Ad Hoc WEAI Master Data
Description
This is cross-sectional data of 1148 individuals across 585 dual-households collected in 2019. It records the indicators of WEAI construction from selected samples of four states in India for a study on understanding Gender-Nutrition linkages in agrarian households. Objectives:- • To explore the levels of women empowerment and nutritional scenarios of the study area in terms of nutritional intake and outcome of population. • To decompose women’s access to farming decisions by its constituent factors to observe the most important contributor to empowerment. • Explore the linkages between women’s empowerment within the farming system – its contributory indicators and nutritional outcome and the gender gap in nutrition within households. The questionnaire collected 1. Indicators of Household Characteristics – socio-economic attributes, demographic composition, and geographical characteristics. 2. All Household Members- age, sex, education, height, weight, occupation, duration of stay in the village, access to social protection schemes. To capture nutritional intake, other than the height and weight of cooperating participants, a standard consumption intake module of the India Human Development Survey and socio-economic characteristics as controls for quantitative analysis were used. 3. Asked only to Women- Indicators of autonomy and empowerment in agriculture (following the indicators in Women Empowerment in Agriculture Index); item-wise consumption of food last 24 hours. To capture women’s access to productive resources, information was collected to create an Adhoc Women Empowerment in Agriculture Index (WEAI) using Alkire Method (Hazel et al. 2015). Daily dietary consumption data was collected for each individual to construct an Individual level Dietary Diversity Index. The height and weight of members were collected to identify persons with malnutrition in the households. The index of women empowerment was a key predictor along with the geography or agroecological location of the household.
Files
Steps to reproduce
A multi-stage, systematic random sampling procedure was adopted. Sample calculations based on NSSO Employment Unemployment Survey 2011-12 (68h Rounds) and NFHS IV (2015-16) on women’s autonomy, participation in agriculture, and nutritional outcome at a district level. Since we wanted to generalize our results for a set of agricultural regions these were our first stage units of the sample. Two major agrarian regions “Sub-humid” regions (coastal excluded) and “Semi-arid” Regions (arid excluded) were identified using the Meso-data base of ICRISAT. After identifying the first stage units, we moved on to identify the second stage units i.e. states to draw the samples for this study. Although political state boundaries do not conform to ecological boundaries, we selected states as they have homogeneity regarding language, food, and cultural practices. States were selected based on % of women in farming(rural) using NSS EUS (68th Round) 2011 data(NIC 2008 codes 01 to 03) and % of women who rank above 0.6 in Average Index of Autonomy (rural) from NFHS Data 2015. The index was constructed by collapsing responses of women in the 15-49 age group in the autonomy section. We ranked women in the sum of these two indicators (% in the farm sector and % with autonomy score above .6) by states within each Agroecological region. From the NFHS data, we also worked out the % of women who have Chronic Energy Deficiency (BMI<18.5) as a proxy for our outcome indicator. Within the 2 meso-regions, 2 states each were selected based on the following criteria: State 1: Min ∑ (% of Women in farming sector + % women with Autonomy Score >0.6) State 2: Max∑ (% of Women in farming sector + % women with Autonomy Score >0.6) Third stage units i.e. districts were identified using the same set of criteria. To capture the micro-regions, while ranking districts, a combination of districts in these two criteria was followed: a) Districts that fell either on Sub-humid or semi-arid ecological zones. b) Districts ranking high and low on the following indicators: District 1: Min ∑ (% of Women in farming sector + % women with Autonomy Score >0.6) District 2: Max∑ (% of Women in the farming sector + % women with Autonomy Score >0.6) Within the districts, we randomly selected Gram Panchayats. Local administration provided contacts for agricultural dual households. A random sampling of households was done out of the list of households. The probability proportionate sample size was calculated using the % gap in malnourishment between the states with maximum score (p2) and minimum score (p1) in the index of autonomy computed from NFHS IV. Adjusting for errors, 574 households were kept in the sample.