A job-related social skills program for high school students with mild cognitive impairments, using a range of media materials has been implemented in several school districts in Virginia. The program, developed under the auspices of the U.S. Office of Special Education Programs, will be examined in this study for issues of implementation and adoption in public high schools. The use of traditional experimental designs in program evaluation is seriously questioned when complex issues of implementation are involved. These issues are magnified in special education settings where subject assignment, sample size, individualized instruction, and teacher choice of materials are uniquely present. Consequently, a case study approach of four classrooms following the techniques of Miles and Huberman and Yin was used to examine the implementation process. Three major factors were used to organize data collection: teacher understanding of existing curriculum and goals and teaching style, congruence between the existing and new content and decision making related to implementation, and instructional delivery and the way program use actually looked. While all teachers appeared to like and accept the program, they did not want it to alter their existing classroom plans, teaching styles, and personal interaction approaches. Each teacher showed a unique defense of his or her existing educational style, and the new program was adapted to the ongoing classroom situation, rather than vice versa. If faced with decisions between using the new program and fulfilling existing requirements, these teachers chose to reject the program. The great variation in implementation and modification of the program across these sites supports the need for more careful descriptive site by site studies that allow for differences that cannot easily be identified in quasi-experimental designs. / Ph. D.