View report page

Only children are shown separately because order of position in family does not arise
in their case. The proportion of only children in the survey was 12 per cent. for boys and
10 per cent. for girls; for both sexes these percentages were higher for the defects of stammer
(14 per cent. boys, 16 per cent. girls) and dysarthria (23 per cent. boys, 20 per cent. girls).
In a self-selected sample of families having children with a specific defect, there will
invariably be an over-representation of the larger families. If, for example, the incidence
of the defect was ten per cent., every ten-child family would be statistically likely to be
represented but only half the five-child ones and one-fifth of the two-child families. Hence,
if it is desired to see whether the position in the family has any bearing on the presence or
absence of a particular defect, it is difficult to arrive at a population against which to test
such a hypothesis. Practical alternatives are to assume that in, say, five-child families the
theoretical distribution of defective children would be 20 per cent. for each position; or
to relate the incidence to the numbers of children in each position to the numbers of children
contained in all the families in the sample. There are complications with either course—for
example, bearing in mind that this was a survey of schoolchildren, there is the risk that,
in the larger families in particular, some of the children would be under or over the school
age range and might not have been included. Fortunately, whichever method is used the
results are of the same order and line (b) shows the children with defects (line (a)) as a
percentage of the total children in the families occupying the respective positions.
157