A statistical trap associated with family size.

A recent paper published in this journal (Poole & Kuhn, 1973) has reported that ‘a relatively large family appeared to be no impediment to the educational achievements of middle-class children’. This finding, however, is based upon an analysis which fell into a relatively unknown statistical trap. Poole & Kuhn are only the latest in a long line of victims of this trap and some of their predecessors are reviewed below. [ABSTRACT FROM PUBLISHER]