| This dissertation addressed a current interest and need to improve evaluation practices in ESEA Title VII bilingual education programs.;A survey questionnaire regarding technical assistance needs in conducting Title VII evaluations was also sent to 128 Title VII Bilingual Project Directors. Findings indicated common practices in California bilingual program evaluations and technical assistance needs in conducting evaluations.;Evaluation designs frequently emphasized an evaluation based on instructional objectives in relation to student achievement. There was a need for more coverage on data collection techniques, timelines for evaluation activities, descriptions of the tests and instruments to be utilized, proposed data analyses and definition of evaluation resources in terms of funds and staffing.;Evaluation reports focused on student achievement data and included survey questionnaire results regarding staff development and community involvement in some cases. Student achievement data was frequently reported by grade only without separating non-English speakers' scores from fluent-English speakers' scores. Also, the number of students at various grade levels was not always apparent in averaged scores. Raw scores, percentiles, and grade-equivalent scores were used most frequently. Reports needed to include a more in-depth analysis of test scores, stronger conclusions and specific recommendations that could be used in future planning and for program revisions.;The study examined existing efforts in bilingual program evaluations through a meta-evaluation of a sample of ESEA Title VII projects funded in 1979-80 that were representative of five regions of California. This meta-evaluation was done by examining the evaluation designs as stated in project proposals and the actual implementation of evaluation activities in the projects' evaluation reports that were submitted to the U.S. Department of Education in Washington, D.C.;Some projects did, however, document the extent to which they had met their program objectives and had moved toward record-keeping systems and data analyses of test scores over time and in relation to such confounding variables as attendance, time in program, and language proficiency.;Survey results from 93 of the 128 Title VII Directors in California indicated that their technical assistance needs for conducting Title VII evaluations involved measuring the affective domain, using evaluation information effectively, developing in-house evaluation capabilities and analyzing data.;The meta-evaluation results of this study indicated that Title VII projects still need to improve their evaluation designs and reports before evaluation reports can truly be considered as data bases for future meta-analyses studies or as statements on the effectiveness of bilingual education. |
