ANNEX I: FOOD CONSUMPTION DATA: THE NATIONAL NUTRITION MONITORING BUREAU AND NATIONAL SAMPLE SURVEY ORGANISATION

The National Nutrition Monitoring Bureau (NNMB) since its inception in 1972 has been collecting data on diet and the nutritional status of representative population groups in different states of the country. These include the states of Andhra Pradesh, Gujarat, Karnataka, Kerala, Madhya Pradesh, Maharashtra, Orissa, Tamil Nadu, Uttar Pradesh and West Bengal. In each state, every year, 500 rural households are covered by diet and nutrition surveys. In 80% of these households food consumption levels are assessed by the one-day weighment method, wherein the investigator weighs all the raw foods to be cooked by the family for the day. The total amounts of nutrients such as energy, protein, vitamins and minerals are derived by referring to the Food Composition Tables of Indian Foodstuffs. Data on nutrient intakes are expressed on a per consumption unit (CU) basis by converting the number of individuals partaking meals into consumption units using the calorie coefficients recommended by the Indian Council of Medical Research (ICMR) for the Indian population. In the remaining 20% of the households, the 24 hour recall method of diet survey (oral questionnaire method) is carried out. The volume of cooked food consumed by each member in the household is assessed using standardized cups and then converted into raw amounts by using appropriate conversion factors derived for each food item. The nutrients derived from these foods are calculated using the food composition tables.

Apart from NNMB, the National Sample Survey Organisation (NSSO) is the only other agency collecting large-scale data on food consumption, based on consumer expenditure. They have been conducting consumer expenditure surveys annually up to 1973-74 and thereafter quinquennially, dividing expenditure into food and non-food. Calorie and protein intakes are expressed per consumption unit per day, based on the consumption of all food items collected during the reference period. The reference period for data collection on all items of consumer expenditure is the previous 30 days. A stratified two-stage sampling design is employed with census villages in rural and blocks in urban areas as the first stage units. Households are the second stage units. Sample villages are selected with probability proportionate to population, with households being selected by a circular systematic sampling method. The entire one year survey period is divided into four sub-rounds each of three months duration. The estimates based on pooled data taking by sub-rounds are free from the effect of seasonality.

The intake level of the rural population, as assessed by NSSO is quite high when compared to NNMB data, indicating differences in the main objective, sampling design and method of data collection between NNMB and NSSO. These are shown in the table below.

Table A1: Comparability of NSSO and NNMB Survey Designs

Criterion	NSSO	NNMB
Objective	Collection of information of household consumer expenditure	To assess nutritional status of individuals	To assess nutritional status of households
Method of collection	Recall	Recall	Weighment
Sample design	Stratified two-stage	Stratified three-stage	Stratified three-stage
Sample size	Large, varying size	Very small (one fourth of weighment) fixed size for each state	Small (one thirty third of NSSO) fixed size for each state
Reference period	One month preceding the day visit	One day preceding the day of visit	Day of visit
Unit of data collection	Household	Individual	Household

Source: Kulkarni and Kumar (1984)

ABBREVIATIONS OF STATES

Rajasthan (Ra)
Jammu and Kashmir (JK)
Himachal Pradesh (HP)
Punjab (Pu)
Haryana (Ha)
Delhi (De)
Uttar Pradesh (UP)
Bihar (Bi)
West Bengal (WB)
Orissa (Or)
Andhra Pradesh (AP)
Tamil Nadu (TN)
Karnataka (Ka)
Kerala (Ke)
Sikkim (Si)
Meghalaya (Me)
Assam (As)
Nagaland (Na)
Manipur (Man)
Mizoram (Mi)
Tripura (Tr)
Goa (Go)
Gujarat (Gu)
Madhya Pradesh (MP)