Open AccessPROCEDURES FOR OBTAINING SEPARATE SET AND CONTENT COMPONENTS OF A TEST SCORE 1
Author(s) -
Helmstadter Gerald C.
Publication year - 1956
Publication title -
ets research bulletin series
Language(s) - English
Resource type - Journals
eISSN - 2333-8504
pISSN - 0424-6144
DOI - 10.1002/j.2333-8504.1956.tb00064.x
Subject(s) - set (abstract data type) , content (measure theory) , test (biology) , cronbach's alpha , statistics , test set , mathematics , computer science , natural language processing , psychometrics , mathematical analysis , paleontology , biology , programming language
Cronbach has defined response set as “any tendency causing a person to give different responses to test items than he would when the same content is presented in different form.” The purpose of this study was to develop and compare different procedures for obtaining separate measures of set and content components of a test score. Using two distinct models, several ways of obtaining such scores were developed. Each scoring system not producing results algebraically or linearly equivalent to another was then applied to the same data obtained in a situation in which a response set had occurred. Intercorrelations of all the measures were obtained and used for the comparison. The median intercorrelations among content scores, among set scores and between content and set scores were found to be .82, .96, and .05 respectively. It was therefore concluded that separate measures of set and content could be obtained and that a simplified ratio procedure would produce adequate results for most situations.