Published online by Cambridge University Press: 31 July 2008
Frequently it is of interest to compare the distributions of birth intervals for two or more population subgroups. Such analysis can serve as a useful adjunct to conventional studies of differential fertility.
This paper discusses several statistical tests which can be used to test differences in the distributions of the length of birth order specific intervals when these distributions are obtained using life table techniques. Since such estimates involve incomplete or arbitrarily censored intervals, conventional statistical tests are not appropriate.
The tests are illustrated through application to data from the 1965 National Fertility Survey. Data on the occurrence and timing of third and fourth births are shown to be significantly different for black and white currently married women. Comparisons among income groups, however, fail to show significant differences.