z-logo
open-access-imgOpen Access
Multinomial Logistic Regression Model to Identify Factors Associated with Food Insecurity in Rural Households in Nepal
Author(s) -
Sameena Shah
Publication year - 2020
Publication title -
nepalese journal of statistics
Language(s) - English
Resource type - Journals
eISSN - 2645-839X
pISSN - 2565-5213
DOI - 10.3126/njs.v4i0.33448
Subject(s) - multinomial logistic regression , food security , logistic regression , ordered logit , socioeconomic status , poverty , variables , categorical variable , socioeconomics , geography , agriculture , environmental health , economics , statistics , economic growth , mathematics , medicine , population , archaeology
Background: Food is basics of our lives and many people experiences food insecurity at some time because of food deprivation and lack of access to food due to different resource constraints. It is a global challenge and threatens the rural people in developing countries like Nepal. Objective: The objective of the study is to identify the factors associated with food insecurity in rural area of Nepal. Materials and Methods: The analysis is based on rural household data extracted from the data of Nepal Demographic and Health Survey 2016. The dependent variable food insecurity status was measured in four levels namely food secure, mildly food insecure, moderately food insecure and severely food insecure household using Household Food Insecurity Access Scale. Independent variables were categorical and quantitative variables. In order to identify the factors associated with food insecurity, ordinal logit model was fitted initially. Due to violation of test of parallel lines by overall as well as some of the independent variables, multinomial logistic regression model was finally adopted by examining the model adequacy test. Results: The fitted multinomial logistic regression satisfied the diagnostic test including tests of goodness of fit, multicolinearity diagnostic criteria and minimum criteria of utilization of the model with about 29% predictive power. The variables ecological region, wealth index, size of agriculture land, any member(s) having saving account in any financial institution, any member(s) had gone to foreign employment in last 5 years other than India, family size, number of members completed secondary education and household member rearing cattle(s) were found to be significant. The poorest households (HHs) had 3.14 (CI: 1.88-5.26) times, poorer HHs 2.51 (CI: 1.55-4.07) times and moderate HHs 1.42 times higher chances of being severely food insecure relative to rich HHs. Conclusion: The study revealed that food insecurity of the rural HHs increases with decrease in the wealth index, size of land and number of members of the HHs with completed secondary education. The food insecurity of the households decreases with increase in the access to bank service.

The content you want is available to Zendy users.

Already have an account? Click here to sign in.
Having issues? You can contact us here