This study was conducted to develop a statistical model of the influence of
course characteristics on student ratings of instruction. Unique to this study was
the inclusion of the student participation rate as a variable. Other course
characteristics studied were: expected grade, attendance, grade point average,
sex of students, reason for taking the course, academic field, student level,
course level, major, and class size. Stepwise multiple regression analysis was
employed using two measures of global ratings of instruction as criterion
variables: 1) a single item rating the overall "favorable impression" the instructor
had on the students (global), and 2) the average of twelve items rating the
instructor's effectiveness (overall). Course averages or proportions were
calculated for each variable. A model was developed using 1989 Fall Term data
(952 courses), and tested using 1990 Fall Term data (1,052 courses) from
Oregon State University. Results of the regression analysis showed that
approximately 20% of the variance found in the equations could be explained by
five variables: expected grade, participation rate, attendance, sex, and grade
point average. Expected grade was by far the best predictor accounting for 13%
to 14% of the variance (at least r=.36 with criterion variables). Participation rate
accounted for as much as 3% of the variance (roughly r=.2) and contributed
significantly to all four equations. A correlation of -.42 was found for participation
rate and class size. Results suggest participation rate has a relationship with
student ratings of instruction and further study is warranted. Reason and
academic field also explained small significant amounts of the variance in at
least one equation each. The results were determined to be reproducible;
equations generated with the 1990 test data were very similar to the 1989 data.
Very high correlations were found between the global and overall criterion
variables (r> .95); the single item variable produced the same results as the
average of twelve items. A significant correlation of more than .5 was found
between academic field and sex of student. The 80% unexplained variance is
discussed. / Graduation date: 1993
| Identifer | oai:union.ndltd.org:ORGSU/oai:ir.library.oregonstate.edu:1957/27987 |
| Date | 26 April 1993 |
| Creators | Frost, Jacquelyn Lee |
| Contributors | Courtney, E. Wayne |
| Source Sets | Oregon State University |
| Language | en_US |
| Detected Language | English |
| Type | Thesis/Dissertation |
Page generated in 0.0023 seconds