The DHS Program User Forum - RDF feed
https://userforum.dhsprogram.com/index.php
Re: Wealth Index On SES
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28739&th=13489#msg_28739
I believe the variable you are looking for is hv270 in the HR and PR files, v190 in the IR, BR, and KR files, and mv190 in the MR file. This is a categorical variable with 5 ordered categories. The wealth quintiles are defined at the level of the household and are the same for everyone in the same household--man, woman, or child.
The all-women factors, such as awfactw, are not relevant. They are used for some analyses to adjust for the fact that the BD surveys of women are limited to ever-married women. For example, if you want to estimate the mean number of children ever born, for all women in Bangladesh, you need to compensate for the omission of never-married women, who are assumed to have had no children. The adjustment can be made specific for the category of v190 by using awfactw, just as it can be made specific for region (v024) with awfactr, education (v106) with awfacte, etc. But I don't think you have any need for awfactw (or any other aw factor) at all.
The first thing you need to decide is whether your units of analysis are households (use the HR file) or all individuals in the households (use the PR file) or ever-married women age 15-49 (use the IR file). As I said, the variable you want to use is probably hv270 or v190, depending on the units of analysis and the file.]]>Janet-DHS2024-03-01T13:54:33-00:00Re: Wealth Index (DHS)
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28715&th=13485#msg_28715
There is a lot of information about the wealth index on the DHS website: https://www.dhsprogram.com/topics/wealth-index/index.cfm. Yes, other researchers have constructed alternatives, particularly for non-DHS surveys. When using DHS surveys and data files, I don't see a reason to use an alternative.
The main distinction is between the continuous index (hv271 or v191 or mv191, depending on which file you are working with) and the quintiles (hv270, v190, or mv190), which divide the continuous index into 5 intervals with the same weighted numbers of de jure household residents in the PR file. If you are using all of India, you will use hv270 (etc.) rather than the state versions, which would only be used if you are restricting your analysis to a single state. hv270a would be used for a national analysis if you want to adjust for the fact that wealth tends to be higher in urban areas than in rural areas.
I suggest that you open the PR file and enter "tab hv270 hv025 [iweight=hv005/1000000] if hv102==1, col" and "tab hv270a hv025 [iweight=hv005/1000000] if hv102==1, col". The first command will give equal-sized quintiles at the national level but not within the urban and rural subpopulations. The second command will give equal-sized quintiles at the national level AND within both the urban and rural subpopulations. Whether you select hv270 or hv270a depends on whether you want a measure of wealth that does not differentiate between urban and rural status. I personally prefer to use hv270, rather than hv270a. The tables in the final report use hv270.]]>Janet-DHS2024-02-28T19:02:31-00:00Re: How to arrive at final wealth index score using PCA
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28714&th=13482#msg_28714
Yes, you can match the SPSS procedure with Stata. Have you looked at the DHS website (https://www.dhsprogram.com/topics/wealth-index/index.cfm)? MANY variables go into that PCA. Various people have tried to simplify the construction, and I support that, but if you simplify it you will not get a match.
The conversion to quintiles is a little more complex. The following lines, applied the HR file, show how to do it ("mem" is the number of household members). You have to do it with the HR file because everyone in the same household has the same value of the continuous index and is in the same quintile.
tab hv270 hv270_test1 [iweight=wt]]]>Janet-DHS2024-02-28T19:01:53-00:00Re: Wealth indices compare to income and consumption measures
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28711&th=13479#msg_28711
There has been much discussion of the wealth index on the forum. You can also look here: https://www.dhsprogram.com/topics/wealth-index/index.cfm.]]>Janet-DHS2024-02-28T18:58:28-00:00Wealth Index On SES
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28678&th=13489#msg_28678
I am using the BDHS 2017-18 data.
I am pretty much confused what to do in this circumstance. Please, help me.]]>Jobayer H2024-02-22T01:03:29-00:00Wealth Index (DHS)
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28673&th=13485#msg_28673
blossom word game
Hello,
I working on the DHS dataset for India, and have a few doubts regarding the wealth index. I see that the DHS data already includes the wealth index, but I am confused about the following points:
1. I have come across a few papers which use the DHS dataset, and construct their own wealth index instead of using the one already given in the DHS dataset. Often, there is not much explanation on why and how they construct a new index. I can guess that this may be because the newly constructed index is more suited to their specific analysis. I was wondering if there are any papers that guide on when and when not to construct a new wealth index or use the one given in the DHS dataset.
