In lieu of an abstract, here is a brief excerpt of the content:

The Review of Higher Education Fall 1980, Volume 4, No. 1 Pages 13 to 24 Copyright © 1980 Association for the Study of Higher Education All Rights Reserved THE INFLUENCE OF STUDENTS’ AFFECTIVE ENTRY ON INSTRUCTOR AND COURSE EVALUATIONS John C. Ory* Research on student ratings of instruction has demonstrated that ratings based on responses to objective items may be biased indicators of instructor and course quality. A general concern of instructors is that ratings are greatly influenced by conditions or circumstances that occur independent of the classroom experience, influences that are often out of the control of the instructor being evaluated. For example, research has shown that student ratings are influenced, often in complex ways, by class size, elective-required status of the course, expected grade and potential use of the ratings (Feldman, 1976, 1978). To date, most of the studies examining biasing influences on student ratings have examined the impact of course and student demographic factors (Costin, Greenough and Menges, 1971; Centra and Creech, 1976; Marsh, 1978). Few studies have investigated biasing influences brought to a course in the way of pre-course expectations or the “ affective entry level” of the students. Granzin * Coordinator of Examination Services, Office of Instructional Resources, Univer­ sity of Illinois at Urbana—Champaign. 13 14 The Review of Higher Education and Painter (1973) examined the relationship between course ratings and the static and dynamic (i.e., reflecting change over time) attitudes of the students as measured prior to the course and during the course, respectively. These static and dynamic attitudes were regressed on three post-course ratings. However, the dynamic influences analyzed in their study were obtained by subtracting pre-course measurement from corresponding post-course meas­ urements. As Chermesh (1977) commented, “ Their discussion and design overlooked many of the statistical artifacts that plague ‘change scores’ such as regression effects, floor and ceiling effects, and the well known unreliability of such measures” (p. 291). Chermesh avoided the use of change scores in his study of student affect and course ratings. However, his two-step causal analysis examined attitudinal assessments collected during the middle and end of the academic year, thus, ignoring the pre-course expectations of the students. Chermesh concluded that his study could have been improved by the addition of the pre-course measurement. In another study, Painter and Granshin (1972) collected student’s grade expectations on the first and last day of class to examine the strength of relationship between change in course evaluation and change in student perception of performance. They concluded that “ students who revise grade expectations significantly more often shifted their evaluation in the same direction (higher grade expectations with higher course ratings)” (p.78). Given these results the question could also be raised whether or not revisions of student attitude toward the instructor and the course are also associated with changes in course evaluations. Or, what is the effect of student attitude change on objective ratings of instruction? The intent of the study was to examine possible biasing influences on student ratings of instruction brought about by the pre-course expectations of the students. The study first attempted to identify factors which create or affect the pre-course expectation of students. Student and course demographic variables were analyzed as possible determiners of affective entry. Second, affective entry data was compared to course and student demographic information in terms of their relative influence on student ratings of instruc­ tion. The question addressed was whether or not course evaluations are more influenced by the expectations of the student than by the background of the student or the circumstances of the course. Finally, the validity of objective student ratings of instruction was chal­ lenged through a comparison of rating bias due to out-of-class factors (e.g., pre-course affect and demographic information) versus in-class factors (e.g., attitude change due to instruction). Affective entry and exit data were used to assess student attitude change and to test the hypothesis that valid measures of instructional quality are more influenced by the instructional process (in-class factors) than by pre-course student expectations or demographic variables (out-of-class factors). Ory—The Influence of Students’ Affective Entry 15 Method Instruments...

pdf

Share