The DHS Program User Forum
Discussions regarding The DHS Program data and results
Today's Messages (on)  | Unanswered Messages (off)

Forum: Nutrition and Anthropometry
 Topic: Maternal Education and Stunting
Re: Maternal Education and Stunting [message #28928 is a reply to message #28900] Thu, 28 March 2024 15:25
Janet-DHS is currently offline  Janet-DHS
Messages: 666
Registered: April 2022
Senior Member
Following is a response from DHS staff member, Tom Pullum:

To proceed with your analysis in Stata, you will need to use regression commands. For your first question, you can use a logistic regression model where the dependent variable is stunting (coded as 1 for stunted, 0 for not stunted) and the independent variable is mother's education (measured in years). You can also include an interaction term between mother's education and a binary variable for urban/rural location to see how the impact of mother's education on stunting varies across urban and rural areas. The Stata command for this would be:
logit stunting c.mother_education##c.urban_rural

For your second question, you will need to recode mother's education into categories (e.g., no education, incomplete primary, complete primary, etc.) and then run a logistic regression model with these categories as independent variables. You can then compare the coefficients to determine which category yields the greatest relative return. The Stata command for this would be:
xi: logit stunting i.education_category

Please note that these are basic commands and you may need to adjust them based on the specifics of your dataset and research question.



Current Time: Thu Mar 28 22:00:32 Coordinated Universal Time 2024