G. Angeles et al., PURPOSIVE PROGRAM PLACEMENT AND THE ESTIMATION OF FAMILY-PLANNING PROGRAM EFFECTS IN TANZANIA, Journal of the American Statistical Association, 93(443), 1998, pp. 884-899
Most studies evaluating the impact of family planning on fertility tre
at the presence of family planning clinics as being ''randomly'' assig
ned among the areas included in the study. They tend to ignore the pos
sibility that the distribution of services may be related to the ferti
lity level observed ina particular area. In some cases the distributio
n of services may respond to a conscious effort by public authorities
or funding agencies to target areas with observed higher fertility. Ev
en in absence of program planning, the factors determining service pla
cement might be related to the determinants of high, or low,fertility
in a particular area. If that is the case, and one fails to account fo
r the endogeneity of family planning services, then the estimated impa
ct of family planning programs will be biased. This article presents a
modeling approach to address this issue. The model extends the simult
aneous equation framework by integrating an individual-level model of
timing and spacing of children with the dynamic process of program pla
cement. Individual-level data from the 1991/1992 Tanzania Demographic
and Health Survey are augmented with data on the timing or and factors
influencing, family planning service placement to demonstrate the app
roach. The empirical results show that standard methods yield misleadi
ng results on the impact of different components of the family plannin
g program on fertility. In particular, the effect of access to family
planning hospitals on births is overstated, and the impact of access t
o health centers that offer family planning is understated significant
ly. We quantify the size of these effects through simulations.