Abstract

Fertility research and the analysis of fertility trends are central to population studies. However, in countries such as Iran, the absence of detailed birth history data restricts the application of many advanced fertility analysis methods. This study addresses this limitation by developing and evaluating a birth history reconstruction approach using the two-percent sample of the 2016 Iranian Population and Housing Census. The method first links children living in households to their mothers. It then estimates the number and probable ages of deceased and absent children using conditional fertility probabilities derived from cohort fertility patterns, survival probabilities, mortality patterns, and the non-residence adjustment factor (K). Based on these estimates, complete birth histories are reconstructed for individual women.

The results demonstrate a high level of consistency between the reconstructed birth histories and estimates obtained from the own-children method. The reconstructed total fertility rate (TFR) for 2016 was 1.95 children per woman, identical to the estimate derived from the own-children method. The RMSE, MAE, and MAPE values for the national level were 0.034, 0.025, and 1.33%, respectively, while the Pearson correlation coefficient reached 0.95. The method also showed strong performance across urban and rural areas, with correlation coefficients of 0.98 and 0.97.