ソシオメトリーにおけるテクニックに関する一検討

概要

論文の詳細を見る
PURPOSE There have been not a few studies of the relationship among various sociometric techniques, though the methods employed are fairly different. Here in Japan, we can hardly find an intensive as well as extensive research of this kind. This study is an attempt to explore: (1) the intercorrelations between pupils' sociometric status scores obtained in a rating scale , on one hand , and those obtained by use of several sociometric tests with choice-rejection levels of unlimited, five limit, four limit, three limit, two limit and one limit, on the other, and, (2) the intercorrelations among pupils' sociometric status scores obtained in those various sociometric tests.PROCEDURE In the present study, several sociometric tests, with one criterion and different numbers of choice and rejection, were administered to the pupils in six public school classrooms, three each both in the third and fifth grade, on six different occasions in February, 1962. Subjects were requested to choose the classmates with whom they most and least preferred to study. At the end of the test series, every pupil was asked to rate all the rest of the classmates on the 5-point scale where 5 was given to those with whom he most preferred to study, and 1 to those with whom he disliked to do. Sociometric status was determined by SSS (PVR+MPV-NVR-MNV) in the sociometric tests, and simply by the total points he received from other members of the class in the rating. When the sociometric status had been determined for each test and scale, the scores were all converted into the standard scores. Pearson product-moment correlation coefficients were calculated to evaluate the interrelationships among the scores obtained by use of different sociometric techniques.RESULTS AND CONCLUSIONS Main findings on the basis of the obtained data are as follows: (1) As shown in Table 3, the intercorrelations between the status scores obtained in the rating and those in other scoiometric tests are, in general, fairly high and statistically very significant. The correlation coefficients ranged from .652 to .888 with a median of .83 approximately in the third grade level, and, in the fifth grade, they ranged from .485 to. 86 with a median of .75 approximately. Inspection of Table 4 will reveal that ,as far as this sample of pupils is concerned, there seems to exist a tendency for the coefficients in the fifth grade to be lower apparently, and this data, grossly interpreted, seems to suggest that there is a trend for the coefficients to decline as the choice and rejection level decreases. (2) As shown in Table 5, 6, 7, 8, 9, 10 and 11, the intercorrelations between the sociometric status scores obtained in the test with unlimited choice and rejection level on the one hand, and those obtained in tests with various limited choice and rejection levels on the other, are also very high and statistically significant. The correlation coefficients ranged from .614 to .937 with a median of .82 approximately in the third grade level, and, they ranged from .502 to .909 with a median of .77. Grossly interpreted, Table 11 seems to indicate nearly the same trend as recognized in Table 4. (3) Comparison of Table 4 with Table 11 seems to suggest that, in the fifth grade level, the correlation coefficients obtained between the unlimited sociometric technique and several other limited ones are consistently higher, apparently at least, than those between the rating and limited sociometric techniques. This trend ,however, could hardly be observed in the third grade level. (4) Generally speaking, the intercorrelations obtained between the two sociometric techniques where choice and rejection level is close to each other were found to be higher than those between the two measures where the difference of the level is rather great. Correlation coefficients between the sociometric status scores obtained with one, and very often two, limit of choice and rejection level, and those with larger limit, were found to be lower, as compared with other coefficients.A discussion of the results leads to the tentative conclusion that we need not employ the rating method nor the unlimited sociometric test at the vast expense of time and energy only to determine the sociometric status of pupils at large in the classrooms. A sociometric test with three or more choice and rejection level will be satisfactory as a technique to evaluate pupils' sociometric status.
奈良学芸大学の論文

ソシオメトリーにおけるテクニックに関する一検討

スポンサーリンク

概要

奈良学芸大学 | 論文

スポンサーリンク