ERIC Number: ED343959
Record Type: Non-Journal
Publication Date: 1992
Using Restricted Factor Analysis in Test Construction.
Oort, Frans J.
The use of restrictive factor analysis (RFA) is illustrated with data from the field of educational research. An attitude scale measuring the degree to which students feel they are accepted by their classmates is constructed in accordance with the Theory of Violators. The main idea of the Theory of Violators is that most problems associated with measuring psychological constructs can be reduced to the question of unidimensionality. The theory consists of three components: (1) a formal base of definitions of item purity and unidimensionality that introduces the concept of violator, a variable with respect to which the test item is biased; (2) a typology of violators; and (3) instruction on how to construct a scale that is free of item bias, unidimensional, and efficient. As a first step, test items were selected from an existing measure, the School Attitudes Questionnaire of H. C. M. Voorst (1990) administered in the Netherlands to 915 students aged nine to 12 years. Item bias detection was then performed through statistical analysis. After eliminating biased items, the remaining items were tested for unidimensionality through factor analysis. Inefficient items were then detected through RFA. This exercise demonstrates that RFA offers a single approach to various measurement problems. There is an eight-item list of references. (SLD)
Publication Type: Reports - Evaluative
Education Level: N/A
Authoring Institution: N/A
Identifiers - Location: Netherlands