Correlates of Course Ratings

The purpose of this study was to assess the extent to which selected variables are related to scores on the Utah State University Faculty Evaluation Questionnaire (USU FEQ). The variables examined were size of class, level of class, college under whose auspices a class is offered and whether the class is required or elective. In addition, the relationship between total mean score and responses to question 23, a percentage ranking of instructors, was assessed. The fall quarter, 1975, FEQ results were used in this assessment. Employing the same procedures as had been followed in previous quarters, 315 courses were evaluated. Courses at all levels (100-700) and within all colleges of the University were sampled. Through multiple regression analysis, it was found that 87% of the variance in total mean scores was explained by the selected variables. Question 23 was the outstanding contributor; therefore, the analysis was repeated with this variable removed. As a result, 21% of the variance was explained by the remaining variables. Question 23 was highly related to total mean score. Reasonably accurate prediction of instructors ratings can be made from a knowledge of responses to this question. A small negative correlation was found between sample size and total mean score. Instructors of larger classes tended to rate lower than instructors of smaller classes. A small contribution to the explained variance was made by the variable of class size. This contribution has little practical significance. Instructors of different colleges received different average ratings. Those colleges most closely related to high total mean scores were Education, Family Life and Humanities, Arts and Social Sciences. Those colleges receiving lowest ratings were Engineering and Science. Knowledge of the college under whose auspices a course exists improves prediction of instructors ratings on the USU FEQ. Whether a course was required or elective and the level of a course had little relation to the ratings instructors received. The outcomes of this investigation may be limited by any one or a combination of the following: The results of this study were obtained from fall quarter evaluations. Ratings may relate to the quarter in which the course was evaluated. The USU population was the only university used in the sampling. Results, therefore, will be generalizable to USU alone. This study has been concerned with environmental variables, i.e., class size, level, college, required vs. elective. Student, class, or teacher characteristics have not been considered. Significant relationships may exist in these areas. Within University policy, professors are allowed to choose the quarter in which they will be evaluated. Therefore, some volunteer effects may be present in this study. However, extra encouragement to evaluate fall quarter was given to teachers to help control for these effects. From this study of the relationship between selected variables and total mean scores, the following may be concluded; Question 23 could be used to obtain a quick, easy estimate of student ratings of an instructor. The significant differences between ratings of instructors in different colleges receive presents an area of concern. When comparing instructors from different colleges, precautions may need to be taken. Differential norms for the colleges could be considered. The size of an instructor\u27s class is of no significant concern when interpreting FEQ results. The results of this study show no basis for the employment of differential norms for differing class sizes. No basis for considering whether a course is required or elective and level of a course when interpreting questionnaire results is given. From the above summary and conclusions the following recommendations are made; The college under whose auspices an instructor teaches should be considered when interpreting FEQ results. College norm groups should be considered for inclusion on results printouts. The significantly higher ratings received by instructors in the College of Education should be investigated. The following questions arise; A. Do students rate instructors higher if instructors give them higher grades? (Grades received in Education are higher.) B. Are the students in the college more empathetic to their instructors and, therefore, more lenient in their ratings? C. Is the course content less demanding or more interesting so as to receive higher student ratings? D. Are, in fact, the instructors in the College of Education better teachers? If so, what are the characteristics that make them better? These questions and others should be researched. 3. Since relationships may vary according to the particular aspect of teaching performance that the student is asked to rate (Clark & Keller, 1954), a factor analytic study including the selected variables of this study and all of the questions on the USU FEQ may be profitable. 4. Scores on the USU FEQ may be related to the department in which the course is taught. Correlations might be determined for the departments which have a large enough N. 5. Determination of whether students\u27 subjective criteria in rating faculty match the faculty members\u27 goals in teaching may disclose some valuable information about the USU FEQ. 6. The benefits from student evaluation of instruction can only exist to the extent that ratings represent valid appraisals of classroom instruction. Research must be conducted in answer to the question, Is the USU Faculty Evaluation Questionnaire a valid instrument