The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Other countries » Haiti water DHS 2016 - why skip sh102 except for bottled water?
Haiti water DHS 2016 - why skip sh102 except for bottled water? [message #18172] Thu, 03 October 2019 10:27 Go to next message
ianross is currently offline  ianross
Messages: 3
Registered: September 2018
Member
Hi DHS-ers,

I am using Haiti DHS 2016-17, with the HTPR70FL.DTA file in Stata.

Please can you explain the logic for the questionnaire skipping sh102 for all sources of drinking water except bottled water? This is a real shame as data for sh102 would be really important for other purposes. Often people drink from one source and use another source for other purposes, and it's important to know this for everyone, not just bottled water.

cheers,
Ian

[Updated on: Thu, 03 October 2019 10:38]

Report message to a moderator

Re: Haiti water DHS 2016 - why skip sh102 except for bottled water? [message #18251 is a reply to message #18172] Mon, 21 October 2019 06:48 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 1701
Registered: February 2013
Senior Member

Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

When you see a situation like this, you should look at the skip pattern of the questionnaire. I did a cross-tabulation of the two variables, including the NA (not applicable) cases with this command: tab sh101 sh102,m. (For this purpose you don't need to include the weights.) It's clear that the question about source of water for cooking or handwashing was only asked of respondents who said that their source of drinking water was bottled water (sh101=81).

hv102 is a duplicate of sh102. hv102 is probably labeled in English just because it is a core variable that appears in all surveys. Variables with labels that begin with "s" are survey-specific variables. I really don't know why sh102 was included as a survey-specific variable when it was also a core variable. The question was not asked twice--this is just a duplication during data processing.
Re: Haiti water DHS 2016 - why skip sh102 except for bottled water? [message #18254 is a reply to message #18251] Mon, 21 October 2019 08:09 Go to previous messageGo to next message
ianross is currently offline  ianross
Messages: 3
Registered: September 2018
Member
thanks Bridgette / Tom,

I understand the skip pattern. My point is that by having this skipping pattern we are losing potentially valuable information about the water market. For example, if people use a private reserve osmosis kiosk for drinking water (as is v common in urban Haiti, especially Port-au-Prince) then it is still important/useful to know what they use for other domestic purposes, e.g. whether they use the public network or not. If DHS is already taking up questionnaire space for this Q for people using bottled water, why not just ask it of everybody? The marginal questionnaire space is zero and the marginal questionnaire time is possibly 10-15 seconds. The data would be very useful from a water security perspective. Drinking water is not everything. People still have to collect and pay for water for other purposes.

thanks
Ian
Re: Haiti water DHS 2016 - why skip sh102 except for bottled water? [message #18332 is a reply to message #18254] Mon, 11 November 2019 09:29 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 1701
Registered: February 2013
Senior Member

Following is a response from DHS Research & Data Analysis Director, Tom Pullum:

Thank you for the comment. I agree that the skip was not desirable, but it would have been approved by an advisory committee of local experts (as in every country). I will make sure that the survey manager sees it.

Previous Topic: Senegal 2012 regions
Next Topic: Timor Leste - Blood Pressure For Female
Goto Forum:
  


Current Time: Mon Nov 18 15:06:14 Eastern Standard Time 2019