The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Domestic Violence » Calculating the estimated marginal effect of spousal violence on women's employment using NFHS 5 dta
Calculating the estimated marginal effect of spousal violence on women's employment using NFHS 5 dta Mon, 08 May 2023 12:14
 567241vp Messages: 3Registered: April 2023 Member
Hi,

I'm currently working on a paper that seeks to understand the relationship between spousal violence and women's employment in India. I'm trying to run a logit regression with the following equation: E = α +Vδ1 + Wδ2 + Mδ3 + Hδ4 + e in Stata
, where E is the dependent variable representing employment for married women, V is the spousal violence experience indicator for the woman, W represents the set of attributes of the woman including her education level and age group; M represents the set of attributes of her husband, such as his propensity to consume alcohol and get drunk, education, and employment status (employed or unemployed); and H represents the set of attributes of the household, including its caste category, religion, and location.

I'm having trouble running the logit regression as I'm unsure which variables to use. Is there any advice someone could give me on this? I can also provide further details if necessary.

Thank you.

[Updated on: Mon, 08 May 2023 12:15]

Report message to a moderator

Re: Calculating the estimated marginal effect of spousal violence on women's employment using NFHS 5 dta [message #26878 is a reply to message #26839] Mon, 15 May 2023 12:54
 Janet-DHS Messages: 816Registered: April 2022 Senior Member
Following is a response from DHS staff member, Tom Pullum:

You are asking for more assistance than DHS staff can provide. All of your groups of variables are very vague. Perhaps other users can help.
Re: Calculating the estimated marginal effect of spousal violence on women's employment using NFHS 5 dta [message #27552 is a reply to message #26839] Fri, 01 September 2023 11:17
 dcpathak Messages: 6Registered: September 2022 Member
Hi,
I came across this query today. I know it was asked way back, so apologies in advance if you do not need this response anymore.
I think you are trying to see how a woman's employment status affects her odds of spousal violence. In this case, the dependent variable should be spousal violence (a binary variable) and not employment status. You may have a look at my related paper https://doi.org/10.1177/09722661221132968
I hope this helps.
Durgesh
 Previous Topic: Data on Domestic Violence in 2017 and 2022 Philippines DHS Next Topic: Merging data files
Goto Forum:

Current Time: Sun Sep 8 02:45:03 Coordinated Universal Time 2024