Advances in the P/F ratio method for the analysis of birth histories

Abstract

One of the most frequently used indirect techniques for deriving estimates of recent fertility from simple questions in censuses and surveys is the ‘P/F ratio’ method. Availability of detailed birth-history data, as in the World Fertility Survey, and applications of the P/F procedure as a diagnostic tool in the evaluation of the quality of data have led to simplifications and extensions of the original method. This analysis illustrates that when complete maternity histories are available, the P/F procedure can be simplified and made more powerful by (1) calculation of P/F values from cohort-period fertility rates and (2) use of two further indexing variables, namely duration since first marriage and duration since first birth, in addition to age. More generally, the paper indicates that a set of P/F values is only one of a battery of measures which aid in the analysis of trends and errors in data from maternity histories. Illustrative examples are given from various analyses of world Fertility Survey data.

Howard Goldberg has been independently pursuing an investigation of the P/F procedure by marriage duration at the Office of Population Research (Princeton University), and we have profited from recent discussions with him. We would also like to acknowledge useful comments and criticisms on earlier drafts from James Trussell and Kenneth Hill.

We would like to thank participants in the Latin American and Asian WFS Workshops for stimulating much of this work and for thoroughly testing new ideas. The workshop staff, Shea Rutstein and Joe Potter, were particularly helpful in re-defining the P/F concept for survey data. Although demographers have often thought of using the P/F procedure by marriage duration, we are especially indebted to Ansley Coale for his suggesting the idea to us.

We would like to thank participants in the Latin American and Asian WFS Workshops for stimulating much of this work and for thoroughly testing new ideas. The workshop staff, Shea Rutstein and Joe Potter, were particularly helpful in re-defining the P/F concept for survey data. Although demographers have often thought of using the P/F procedure by marriage duration, we are especially indebted to Ansley Coale for his suggesting the idea to us.

Notes

We would like to thank participants in the Latin American and Asian WFS Workshops for stimulating much of this work and for thoroughly testing new ideas. The workshop staff, Shea Rutstein and Joe Potter, were particularly helpful in re-defining the P/F concept for survey data. Although demographers have often thought of using the P/F procedure by marriage duration, we are especially indebted to Ansley Coale for his suggesting the idea to us.