Published online by Cambridge University Press: 27 September 2011
This paper attempts a new look at an old problem. Throughout this century there have been many reports showing that certain characteristics of family structure are associated with the individual's performance in evaluative situations, be these IQ tests, tests of achievement, school and university examinations and even occupational success (for an excellent summary, see Anastasi, 1956). It is well known, for example, that children from large families tend not to do so well in such situations as children from small families, and that this phenomenon appears to be, in some degree, independent of socio-economic differences. This we can illustrate with our own data (Text-fig. 1) in which we see a steady decline in score on a nonverbal group test of intelligence as the size of the family increases. Less clear is whether other features of family composition, such as the spacing between siblings, the sex composition of the sibship and the ordinal position of the individual within the sibship, also affect achievement. (There is no lack of reports but, as we shall show, the evidence they provide is conflicting.)