Does Social Parameters and Landholding Size Affect Household Food Security in Rural India?

Past studies have proved that various factors affect household food security. This paper presents whether social groups, family size and landholding sizes are related to food security of the households. Consumption of food groups as per household recommended requirement is considered as food security in this paper. Subsample of 315 households from 39 villages of Mathapada Gram Panchayat and 8 villages from Doraguda Gram Panchayat in Boipariguda block in Koraput district of Odisha state, India were taken for the study. Results of the study show that social group and family size have significant effect on the gap between consumption and household food groups and nutrients requirement, while landholding did not show any effect. Intervention studies and knowledge on consumption of own production and marketing of excess produce should be emphasized. Government entitlements should focus on decreasing the gap between the consumption and household requirement by increasing the accessibility to foods and making the rural households food secure.


Introduction
Food security is defined as "a situation that exists when all people, at all times, have physical, social and economic access to sufficient, safe and nutritious food that meets their dietary needs and food preferences for an active and healthy life" (FAO, 2003). This definition consists of four central parts: availability, stability, accessibility and utilization. A food system is said to be vulnerable when one or more of the four components of food security are uncertain and insecure (FAO, 2008). Household food security is the application of this concept to the 2. Methodology

Study Location
Thirty nine villages of Mathapada Gram Panchayat and eight villages from Doraguda Gram Panchayat in Boipariguda block were taken for the study. A house listing survey was undertaken to understand the socio-demographic profile of the study villages. Totally, there are 1575 households in the selected 47 villages with a population of 6795 (3360 males and 3435 females). Majority are scheduled tribes (74%) and lived in semi pucca house (54%). About 53% of the head of the households are farmers; 22% are agriculture labours and 20% are engaged in non-agriculture wage labour. The total operational land is 1528 hectares and since Koraput is a hilly region, different land pattern is observed: 54% of the total lands are under upland, 30% under lowland and 16% under medium land. About 42% of the households have marginal land (<1 ha) followed by small land (1 to 2 ha) (24%); semi medium land (2 to 4 ha) (11%); medium land (4 to 10 ha) (3%); 21% of the households are landless.

Sampling
Subsample of 315 households (20% of total households) was selected by random sampling method using SPSS statistical software (version 20). A survey was conducted using structured questionnaires for the subsample households during March-April 2019, in order to understand the agriculture pattern, home garden, fishery, livestock and household food consumption pattern and nutrition knowledge. It was ensured that the distribution of socio-economic variables and land class remained similar between the subsample households and total population.

Agriculture, Home Garden, Fishery and Poultry
Details on crops cultivated in upland, medium land and lowland in Kharif season for the period of June to November 2018 and rabi season for the period of December 2017 to May 2018, total operational land, crops grown, home garden details on vegetables and fruits grown last year were collected. Apart from these existing status of pisiciculture and backyard poultry were also collected.

Food Consumption
Quantity of food items consumed by household during the month of February  Tables published by the Indian Council for Medical Research (ICMR, 2017). The gap in the food group intake and nutrient intake was calculated by finding the difference between the amount of foods or nutrient required for the family based on the RDI and RDA (recommended dietary intake and recommended dietary allowance) and actual consumption.

Statistical Analysis
ANOVA and linear regression was used to assess the effect of social parameters, family size and land size on the food consumption. STATA and SPSS are the statistical packages that were used for the analysis.

