Estimation and Projection of the Fertility: National, Provincial and Local Level in Nepal

Abstract

Fertility levels and patterns provide an important demographic information regarding the population change, as well as socio-economic development and human well-being. There are very few specific studies in Nepal that estimate and project fertility among different caste/ethnic groups at the national, provincial and local levels. This study compares the fertility estimation and projection at national and its sub-domains, and verifies and validates in Nepal. National household censuses (2001 and 2011) were carried in 12.5 percent of the total households and (649,476 and 1,091,337) reproductive age group of sample women were identified through analysis respectively. Age sex pyramids and frequency table represent demographic scenario of national and provincial levels. The study was carried out adhering to the Arriaga method and changing P/F ratio method. Algorithm first smoothed local age specific rates (ASFR) using Empirical Bayes method and then applied a new variant of Brass’s P/F parity that is robust under conditions of rapid fertility decline at local level. The small area estimation (SAE) was applied at local level and different caste/ethnicity were selected to estimate the fertility which is the contribution of the study. Total fertility rate (TFR) values will reach at national level using linear interpolation, and extrapolation by 2031, it reaches replacement level. The study showed that the Muslim, Hill Janajati, Madhesi Dalit, Madheshi Other Caste, Hill Dalit and Others Minor Caste will have (2.37, 2.31, 2.32, 2.20, 2.37, 2.51) high fertility rate which is above the replacement level of fertility at the end of 2031. Similarly, the fertility rate of Newar, Tarai Janajiti, Brahman/Chhetri and Madheshi Brahman (1.58, 2.03, 2.09, 1.8) will have below the replacement level of fertility. At the province level, Karnali (3.42), Sudurpashchim (2.59) and Lumbini (2.14) will have high fertility rates; Madhesh Province and Gandaki will reach 2.1; Province 1 (2.05), Bagmati (1.9) will be below the replacement level in the same period. SAE is most useful when the vital registration system is incomplete and small local fertility samples made it difficult to estimate rates reliably; applying 742 (2001) and 753(2011) local levels in household census; mainly standardising the empirical Bayes Brass (EBB) method in Kanda (Smallest), Dhanushadam (middle) and Kathmandu (largest) at rural and urban municipal levels were selected respectively. The fertility of SAE is valuable for analysing demographic change and is important for local planning and programme. Future researchers can study to ward levels for more effective results.