District-Level Household Survey

District-Level Household Surveys (DLHSs) were representative surveys conducted at the district level and measured demographics, health, and nutrition status. Four rounds of the survey have been carried out: 1998-99

(DLHS-1), 2002-04 (DLHS-2), 2007-08 (DLHS-3) and 2011-12 (DLHS-4). The DLHS has many similarities with the NFHS, although the detailed survey covers only women. Unlike the initial rounds of the NFHS, it was devised precisely for district-level analysis. The number of households and women covered in successive rounds of the survey has varied. Data are published at the state level (with rural/urban differentials), and raw data are available for further analysis and research. Although it has collected its data during a reference period which would allow for the calculation of incidence rates from one period to the next, the DLHS has been stopped as of its last round in 2011-12. A few studies on child marriage have drawn from these data (e.g., Srinivasan and others 2015). It is now the NFHS that will be the main source of analysis for health-related data, including at the district level.

The Indian Human Development Survey

The Indian Human Development Survey (IHDS) is a multi-topic survey that collects statistics on current age, marriage status, and age at marriage of all the household members and is conducted by the National Centre of Applied Economic Research. The first round of the survey in 2004-05 covered 41,554 households in 1,503 villages and 971 urban neighbourhoods across India and has collected data on a wide range of themes. The second round of the IHDS re-interviewed most of these households in 2011-12. The large coverage of issues allows analysis of associations across a range of social and economic conditions, more so than with the NFHS, but not below the state level. Since most households were re-interviewed after a period, statistics on incidence of child marriage could be calculated after required standardization. The third round of the IHDS was initiated in 2019. State-level analyses such as those by Desai and Andrist (2010), Desai and Vanneman (2016) (discussed in Chapter 4) and Allendorf and Pandian (2016) have been based on this data set.