Abstract
Summary. Research on allocation at 11+ tends to ignore some aspects of the situation. The concept of failure in the grammar school is too imprecise. When correlations are made more attention should be paid to the distribution of movement between the variables. Prediction research should take account of the difference between the performance of boys and girls in the grammar school, and between girls', boys' and mixed schools. Low correlations may be partly due to bad educational guidance. Streaming in the grammar school leads to a crystallisation of differences and an artificially high correlation. Prediction research might investigate the possible existence of different subject arrays for different social classes.

This publication has 0 references indexed in Scilit: