ERIC Number: ED177222
Record Type: RIE
Publication Date: 1979-Apr
Reference Count: 0
A Comparison of Two Student Instructional Rating Forms Utilizing High-Inference Versus Moderate Inference Items.
Wilson, Pamela W.
Two types of items used in student evaluations of college teaching were compared: high-inference items, which require considerable inferring from what is seen or heard in the classroom to labelling of teacher behavior; and moderate-inference items, such as "teacher listens carefully." Two instruments were administered to random halves of five undergraduate marketing classes; one consisted of 20 high-inference items that had consistently loaded on five factors, and the second contained eight high-inference and 10-20 moderate-inference items. The second instrument contained items that were specific to the individual class situations. Item variability, between classes and within classes, was examined. Results indicated that there was no significant difference between the general scale (the Student Instructional Rating System) and the class-specific scale. Although the data were not conclusive, it was suggested that moderate-inference items were more accurate for evaluating instruction than high-inference items. (Analyses of students' satisfaction with the class-specific questionnaire and the class means for each set of items are included, as well as six different rating forms). (GDC)
Publication Type: Speeches/Meeting Papers; Reports - Research; Tests/Questionnaires
Education Level: N/A
Authoring Institution: N/A
Identifiers: Inference; Student Instructional Rating System
Note: Paper presented at the Annual Meeting of the American Educational Research Association (63rd, San Francisco, California, April 8-12, 1979)