On the Care and Handling of Regression Specifications in Fertility Research

Dennis De Tray, Zubeda Khan


Economists, sociologists and demographers must often attempt
to answer important questions with data not well suited to the problem
at hand. One example that crops up frequently in
socio-economic-demographic literature is the use of samples of women,
whose ages span the entire fecund period, to study the effects of
couples' characteristics on "completed" fertility, or on the demand for
children. In this case, the usual procedure is to control either for age
or for duration of marriage, and to assume that issues concerned with
timing and spacing of children can be ignored. Under this assumption,
differences in the level of the stock of children among families at any
point in time (any age) bear a one-to-one correspondence to differences
in completed fertility observed at the end of the fertile period. In
this paper, we explore some of the pitfalls researchers may encounter
when using data with this characteristic, especially in the absence of
exact knowledge of the functional form of the relationship between age
or duration of marriage and other variables thought to affect actual

Full Text:


DOI: https://doi.org/10.30541/v16i3pp.309-324


  • There are currently no refbacks.