2. The DHS dataset consists of many types of wealth indices. If at all one plans on using the wealth index given in the DHS index- how should one decide on going about which wealth index among the plethora given in the dataset to use? (wealth index combined, wealth index factor score combined (5 decimals), wealth index for rural/urban, wealth index factor score for urban/rural combined (5 decimals), wealth index within state, wealth index factor score within state (5 decimals), wealth-index urban within state, wealth index factors score (5 decimals)-urban, wealth-index rural within state, wealth index factors score (5 decimals)-rural)
Is there a guiding paper on the same?
I would be grateful for any guidance on the above.
Thank you.]]>emmaduke2024-02-20T12:40:37-00:00How to arrive at final wealth index score using PCA
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28666&th=13482#msg_28666
smash karts
I need to arrive at a final wealth index score for my analysis using PCA in stata. I am working with the Demographic and Health Survey dataset which provides only an asset-based wealth index for measuring socio-economic welfare. I am unsure about the variable normalisation process to arrive at the final wealth index score for each household. I have followed the variable preparation process using the Filmer and Pritchett /Demographic and Health Survey and coded each variable as a dummy. I have always assumed that if I predict the scores using the command predict score1 for example and apply the survey weights to generate break the scores up into quintiles, I arrive at quintiles representing different categories of wealth from the poorest to wealthiest. I, however, have been unable to duplicate the survey provided wealth index and quintile distribution using this approach. They provide the code in SPSS, and I am working in stata. I need to be sure what I am doing is right and not necessarily arrive at the very same index.
I did some more reading used the following document;http://www.psi.org/equity-wealth-quintileguide from the Population Service International (PSI). They show how to construct a wealth index based on a survey. In the document which is a guide they use the PCA command and then generate a factor score using the predict command, it appears they do not use this as the final wealth index score. They, however, go on to standardise each variable using the mean and standard deviation. They then proceed to create a final wealth index score for each household by multiplying the standardised variable by the factor scores obtained from the PCA and summing these up for each household. I guess my questions are;
1. Is predicting the relevant factor score using the predict command in stata alone adequate to arrive at a wealth Index score based on which to rank households.
2. Is ranking the households using quintiles and survey weight command sufficient to reflect the distribution of wealth based on this index.
3. is the process of generating the index using the standardised variable as described above same or similar to the stata predict command, ie. another way toobtain the score?
In the interest of brevity my commands are as follows;
pca n1, n2, n3 ... ni [aweight = weight], means
predict wealthscore1 wealthscore2
sort wealthscore1
xtile quintile= wealthscore1 [pweight=weight], nq(5)
Some clarity as soon as possible will be greatly appreciated. regards,
]]>annabellebronson2024-02-19T10:10:10-00:00Wealth indices compare to income and consumption measures
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28662&th=13479#msg_28662
]]>aprilroland2024-02-16T04:43:15-00:00Re: understanding hv270 & hv270a
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28635&th=13459#msg_28635
Rae2024-02-10T19:53:23-00:00Re: understanding hv270 & hv270a
https://userforum.dhsprogram.com/index.phpindex.php?t=rview&goto=28620&th=13459#msg_28620
To understand hv270a you have to go back to the continuous versions of the wealth index, hv271 and hv271a. hv271a is constructed as two linear functions of hv271, one for urban households and one for rural households. You will see this if you regress hv271a on hv270, separately for hv025=1 and hv025=2. This is done such that the mean of hv271a is approximately the same for urban and rural households.
Then, just as hv270 is constructed from hv271 by converting to quintiles, hv270a is constructed from hv271a by converting to quintiles, but separately for urban households and rural households. You can find more in the Guide to DHS Statistics and here: https://dhsprogram.com/topics/wealth-index/index.cfm.
We are just finishing up some of the DHS documentation for DHS8. The Standard Recode Manual should be out soon, but I can't say when. Sorry for the delay, which goes back to the delay in DHS8 data collection because of covid.]]>Janet-DHS2024-02-07T17:51:25-00:00