The use of students to appraise the performance of college teachers has been questioned by educators because of concerns over potential biasing factors relating to intrinsic characteristics of the students, such as reasons for taking the course, perceptions of progress or success in the course relative to expectations, and perceived course workload. However, the construct validity of questions asked to measure these potentially biasing constructs has not been systematically examined, thus casting a shadow on previously published causal models of the relationships among these factors.
Cross-sample construct validation was conducted on five factors cited in the literature: prior interest in the subject matter, reason for taking the course, strength of motivation to take the course, perceived course workload or difficulty, and estimated relative progress in the course. A sixth factor of global teacher evaluation items was added to verify the relationship between ratings and the other five factors. Exploratory and confirmatory factor analysis showed that, of the five, only the workload/difficulty construct exhibited validity as a distinct construct. Prior interest, reason for taking the course, and strength of motivation were basically all components of a broader construct of desire or drive to take the course. Estimated progress turned out to have two separate components, one positive and student-related (satisfaction) and one negative and teacher-related (dissatisfaction).
| Identifer | oai:union.ndltd.org:RICE/oai:scholarship.rice.edu:1911/15895 |
| Date | January 1985 |
| Creators | ELLIS, RANDI S. |
| Source Sets | Rice University |
| Language | English |
| Detected Language | English |
| Type | Thesis, Text |
| Format | application/pdf |
Page generated in 0.0013 seconds