測定指標としての社会性偏差値の検討及びその適用による試論

概要

論文の詳細を見る
The present study was primarily designed to investigate on "the standard score of sociality" devised as a sociometric index based upon the distribution of sociality. Secondly, to find out the relationships among the school records, intelligence and sociality indicated by this index. 1. The standard score of sociality (S.S.S.) : The scores of usual indices generally indicate only the relative heights of sociality. A index indicated by the standard score was devised, as it is convenient in practical use for the score of index to be ranked on the standardized scale. An individual sociality was defined as social status determined by choice-rejection responses from members of class. Accordingly, Subjects were instructed to check responses for all members (n) one another on the 5grades scale under a certain criterion. It was all but conformed to the usual indices. It is assumed that the scores on a valuation scale, being the basis of personal feeling, is normally distributed : each grade (i), in case of arrangement, is represented by z-score transformed from summations of checked frequency (f) at the grade. An individual score is generally indicated as follows : X =Σz_if_i(n-1) Assuming the sociality is normally distributed, X is replaced with S.S.S. indicated by Z score, So,S.S.S = 50+10(X-X^^-)/S.D. the normality was confirmed by X^2 test in 209 pupils of a secondary school. The reliability coefficient, by the retest method, was 0.97 while it was 0.90 in case of the split half one. 2. The relations among school records, intelligence and sociality : Subjects were the pupils described in 1. At first, the correlation and partial correlation coefficients among the school records (1), intelligence (2) and sociality (3), were calculated. Also each score of the former two elements was indicated by the standard score. Results are shown as follows : Correlations : γ_<12>=0.62 γ_<l3>=0.60, γ_<23>=0.41 partial correlations : γ_<l2・3>=0.51, γ_<13・2>=0.48, γ_<23・1>=0.07 They are all significate at 1% level except γ_<23・1> These results suggest that the correlation between the sociality and school records is relatively high and, more ever, that the former is naturally independent of the intelligence. In other words, the sociality relates to the factors of being the over- or under- achiever Over- and under achievers were, therefore, selected out under a criterion. In regards to both groups, the mentioned deduction was confirmed at 1% level of significance. Secondly, to investigate the developmental transition, 4 groups, selected out of primary, secondary and high schools, were examined with the same procedure. The same tendency was found also in all groups but a group of high schools.
日本教育心理学会の論文
1958-06-25

著者

寺岡隆
北海道大学

測定指標としての社会性偏差値の検討及びその適用による試論

スポンサーリンク

概要

著者

関連論文

スポンサーリンク