In order to clarify the relationships between school achievement variances, the following studies were carried out.
Study I
Problem. The purpose of this study was to examine upper level and lower level relationships between seven subjects (Japanese Language, Social Studies, Arithmetic, Science, Music, Art & Handicraft, Physical Education).
Method. 143 pupils in the fourth grade elementary school comprised the sample. Principal Component Analysis (PCA) and Cluster Analysis were used to treat the data from this sample.
Result. The results of this study are summarized as follows:
1) One principal component was found to be statistically significant by PCA. It is composed of six subjects. They are Japanese Language, Social Studies, Arithmetic, Science, Music, Art & Handicraft.
2) By Cluster Analysis, two clusters were found. Two subjects made up each cluster -- Japanese Language and Arithmetic for one, Social Studies and Science for the other.
Study II
Problem. Study II had two purposes. One was to examine lower level relationships between ten subjects (Modern Japanese, Classics, Geography, Mathematics, Biology, Earth Science, Physical Education, Health, Music, English). The other was to determine which subject would increase the predictability of scores of biology or earth science.
Method. 401 first year students in senior high school made up the sample. To analize the data of this sample, Cluster Analysis and Multiple Regression Analysis were employed.
Result. The following were the results of this study:
1) Two clusters were found by Cluster Analysis. Three subjects comprised each cluster. Modern Japanese, Classics and English made up one cluster. Biology, earth science and mathematics formed another cluster.
2) With biology or earth science as criterion variables, the magnitude of standard regression coefficient for both mathematics and geography were greater than those of the other subjects.
Considering these results, we may conclude two points.
1. Subject clustering in the fourth grade elementary school is different from that of the first year high school.
2. Achievement scores in mathematics and geography are valuable for predicting performance in biology or earth science.