We use two data sources for the estimation of the India State Hunger Index. These are the third round of the National Family Health Survey (2005‐2006) for India (referred to as the NFHS‐III data)2 and the 61st round of the National Sample Survey (NSS) data from 2004‐20053.
The India State Hunger Index is calculated and presented for 17 major states in the country. These states cover 95 percent of the country’s population, according to the 2001 census in India. While the NFHS‐III has a large enough sample size to yield representative estimates of the child underweight and mortality rates even for the smaller states and union territories in the country, the sample size of the NSS 61st round is insufficient for estimating undernourishment rates in these places. Therefore, we restricted our sample to those states for which the NSS yields precise state‐level estimates.
Population undernourished based on calorie consumption: The proportion of the population undernourished based on calorie consumption was estimated using unit‐level food consumption data from the 61st round of the National Sample Survey, conducted in 2004‐05. The NSS obtained household consumption data on more than 225 individual foods; these were converted to calories using food‐to‐calorie conversion factors reported by NSSO (2007). Allowances for calories from meals eaten outside the home were made using the procedures suggested and followed by NSSO (2007).
The GHI 2008 (von Grebmer et. al, 2008) reports the calorie‐based undernutrition for India to be 20%, based on the FAO undernourishment cut‐off of 1,820 kcals per person.4 The FAO estimate of calorie undernutrition is based on data on food availability, collected from national food balance sheets. As such, it is likely to differ from calorie undernutrition estimates that are based
2 The National Family Health Survey III is the third one in a series of such surveys. It was conducted by the International Institute for Population Sciences (IIPS, Mumbai which acted as the nodal agency. The survey collected data on 51,555 children under 5 years of age. Technical support was provided by Macro International while USAID provided financial assistance for the study.
3 The National Sample Survey Organisaton conducts surveys on various socio‐economic issues annually. The 61st round of the National Sample Survey (NSS) was conducted between July 2004 to June 2005. The Household Consumer Expenditure Survey was conducted on a large sample basis and was the seventh quinquennial survey on the subject. It covered a sample of 79,298 rural and 45,346 urban households in all the states and union territories of India.
4 It should be noted that the calorie norms for poverty used within India differ significantly from the FAO‐ recommended norms for undernourishment. Dandekar and Rath (1971) used a norm of 2,250 calories per day per person to set a poverty line for India. A task force of the Indian Planning Commission subsequently revised these calorie norms to 2,400 in rural areas and 2,100 in urban areas (the difference being attributed to the lower rates of physical activity in the urban areas (GOI 1979). These are the calorie norms that underlie the official poverty line currently in use by the Government of India. To complicate matters further, the National Sample Survey Organisation (NSSO) uses a daily calorie norm of 2,700 calories per consumer unit (not per capita) (NSSO 2007).