社会空間と学力の階層差 [in Japanese] Social Space and Inequality in Educational Performance [in Japanese]
Access this Article
Search this Article
学力に階層差のあることは広く知られているが，マクロな社会の変化とともにそれがどう変容していくかについては，必ずしも明確な展望が描けていない。本稿では，OECD の PISA 調査データ（2009）に多次元階層分析を適用し，国際比較の観点からこの問題に取り組んでいる。まず，生徒の家庭背景について，多重対応分析からブルデュー流の社会空間を構築し，経済発展により階層の多次元化が進むかどうかを検討した。つぎに，社会空間における個人座標を階層変数として利用し，それが PISA テスト得点をどの程度説明するかを吟味した。その結果，〈資本総量〉に対応する第１軸得点が生徒の成績差をよく説明すること，それに〈資本構成〉の違いを反映した第２軸得点を追加すると説明力がさらに高まることが明らかとなった。つぎに，各国におけるそれらの説明力の差異をマクロ水準の回帰分析によって検討した。その結果，第１軸得点の場合は，経済水準の上昇が階層差を縮小する効果をもつものの，平均学校余命が逆に階層差を拡大させる効果をもつことから全体の傾向が曖昧になること，第２軸得点の場合は，教育制度の特徴によらず経済水準の上昇とともに文化的資源の影響力が単調に強くなっていくことが確認された。結局，マクロな社会の変化とともに学力差に対する要因構造の転換が進み，教育達成の階層差は単純には縮小していかないとの結論が導かれた。
While class inequality in educational performance is well known across countries, the direction of its change in social and economic development is less clear. This paper examines international variations in educational inequality by applying multidimensional class analysis to PISA2009 data, of which forty-one countries are used for the analysis. In the first step of the analysis, a Bourdieuian social space for each country was built from variables of the studentsʼ family socio-economic background and investigated as to whether its degree of multi-dimensionality correlated with the level of economic development of each country. In the second step, the relationship between studentsʼ test performances and their class positions was examined. Individual class positions were identified as coordinates in the social space of each country. According to Bourdieuʼs <I>Distinction</I>, the ordinates and the abscissa of social space can be interpreted as corresponding to ʻvolume of capitalʼ and ʻcomposition of capital,ʼ respectively. In almost all countries, especially advanced ones, differences in studentsʼ test performances were explained to a substantial degree by their positions in the social space. In the third step, the differences in explanatory power of social space for test performance among countries were examined by aregression analysis. This was conducted at the country level using GDP and a number of educational indicators as independent variables, including enrolment ratio of preprimary school, pupil-teacher ratio in primary school, first age of selection, and school life expectancy after compulsory education. Dependent variables were standardized partial regression coefficients of the coordinates in the social space of each country, and were derived from the precedent regression analyses of test performance. The main results were as follows: 1)From the analysis of the ordinates, economic development appeared to basically diminish class inequality in test performance. 2) However, the trend towards equality was not clear because prolonged schooling came to emphasize class differences in educational selection or choices and to offset indirectly the basic trend of equality brought about by economic development. 3) On the other hand, from the analysis of the abscissa,cultural resources appeared to become more important with economic development, irrespective of educational changes. 4) In the end, economic development enhanced the degree ofmulti-dimensionality in the stratification system and transformed the factor structures of educational performance, but such changes did not diminish class inequality in educational attainment and failed to reform school characteristics towards equality. Throughout this paper, the usefulness of Bourdieuian multidimensional class analysis for sociological research into education was recognized.
- The Journal of Educational Sociology
The Journal of Educational Sociology 90(0), 101-121, 2012
THE JAPAN SOCIETY OF EDUCATIONAL SOCIOLOGY