Home » Data » Merging data files » Disability in HR with characteristics from IR and MR (Get the disability details from HR file into IR and MR files)
| Disability in HR with characteristics from IR and MR [message #26660] |
Thu, 13 April 2023 23:51  |
kmdshoyaib
Messages: 13
Registered: May 2022
|
Member |
|
|
Hi all.
I am working with Dataset of India for DHS survey of 2019
I am looking to get the disability related variable in HR file to be merged with IR file for women and MR file for men. As I can get more details about the women from IR file and for men from MR file respectively.
I would like to know what the sequence of variables I need to manipulate for me to successfully merge HR file with IR and HR file with MR, as HR file contains disability details for about six persons in the form of person 1, person 2 etc.
Hope to hear soon.
Thanks and Regards
|
|
|
|
| Re: Disability in HR with characteristics from IR and MR [message #26670 is a reply to message #26660] |
Fri, 14 April 2023 11:34  |
Bridgette-DHS
Messages: 3189
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS staff member, Tom Pullum:
We don't get many questions about these variables. Glad you are using them. I will paste below some Stata lines that you can run on the PR file (not the HR file) to re-state the variables in terms of the lines numbers in the PR file (hvidx). Without some such reshaping it will be hard to use the disability data. The file constructed with these lines will be ready for a merge with the IR or MR files AFTER you have renamed state, cluster, hh, line (for example, v024, v001, v002, v003 in the IR file) and sorted.
The disability questions are asked for a maximum of 4 people (not 6) in a household that is reported to have one or more people with a disability. Please check the report and the questionnaire for more information about the selection and interpretation.
use "...IAPR7DFL.DTA", clear
gen dis_hearing =9
gen dis_speech =9
gen dis_visual =9
gen dis_mental =9
gen dis_locomotor=9
gen dis_other =9
local letters "a b c d"
foreach ll of local letters {
replace dis_hearing =sh33`ll'1a if sh33`ll'==hvidx
replace dis_speech =sh33`ll'1b if sh33`ll'==hvidx
replace dis_visual =sh33`ll'1c if sh33`ll'==hvidx
replace dis_mental =sh33`ll'1d if sh33`ll'==hvidx
replace dis_locomotor=sh33`ll'1e if sh33`ll'==hvidx
replace dis_other =sh33`ll'1x if sh33`ll'==hvidx
}
label define dis 0 "No" 1 "Yes" 8 "DK" 9 "No one in hh has a disability"
label values dis_hearing dis
label values dis_speech dis
label values dis_visual dis
label values dis_mental dis
label values dis_locomotor dis
label values dis_other dis
keep hv024 hv001 hv002 hvidx dis*
gen state=hv024
gen cluster=hv001
gen hh=hv002
gen line=hvidx
sort state cluster hh line
|
|
|
|
|
|
|
|
| Re: Disability in HR with characteristics from IR and MR [message #26701 is a reply to message #26672] |
Wed, 19 April 2023 14:06 |
Bridgette-DHS
Messages: 3189
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS staff member, Tom Pullum:
The DHS disability module does include a Likert scale for each type of disability, but the questions about disabilities in NFHS-5 did not come from the module. There were three questions related to disabilities in the NFHS-5 household questionnaire, which were answered by the respondent to the household questionnaire. Question 31 asks "Does any usual resident of your household including you have any disability?" If yes, Question 32 asks "Please tell me the names of those persons." The interviewer records the name and line number of each person with a disability. For each of those persons, Question 33 asks "What type of disability does (NAME) have?" For each person with a disability, the type(s) of disabilities (hearing, speech, visual, mental, locomotor, other) are recorded. The extent of the disabilities is not asked or recorded.
|
|
|
|
| Re: Disability in HR with characteristics from IR and MR [message #26888 is a reply to message #26701] |
Tue, 16 May 2023 11:53 |
kmdshoyaib
Messages: 13
Registered: May 2022
|
Member |
|
|
Thank you for the response. I have used the previous code posted and it worked. Thanks a ton for the previous STATA code. With that code, I could come to a conclusion that there was around 1 percent of women with disabilities in IR file. Here comes the next part of my study,
For each of the case in the dataset, I have a new variable which denotes if the case is having disability or not using 1 or 0.
Is there any way or STATA code using which from those who are not having disability, I will be able to pick up a sample (having equal number of cases as that of disabled) which I can match with the set of cases who are having disability?
Matching with respect to
Current age of respondent
District
Type of place of residence
Wealth Index
State
After matching I was looking for a case control type of study to test the hypothesis that,
Does utilization of healthcare services, contraception, etc, differ significantly between cases (disabled) and controls (not disabled)
Also after selection of such sample, how do I proceed with weights? I think I will have to adjust the weights for excluding such high number of cases. Please guide.
[Updated on: Tue, 16 May 2023 11:54] Report message to a moderator |
|
|
|
| Re: Disability in HR with characteristics from IR and MR [message #26890 is a reply to message #26888] |
Wed, 17 May 2023 08:00 |
Bridgette-DHS
Messages: 3189
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS staff member, Tom Pullum:
Although matching may seem like a good strategy, I would not recommend it. You would be subsampling the respondents who do not have a disability. One issue then is that there would be multiple potential matches, and someone else who tried the same thing might match different respondents and get different results. That is, there is a problem of replicability. A second issue is that by keeping respondents with a disability and an equal number of respondents without a disability, you would be throwing out approximately 98% of the sample. You would lose a lot of statistical power, more than what you would have gained from the matched pairs strategy. I recommend just using logit regression, with utilization of services as a binary outcome variable, disability as the predictor of interest, and the variables you describe as control variables. District is nested in state, and there are a huge number of districts, so you may want to have a random effect for district.
|
|
|
|
Goto Forum:
Current Time: Thu Nov 7 13:32:39 Coordinated Universal Time 2024
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
