement by all students within the context of cooperative learning groups. Journal of the Association for Persons with Severe Handicaps, 19, 290-301.

Kohler, F. W., Anthony, L. J., Steighner, S. A., & Hoyson, M. (2001). Teaching social interaction skills in the integrated preschool: An examination of naturalistic tactics. Topics in Early Childhood Special Education, 21, 93-103.

Lau, C., Higgins, K., Gelfer, J., Hong, E., & Miller, S. (2005). The effects of teacher facilitation on the social interactions of young children during computer activities. Topics in Early Childhood Special Education, 25, 208-217.

Mastropieri, M. A., & Scruggs, T. E. (2007). The inclusive classroom: Strategies for instruction (3rd ed.). Upper Saddle River, NJ: Pearson Education.

McConnell, S. R., McEvoy, M. A., & Odom, S. A. (1992). Implementation of social competence interventions in early childhood special education classes: Current practices and future directions. In S. L. Odom, S. R. McConnell, & M. A. McEvoy (Eds.), Social competence of young children with disabilities (pp. 277-306). Baltimore: Paul H. Brookes.

McEvoy, M. A., Odom, S. L., & McConnell, S. R. (1992). Peer social competence intervention for young children with disabilities. In S. L. Odom, S. R. McConnell, & M. A. McEvoy (Eds.), Social competence of young children with disabilities (pp. 113-133). Baltimore: Paul H. Brookes.

McFall, F. (1982). A review and reformulation of the concept of social skills. Behavioral Assessment, 4, 1-33.

McGinnis, E., & Goldstein, A. P. (1984). Skill-streaming in early childhood. Champaign, IL: Research Press.

Odom, S. L., McConnell, S. R., McEvoy, M. A., Peterson, C., Ostrosky, M., Chandler, L. K., et al. (1999). Relative effects of interventions supporting the social competence of young children with disabilities. Topics in Early Childhood Special Education, 19, 75-91.

Parker, J. G., & Asher, S. R. (1987). Peer relations and later personal adjustment: Are low-accepted children at risk? Psychological Bulletin, 102, 357-389.

Peck, C. S., Carlson, P., & Helmstetter, E. (1992). Parent and teacher perceptions of outcomes for typically developing children enrolled in integrated early childhood programs: A statewide survey. Journal of Early Intervention, 16, 53-63.

Sheridan, S. M. (1998). Why don't they like me? Helping your child make and keep friends. Longmont, CO: Sopris West.

Spohn, J. R., Timko, T. C., & Sainato, D. M. (1999). Increasing social interactions of preschool children with disabilities during mealtimes: The effects of an interactive placemat game. Education and Treatment of Children, 22, 1-18.

Vaughn, S., Kim, A., Sloan, C. V. M., Hughes, M. T., Elbaum, B., & Sridhar, D. (2003). Social skills interventions for young children with disabilities: A synthesis of group design studies. Remedial and Special Education, 24, 2-15.

Keith J. Hyatt

Western Washington University 

John W. Filler

University of Nevada, Las Vegas 

Table 1
Descriptive Information on Children With Disabilities

        Reasons for
Group   Eligibility   Initial Referral              IQ range

1       DD            motor, language, social       average
1       Orthopedic    motor                         borderline
2       DD            language, motor, cognitive    borderline
2       DD            language, social, self-help   borderline
3       DD            social                        average
3       DD            social                        average

Note. Group 1 = Proactive, Group 2 = Reactive, Group 3 = Comparison
DD signifies developmental delay