Open AccessMEASURING INTRA‐INDIVIDUAL VARIABILITY WITHIN ONE TESTING 1
Author(s) -
Raine Walter J.,
Hills John R.
Publication year - 1957
Publication title -
ets research bulletin series
Language(s) - English
Resource type - Journals
eISSN - 2333-8504
pISSN - 0424-6144
DOI - 10.1002/j.2333-8504.1957.tb00076.x
Subject(s) - test (biology) , homogeneous , psychology , sentence , correlation , variation (astronomy) , variance (accounting) , sentence completion tests , statistics , social psychology , mathematics , computer science , natural language processing , combinatorics , paleontology , physics , geometry , accounting , astrophysics , business , biology
ABSTRACT This study attempts to answer two questions: (1) When people take the same Sentence Completion test twice within a few hours will they display individual differences in intra‐individual response variability? (2) If individual differences in this characteristic appear, will they be related to variance in intra‐individual response variability as measured by comparing a person's responses to homogeneous items presented within a single test? A twenty‐five item Sentence Completion test was modified by adding fifteen items. Each additional item was identical with one of the first twenty‐five items except that personal references were made impersonal and general. The test was given to a group of entering law school students at the beginning and end of a three‐hour testing period. It was found that intra‐individual response variability does appear with this short period between the two administrations or the test. Further, the score based on similar responses to homogeneous items within, the first administration correlated .61 with the score based on similar responses to identical items answered in two administrations separated by a several hour period. Corrected for attenuation in both scores the correlation is .90. These findings should facilitate and encourage study of the ramifications of the concept of intra‐individual response variability.