Results and Discussion
Majority of the population in the sample households were schedules tribes (74%) followed by schedules caste (13%), other backward classes (12%) and only 1% belonging to general category. The house types were semi pucca houses (56%) and kutcha houses (33%). Family size in 60% of the households was 4 to 6 members, 1 to 3 members in 31% of households and more than or equal 7 members in 9% of households. Drinking water was sourced from tube well or bore well by 85% of the households; only 10% and 4% of the household sourced tap water and open well or dug well, respectively and 1% household sourced drinking water from surface water. Toilet facility was available in 70% of the household but not used due to unavailability of water. Eighty percent of the households had home garden in their backyard area; 81% of the households had poultry and a few households (8%) had pond. Most of the households (48%) had less than 1 hectare of land, 31% of households had 1 to 2 hectares of land and 11% of the households were landless.  Table 2 shows the consumption of foods according to food groups (g/CU/day).
The household requirement of cereals and millets, pulses, other vegetables, roots and tubers and fruits was met as per the recommended allowances in the general category which is considered as the most advantaged group in a community. Food requirement of OBC category households, which was second in order of social groups met the household requirement of cereals and millets and roots and tubers while all other food groups did not meet the household requirement. Figure 1  Family size played a significant role in the household requirement and consumption of all food groups, except in the requirement of milk and milk products  which was consumed less irrespective of the family size. In a small family size of 1 to 3 member household, the consumption of cereals and millets, pulses, vegetables and sugars met RDI while in large family sizes except cereals all the other food groups were below the required quantity. Pulses and legumes were consumed on an average of 174 g/day against the household requirement of 345 g/day in 4 to 6 members and 198 g/day against the household requirement of 518 g/day in 7 members and above. Other vegetables were consumed on an average of 775 g/day against the household requirement of 914 g/day in 4 to 6 members and 879 g/day against the household requirement of 1382 g/day in 7 members and above. Similar results were reported by Sekhampu (2013) and Olayemi (2012) that household size was negatively associated with household food security. Shone et al. (2017) also reported that odds of households with larger family size to be food insecure was higher than households with smaller family sizes. Except for milk and milk products, family size has significant effect on household food requirement. Figure 2 shows the gap in average consumption of different food groups/ household/day compared with the household requirement based on recommended dietary intake based on family size.
Landholding size had significant effect on the consumption of cereals and millets as per requirement. It was observed that all households consumed roots and tubers as per requirement irrespective of the land holding size. The consumption of other vegetables by landless household were found to be as per requirement. All other food groups were consumed less than the requirement in all different land holding sizes. Nkomoki et al. (2019) reported that the size of land was found to have a positive relationship with food security. Owning smaller farm land increases the risk of being food insecure nearly by 2 times compared to larger land size (Shone et al., 2017). Another study by Agidew & Singh (2018) reported that the majority of the food insecure households own less than 1 ha of farmlands. Figure 3 shows  the gap in average consumption of different food groups/household/day compared with the household requirement based on recommended dietary intake based on landholding size. Rammohan & Pritchard (2014) used ordered probit models and indicated that in Myanmar, an increase in land size enhanced household food security status. Muraoka et al. (2018) demonstrated that an increase in land size resulted in a rise in household food security.
Overall, gap between the nutrient requirement and consumption was observed in all nutrients except, vitamin C and folates. However, the gap between the nutrient requirement and consumption was negative, particularly more in the ST groups, significantly in protein, energy, fats and zinc and also in calcium, vitamin A and iron (Table 3). In contrast to the present result, Behrman & Deolalikar (1990) reported that household social groups affiliation does not affect the nutrient intake of household members.
The nutrient gap also increases as the family size increases, significantly. Interesting, the gap in the nutrients like protein, energy, calcium, iron and zinc  was more in large land holding households and lesser in landless households. Sani & Kemaw (2019) reported that the mean kilocalorie intake of food insecure households was 1440.37 kcal/day, with the minimum and maximum being 597.65 kcal and 2048.13 kcal, respectively. The regression (Table 4) shows that as the family size increases, the gap between the consumption of food groups increases except cereals and millets showing that larger family size are more food insecure than lesser family size. But the land size does not have any significant effect on the consumption and requirement of foods except for pulses, however the negative association shows that as the land size increases the gap between consumption and requirement decreases. The significant positive association between general social group and SC, ST and OBC in the gap between consumption and requirement of food groups showing that general group is in better side.
Rural India remains a caste-based society (Anderson, 2012)   food secure than the Schedule Caste household (Sarkar & Shekhar, 2017). The results of the present study also showed that the SC and ST households consumed lesser foods than that is expected to be consumed when compared to the general caste. The gap between the requirement and consumed nutrients like protein, calorie, iron, vitamin A and zinc was also significantly high in ST group.
In general, food insecure households were characterized by smaller household size. Large family size has significant relationship with much greater risk of poverty. This was similar to the results of the present study and showed that the consumption was more than the requirement of cereals and vegetables in households having lesser members.
Average size of land was much higher in the food secure household (2.4 acres) compared to food insecure (0.9 acre), and food insecure with hunger households (0.1 acre) (Sarkar & Shekhar, 2017). Frelat et al. (2016) showed that farm size is a determinant of food security in sub-Saharan Africa and as farm size increases, the probability of a household being food-secure also increases. The results of the present study were in contrast to the above studies. The gap between consumption of foods irrespective of land holding size was more except cereals and millets and the gap was more in the large landholding households. The linear regression also confirmed the results.
The limitation of this study was that the food and nutrient consumption was collected at one time of a year and as the food consumption may vary based on seasons, further study can be done with the data collected at different time of a year to understand the food and nutrient gap in different seasons. Further research can be also done using data collected using 24 hour diet recall data to access the nutrient gap among different age groups and gender.

Conclusion
The study demonstrates social groups, family size and landholding size are important factors for rural household food security. Improving the knowledge on