A Question of Style — A Comment on the Bennett Typology

Abstract

20 In a lengthy review of Neville Bennett's Teaching Styles and Pupil Progress (Bennett, 1976) John Gray and David Satterly have some harsh things to say about the design and analysis of the study (Gray and Satterly, 1976a). So far Bennett has not been afforded the opportunity of a reply and it is not the intention in this comment to take sides in the controversy. On many points at issue, Bennett and Satterly would appear to be operating with different sets of assumptions and the question of the appropriate statistical unit of analysis and the type of analysis to be employed is by no means as clear cut as Gray and Satterly would suggest.1 Nevertheless, one of the points at issue, the composition of the mixed teaching style2 is open to further empirical investigation from which it would appear that Bennett's choice of the category types 3, 4 and 7 to form the mixed style represents about the best compromise available to him. Gray and Satterley's complaint is that a bias towards the informal style has been built into the mixed style. Since the mixed pupils perform as well as the formal pupils on some of the tests, Gray and Satterly's argument would appear to be that part of the contribution to the average mixed scores should be attributed to the informal pupils and that if this were done some of the differences, particularly those on reading, between the formal and informal pupils might disappear. The data for the full typology of teaching styles is provided in an earlier article (Bennett, 1975). For each cluster the distribution of responses for 19 items from the original questionnaire is given. These selected items on which the typology is based were extracted from a factor analysis of the full questionnaire. The figures appear to give a percentage frequency of response for each type so that for the most informal teacher cluster only 30% of the 35 teachers expected pupils to be quiet (item 4) while for the formal group there was unanimity of view on this point between all 36 teachers. Thus these percentage frequencies provide a profile for each of the clusters. It can readily be seen that there are difficulties in labelling each group since although in general the extreme clusters (one and two v eleven and twelve) reflect opposite tendencies, this is not always the case. For example, no teacher from cluster twelve allows a pupil a choice of where to sit (item 1) whereas 77% of those in cluster eleven do so and are therefore closer to the informal teachers in this respect. These discrepancies become more marked as one moves into the middle groupings (five, six and seven). If, as Bennett says, quoting McQuitty (1967), each type defines a category of persons 'wherein everyone in the category is more like every other person in the category than he is like any other person in any other category' then it follows that the 'central profile' of the two informal types should be more highly correlated with each other than with either of the two formal types. Extending this principle to the mixed group (three, four and seven) it follows that if, as Gray and Satterly suggest, this is biased in favour of the informal style it should correlate more highly with informal types (one or two) than with the formal ones (eleven and twelve).