STREAMING IN THE PRIMARY SCHOOL

Professor Lynn suggests that two investigations carried out by Dr. Barker Lunn have yielded inconsistent results: the first (cross‐sectional) study showed that the attainment of children in streamed schools was superior and the second (longitudinal) study showed that streamed children did better on one set of tests and unstreamed children did better on a parallel set of tests. He suggests that the children in the first study were crudely matched for social class, and that both studies fail to solve the problem of streaming because of possible statistical distortions in the analysis.

Dr. Lunn replies that the apparently inconsistent results were due to the differences in the aims of the two stages. The cross‐sectional stage, examining absolute measures of attainment, was exploratory‐‐factors such as the individual pupil's social class, and his initial ability were not taken into account. The second or main longitudinal stage was concerned with the progress of one age group of children comparable in terms of sex, social class and ability throughout their junior school careers. In the latter stage the problem of social class was overcome by examining the progress of children who were comparable in this respect.

Dr. Lunn rejects Professor Lynn's statement that streamed children did better on one set of tests and unstreamed children did better on the other as a misrepresentation of fact. Non‐streamed schools in Sample A (working on both tests) tended to do better and streamed schools in Sample B tended to do better, although there were no significant differences between 85 per cent of the comparisons of mean scores.