Studies in the Reliability and Validity of a Community College Student Rating of Instruction Questionnaire

Publisher Website
Google Scholar
Abstract
A 13-item Student Description of Instruction Questionnaire (SDI), locally developed but modeled somewhat after the University of California at Davis instrument, was subjected to a series of reliability, stability, and validity studies based on data collected over a period of four semesters, from spring 1973 to fall 1974. The validity studies concentrated on examining the degree to which several student and instructor “conditions” were associated with student ratings of instruction. The student-related conditions examined were: expected grade in course, class size, lower-upper class level. The instructor characteristics were: age, sex, tenure status, professional rank, and highest degree earned. A stepwise multiple regression analysis, using all five instructor characteristics and the interactions among them as independent variables, was carried out in four stages to tease out maximum predictability from a minimum of predictor terms. The total amount of criterion variance accounted for by the five instructor-related variables and their interactions was too small to be of any practical consequence, and the remaining conditions also showed little or no relationship to student ratings. With regard to reliability—stability, six studies were conducted: odd—even internal consistency, test—retest reliability of instructor item patterns, test—retest item consistency, instructor item pattern stability over one semester, instructor inter-section correlation for two sections of the same course, and stability of instructor mean ratings over one year. Indices suggested moderate to good reliability and stability, depending on department.

This publication has 0 references indexed in Scilit: