ISSN : 1229-067X
The present study tested seven assumptions concerning the nature of various indicts derived from an endorsement set (a set of scale values of all statements endorsed by a subset of subjects endorsing a particular common statement) including a new item index called sensitivity index. A kind of item reliability, this index is defined by the between-subject variance of the endorsement set of an item (attitude statement). Two studies were carried out. which supported all but one of the seven assumptions tested. The results showed that the sensitivity index is independent of the other indices derived from the same item endorsement set and that its use in the selection of attitude statements in Thurstone-type attitude scales results in more reliable attitude scales than does the use of Thurstone's test of irrelevance. It was maintained that Thurstone's test of irrelevance is a confounded index and that the sensitivity index is superior to the test of irrelevance because it is a purer item index and renders itself to a more precise and objective measurement than the test of irrelevance. Contrary to expectations, item sensitivity was unrelated to rated item social desirability, but the latter was found related unexpectedly to item scale position in an inverted U function.
This study examined the relationship between a set size of stimuli and the context effects in impression formation. It was assumed that the context effects tend be larger when the set size of stimuls list decreases, In an attempt to test this assumption, an experiment was performed, manipulating the set size of the stimulus lists and inducing the context effects. The set size of lists was ranged one to four, while the favorableness of the stimulus lists were constant. The context effects were induced by the presentation order of the stimulus series. Subjects were presented with the stimulus lists and rated their liking for the adjective traits under different contexts. Consistent with our prediction, results showed significantly decreases in context effects as the set size of lists increased and the structure of stimulus were differentiated.
The purpose of this paper was to study genetically in the process of being established our perceptual constancy of objects and GegenstandsbewuBtsein and to try to clarify how they interact. Two Korean girls' response tendency and Kinaesthese were observed for two years (Seok-joo) and one and a half year(Deok-joo), respectively. The observed data were analysed in terms of the constitution of object permanence concept (Piaget) and the occurrence of the perception of shape constancy. The relationship between the two were discussed and the intervention of a certain cognitive activity into the process of perception was supported. Theoretical implication was discussed within the framework of Helmholtz's unconscious inference, Piaget's perceptual activity, and Husserl's Konstituion.
The hypothesis advanced by R. R. Scars and his collaborators that restrictive childrearing by parents leads to femininity in children was tested by means of a self-administered child-rearing attitude test given to mothers rather than by means of ratings of interview data as used by Sears and his coworkers, One hundred ninety male and 160 female, 5th and 6th graders were given an M-F test for children, and on the basis of the scores the mothers of the most masculine and the most feminine students were given a child-rearing attitude test measuring four child-rearing dimensions previously isolated from Korean mothers. Analysis of the test scores from 60 mothers of the masculine and 79 mothers of the feminine students showed that a "cold" mother is associated with a feminine boy whereas an "overprotective or domineering" mother is associated with a feminine girl. Permissiveness and emotional involvement were unrelated to child's femininity. It was concluded that at least for mothers data do not support either the primary identification hypothesis or Sears' activity-restriction hypothesis. An unexpected finding, namely that mothers of the boys were more masculine than those of the girls, was explained in terms of greater number of male siblings expected for a male student as against a female student. Evidence was presented which bore out the latter expectation.