Retrospectively recorded breastfeeding durations show pronounced heaping at 12, 24, and 36 months. Using the interview-reinterview component of the Malaysian Family Life Survey (MFLS), this paper shows that the heaped responses are not drawn from adjacent durations, as is often assumed in the literature. The paper then proposes and estimates a joint model of breastfeeding duration and the recorded duration of breastfeeding conditional on the true duration. The estimated regression coefficients determining the true duration are sensitive in magnitude--and in one case sign--to the correction for imprecise response.
This report is part of the RAND Corporation draft series. The unrestricted draft was a product of the RAND Corporation from 1993 to 2003 that represented preliminary or prepublication versions of other more formal RAND products for distribution to appropriate external audiences. The draft could be considered similar to an academic discussion paper. Although unrestricted drafts had been approved for circulation, they were not usually formally edited or peer reviewed.
Permission is given to duplicate this electronic document for personal use only, as long as it is unaltered and complete. Copies may not be duplicated for commercial purposes. Unauthorized posting of RAND PDFs to a non-RAND Web site is prohibited. RAND PDFs are protected under copyright law. For information on reprint and linking permissions, please visit the RAND Permissions page.
The RAND Corporation is a nonprofit institution that helps improve policy and decisionmaking through research and analysis. RAND's publications do not necessarily reflect the opinions of its research clients and sponsors.