ERIC Number: ED064378
Record Type: RIE
Publication Date: 1968
Reference Count: 0
Test Homogeneity and Response Stability.
Tyler, Thomas A.
It was hypothesized that unstable subject-item interactions on personality scales would be associated with small psychological distances between subjects and items. It was further hypothesized that this relationship would be more demonstrable when the psychological test was more homogeneous. To clarify the rationale of the first hypothesis, an explanation was offered for the commonly observed greater stability of extreme items and extreme subjects. Both of the hypotheses were supported. For the second hypothesis, however, it was demonstrated that a single index of test homogeneity was only partially adequate. The separate contributions of items and of subjects to total test homogeneity had to be considered for the most meaningful interpretation of the results. Implications of the instability of responses associated with small subject-item distances were drawn for those operational definitions of subject variability and subject acquiescence which are based on personality test data. A suggestion is made for controlling acquiescence in personality tests. (Author)
Publication Type: N/A
Education Level: N/A
Sponsor: National Science Foundation, Washington, DC.
Authoring Institution: Chicago Univ., IL. Dept. of Psychology.
Note: Based on a doctoral dissertation submitted to the Department of Psychology, University of Chicago