Determinants of Food Insecurity in the Sundarbans: Multinomial Logit Approach Apropos the Food Insecurity Experience Scale

Abstract

Aims: This study examines socio-economic and demographic determinants of food insecurity in the Sundarbans region using a multinomial logistic regression model, aiming to identify key determinants of food insecurity levels. Study Design: A cross-sectional analytical study employing multinomial logistic regression to assess categorical predictors of food insecurity. Place and Duration of Study: Data were collected over six months (January 2024–December 2024 at regular intervals) across 11 blocks in South and North 24 Parganas, such as Gosaba, Basanti, Kakdwip, and Sagar. Methodology: Structured interviews with 300 randomly selected respondents representing diverse livelihood groups provided socio-economic and environmental data, including household income, cultivated land, indebtedness, and extension services participation. Food insecurity was categorized into food secure, moderately food insecure, and food insecure, and analyzed via multinomial logistic regression. Results: The model demonstrated 80.9% overall predictive accuracy, correctly classifying 78.9% of food-secure households, 91.7% of moderately food-insecure households, and 39.5% of severely food-insecure households. Protective factors like cultivated land access, household education, and social capital mitigated food insecurity, while indebtedness and skipped meals increased food vulnerability. Agricultural extension services and in-kind transfers were effective interventions for food insecurity. Pseudo R-squared values (Nagelkerke: 0.663) confirmed explanatory strength. A one-unit increase in cultivated land reduced food insecurity odds by 91.3% (OR, 0.087; 95% CI 0.020–0.384, p<0.001). Policy Implications: The findings offer actionable policy insights to enhance food security in vulnerable regions. Conclusion: Combating food insecurity in the Sundarbans requires integrated strategies improving land access, debt alleviation, education, and social networks, alongside sustainable practices, ensuring equitable food systems.