SCIENCE teachers of to-day have the twofold task of instructing the specialist and creating in the average citizen an interest in the scope and methods of modern science. The vast majority of their pupils falls within the latter category and the object of this study was to help teachers to understand which factors are associated with interest in science and whether there are any factors which tend t o be associated with indifference towards science. Specific aspects of this problem have received attention throughout the century, but it is only during the last decade that researches such as those of Terman (1) and Lovell and White (2) have shown a trend towards global investigations concerning the relationships of the interests of individuals to as many of the elements of their psychological and sociological environments as it seemed reasonable to postulate. In the main, however, these were not concerned with school children, but with students and adults. Nevertheless, recent studies of the influence of such external factors as the museum (see Brooks andvernon, (3), suggest that the field of influence which either stimulates or aborts the growth of interest of children in science is probably enormously wide. The main part of the study to be described concerned itself with the statistical relationships between interest in science and each of forty-seven variables representing factors in school and home. The selection of the latter was made by combing the literature to collect evidence on any factors which had been found to be associated with interest and also by considering possibilities such as attitudes of pupils and teachers to each other, parental attitudes and peer relationships which arose out of the general trend of psychological findings of the present time.The sample of testees consisted of 150 pupils from a large co-educational bilateral school on the fringe of East London, in an area containing both urban and semi-rural districts. The sample contained roughly equal numbers of first and third year pupils, of boys and girls, and also roughly half the representatives of each age group were from the technical (selected) stream and the rest from the modern (unselected) stream. It was thought to be reasonably representative of pupils from English secondary schools except that it was biased towards the lower middle and middle classes and was deficient in children of working class origin.The objective method of measuring interest through ' information tests ', used by Peel and his fellow workers5 was not considered useful in this experiment because it requires the use of ordinary words or homonyms which do not have established stereotyped definitions, and these would have been difficult t o find in connection with interest in school science. Instead, use was made of subjective interest tests scored by absolute summation without reference to norms outside the group. These took the form of presenting the testees with statements and instructions such as ' I Give four votes to those which are most like what you woul...

Summary. Current views on the problem of correction for undeserved scores likely to be obtained by testees on true‐false achievement tests, given with either ‘guess’ or ‘do not guess’ instructions, are summarised. The desirability of a new approach is discussed, and an investigation which led to the conclusion that ‘S=R —·5W,’ may be an appropriate formula for ‘do not guess’ tests is reported. The experimental results are shown to conform to a simple mathematical theory, and from the latter a conjecture is made about the formula appropriate to ‘guess’ tests.

No abstract