ABSTRACT
The analyses concerned four sets of variables: structural, contextual, pupil involvement and pupil social characteristics. A number of problems arise in the analysis of so many variables, but these can be reduced by the application of theoretical perspectives, and by relating the form of the analysis to the kinds of explanations to be made of the results. Data were gathered for individual pupils, but all of the analyses were for aggregates of pupils. An analysis of the mean scores of all pupils in all grammar schools would not necessarily reveal anything about involvement patterns in any individual grammar school. The choice of the school as a basic unit for the analysis of involvement also follows from the method of constructing the scales used in its measurement. In some analyses mean involvement within individual schools was made a dependent variable in relation to the pupils’ age, sex and social class.
