Open Collections

Student evaluation of teaching : a multilevel analysis

University of British Columbia

Despite problems with methodology and interpretation of results, student evaluations of instruction provide useful data for administrative decisions. This study specified and tested an Hierarchical Linear Model that can be used to ameliorate some of these problems. The study examined the student ratings of 3,689 university courses taught by 260 instructors that were collected over an 11-year period in an education faculty. A longitudinal hierarchical linear model was used to investigate whether individual instructors' scores were stable over time and in a variety of contexts. The model was also used to examine the effects of exogenous course and instructor variables on the scores. Results showed that the effects of the course-level variables class size, course level, percentage offemales in the class and percentage of students taking the course as an elective were significant. Together these variables accounted for approximately 20% of the withininstructor variance and 17% of the between-instructor variance. The effects of the instructorlevel variables rank, sex and number of course taught were not significant, although the sex of the instructor was substantively related to both the average score and the course level. Scores increased significantly over time, but with less reliability. The years of experience of the instructor who taught the course was also significant. The analysis also illustrated the utility of this model for score adjustment and recommendations were made with respect to the use of the scores for summative evaluation.

Thesis/Dissertation

application/pdf

2009-02-17

For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use.

http://hdl.handle.net/2429/4699

Education

University of British Columbia

UBCV

DSpace