Performance-portfolio assessment : measurement issues of validity and generalizability in a problems-based pre-service elementary mathematics and science methods course /
| Main Author: | |
|---|---|
| Other Authors: | , , |
| Format: | Thesis Book |
| Language: | English |
| Published: |
1993.
|
| Subjects: | |
| Online Access: | ProQuest, Abstract Link to OAKTrust copy |
| Abstract: | A specialized form of assessment, the performance-portfolio, was developed to monitor and assess the students' growth in classroom teaching skills in a preservice integrated mathematics and science elementary methods course. The performance-portfolio was an authentic measure of preservice teachers' abilities to plan, design, execute, and assess field-based teaching problems while reflecting on the adequacy of the solution. Validity and generalizability of the performance-portfolio grading system was examined using statistical procedures associated with discriminant analysis, generalizability theory, percent agreement, and Pearson's correlation. The generalizability of performance-portfolio scores was examined by comparing the percent of agreement of student peer ratings to those of the course instructors and the instructor's ratings to each other. In addition, a generalizability study was used to examine sources of variation in the portfolio scores. The validity of the performance-portfolio system was examined by choosing a set of predictor variables to represent the underlying knowledge, skills, and prior experience being demonstrated in the preparation of the students' performance-portfolios. These variables were used in a discriminant analysis to predict performance-portfolio scores. Another validation procedure involved the comparison of the performance-portfolio scores with the final examination scores using Pearson's correlation. The results of these analyses led to conclusions regarding the validity and generalizability of the performance-portfolio grading system. The analyses indicated a moderately high discrimination of performance-portfolio scores by the predictor variables. In addition, a high degree of generalizability in the scoring of performance-portfolios was indicated. A very low correlation between the performance-portfolio scores and a conventional final exam was evident. The implications of these findings to mathematics and science teacher education are discussed. |
|---|---|
| Item Description: | Vita. "Major subject: Curriculum and Instruction." |
| Physical Description: | viii, 113 leaves : illustrations ; 28 cm |
| Bibliography: | Includes bibliographical references. |