A measure of children's reasoning about interpersonal relations

UNCG Author/Contributor (non-UNCG co-authors, if there are any, appear on document)
Therry Nash Deal (Creator)
The University of North Carolina at Greensboro (UNCG )
Web Site: http://library.uncg.edu/
Irwin Sperry

Abstract: The problem investigated in this research was the development, pretesting, and validation of a measure of children's reasoning about interpersonal relations. The test was developed for children from six to ten years of age. Reasoning was operationally defined in the study through hypothetical syllogisms which had a content of interpersonal relations. The content of these items was not that drawn from traditional examples in logic, but the content was empirically derived from the conversations of children in the age range under study. During the development of the reasoning test one hundred fifteen items were written. The form of the item was that of the traditional syllogism consisting of major premise, minor premise, and conclusion. Two alternatives were presented to the child as conclusionss The child's task was to indicate the correct conclusion. The research measure was pretested on a group of thirty-six children. Item-test correlations were computed and plotted against mean scores for each of the one hundred fifteen items. Thirty-six items had item-test correlations between .40 and .80 and difficulty levels of three to seven children out of nine passing the item. Thirty items from among these thirty-six items composed the test of reasoning about interpersonal relations. The reasoning test was administered to a population of one hundred six children in grades one through four. The individual interview technique developed with the pretest was used as the standard procedure. Validity was evaluated in terms of correlation and lack of correlation with Primary Mental Abilities, Form 2-4, Revised 1962, and correlation and lack of correlation with teachers' ranks. Product-moment correlations were computed for a 75 x 75 matrix including the variables of odd score, even score, total score on reasoning test, age, grade, sex, five PMA test scores, proportion of items passed on which the child disagreed with the major premise, number of disagreed premises, teachers' ranks, deviations from grade means, thirty item scores on the reasoning test, and thirty item scores for agreement-disagreement with the premises. Certain intercorrelations for 75 variables were studied. Odd-even split-half reliability obtained was .48, corrected to a .65. A correlation of .57 was obtained between the reasoning test and the PMA, Form 2-4. The highest PMA test score and reasoning test score correlations were with the verbal ana number tests where the correlation was .52 in both cases. The correlation between teachers' ranks and deviations from grade means on the reasoning test was -.45. This negative correlation was in the expected direction due to the manner of assigning ranks. The relation between sex and the reasoning test was a negative correlation of -.12, males having lower scores. Age and scores on the reasoning test were correlated by a relationship of .38. The correlation between grade and the reasoning test score was this was higher than the age correlation with the reasoning test. Two minor hypotheses were examined: (1) that there is a positive correlation between age and passing an item, on which the subject disagrees with the major premise and (2) that there is a positive correlation between age and number of disagreed premises. The magnitude of the relation for the first hypothesis was .27, for the second -.01. Inspection of grade means indicated that only between grades one and two was there any noticeable difference in score. The difference was three points. All of the items in the test were positively correlated with the total score; the range of these correlations was from 0.4. to .55. The correlations between age and item scores and grade and item scores were similar. Only six item premises were accepted by fewer than 75 per cent of the children. Two premises were accepted by all of the children. The conclusions were: (1) that the measure of reasoning about interpersonal relations developed in this research had a moderate level of reliability; (2) that the test of reasoning had a moderate degree of validity with the PMA and teachers1 ranks; (3) that higher correlations throughout between grade and other variables, rather than age and other variables, suggested the ability to reason about interpersonal relations was related to experience as well as maturation; (4) that the low positive correlation between age and proportion of items passed, where there was disagreement with the premise, provided some support for the hypothesis of increasing hypothetical reasoning ability with age; and (5) that there was no support for the hypothesis that number of disagreed premises will increase with age.

Additional Information

Language: English
Date: 1965
interpersonal relations in children, reasoning in children
Reasoning in children
Interpersonal relations

Email this document to