A number of researchers have developed treatment packages to improve social competence in young elementary school age children. Such programs assume that children are deficient in the area of social skills, therefore, by learning appropriate social skills unliked children's social status and resulting prognosis will likely improve. These treatment programs have demonstrated that more positive social behaviors can be learned by the targeted children. Yet, despite behavior improvements within the treatment setting, gains have not consistently generalized across settings or time and low social status is maintained. The purpose of this study was to examine the effectiveness of self-modeling as a social skill training and status improvement technique. Could the use of this technique decrease negative interactions and isolation while increasing positive interactions during the subject's recess play? Also evaluated was whether the treatment and resulting behavioral changes affected the subjects' sociometric status. This research utilized a multiple baseline across subjects design. Subjects were selected by peer nomination from grades one through three in a rural elementary school in western Massachusetts. Three children (2 second grade boys and 1 first grade girl) were selected from those identified as having low social status within their respective grade. The subjects were regular education students and did not exhibit any idiosyncratic behaviors that would set them apart from their peers. Observations and data collection were conducted during morning recess. Observations continued throughout the study and documented decreased rates of negative and isolate behavior and increased rates in positive interactions as a result of the treatment condition for two of the three subjects. Treatment consisted of the targeted children viewing videotapes of themselves playing appropriately with peers during recess. At the end of treatment the peer nomination instrument was again administered to assess whether changes in status accompanied the behavioral changes. Two subjects improved their rates of positive interaction and one of those two also significantly improved her social status. Results for the third subject are less clear. A trend toward the positive is evident but the study was ended before any clear pattern was established.

Experience, buttressed by a school survey, attested to the presence of homeless children in our schools. These children evince special needs which demand attention, or we may suffer their loss from the school systems, and pay a staggering social cost, later. The problem is that their special needs cannot be met if these children are not identified as homeless, and rarely do they self-identify. The dissertation develops a screening test for the identification of the homeless children in our schools in order to treat their special needs. After a pre-test survey, the test was piloted at an inner-city school, refined, and administered as circumstances suggested. The results confirmed the existence of homeless children with special needs in the schools. An increased ability to identify these children, coupled with enhanced teacher training and curriculum development, can help retain and treat these children within the educational process--and, it is hoped, avoid the social cost.