ERIC Number: ED069738
Record Type: Non-Journal
Publication Date: 1972-Aug
Reference Count: N/A
Effects of Empirical Option Weighting on Reliability and Validity of the GRE.
Reilly, Richard R.; Jackson, Rex
Item options of shortened forms of the Graduate Record Examination Verbal and Quantitative tests were empirically weighted by two variants of a method originally attributed to Guttman. The first method assigned to each option of an item the mean standard score on the remaining items of all subjects choosing that option. The second procedure assigned the mean score on a parallel form of all persons choosing the option. When compared with formula scores, it was found that scores generated with the empirical weights were more reliable but less valid when correlated with undergraduate grade-point average (GPA). Test homogeneity was increased through empirical option weighting, and factor analysis revealed large increases in variance accounted for by the first factor. Examination of the actual weights assigned to each option revealed that the weight for omit in most cases differed considerably from the weight which would be assigned under the usual formula score assumptions. It was suggested that the weighting procedures used tended to capitalize on omitting behavior which, although a highly reliable tendency, may actually be negatively related to the GPA criterion used. (Author)
Publication Type: N/A
Education Level: N/A
Authoring Institution: Educational Testing Service, Princeton, NJ.
Identifiers - Assessments and Surveys: Graduate Record Examinations