The DHS Program User Forum: Reproductive Health » Reasons for Non-Use

Home » Topics » Reproductive Health » Reasons for Non-Use

Show: Today's Messages :: Show Polls :: Message Navigator

Switch to threaded view of this topic

Create a new topic

Submit Reply

Reasons for Non-Use [message #5604]

Mon, 15 June 2015 14:37

FG_KJurczynska is currently offline

FG_KJurczynska
Messages: 1
Registered: July 2014
Location: Washington, DC

Member

Greetings,

I'm hoping to reconcile some information concerning measurement of reasons for non-use of contraception, both among current and future (intended) non-users. I've found that Recode VI indicates that both metrics are "no longer part of the DHS VI core questionnaire, but the variable is kept in the DHS VI recode." We nonetheless find both questions in the round 6 questionnaire (Q709), though surveys from that period appear to inconsistently report on one or both indicators. Could you please explain why this is the case?

I checked the round VII questionnaire, and it does include reasons for non-use, this time in Q810. Does this mean that both metrics will be consistently reported moving forward?

Thanks for your help.

Best,
Kaja

[Updated on: Mon, 15 June 2015 14:39]

Report message to a moderator

Re: Reasons for Non-Use [message #6711 is a reply to message #5604]

Wed, 01 July 2015 14:15

Liz-DHS
Messages: 1516
Registered: February 2013

Senior Member

Dear User,
Here is a response from one of our data processing experts, Mr. Guillermo Rojas:
Quote:

In previous DHS phases (up to 5, I believe) we used to have single variable V376 "Main reason not to use a method", that variable was dropped from the DHS 6 questionnaire. In DHS 5 (Q707) and DHS 6 (Q709) there is a multiple answer question (Reason not using a method) which is stored in variables V3A08x, where x goes from A to Z. V376 was left in the DHS 6 recode definition in case a country wants to keep the variable.

Thank you for your post.

[Updated on: Thu, 02 July 2015 13:12]

Report message to a moderator

Re: Reasons for Non-Use [message #10828 is a reply to message #6711]

Thu, 22 September 2016 07:04

Rose is currently offline

Rose
Messages: 4
Registered: September 2016

Member

Dear Liz,

I was just wondering if you could explain what the rationale was behind dropping variable 376? Looking at many of the countries, they seem to not have data on reasons for not intending to use a method in the future. Is there a reason this variable is not used very much?

Thanks,
Rose.

Report message to a moderator

Re: Reasons for Non-Use [message #11114 is a reply to message #10828]

Fri, 28 October 2016 18:53

Trevor-DHS is currently offline

Trevor-DHS
Messages: 802
Registered: January 2013

Senior Member

Yes, the question was changed from a single response question (as was used for v376) to a multiple response question, for which the results are in v3a08a to v3a08x.

Report message to a moderator

Re: Reasons for Non-Use [message #15013 is a reply to message #6711]

Fri, 25 May 2018 04:54

amelia.plant@berkeley.edu is currently offline

amelia.plant@berkeley.edu
Messages: 7
Registered: April 2018
Location: Tunis, Tunisia

Member

Thank you, Liz.

I was wondering if you could clarify the denominator for variables V3A08A-Z. I have been trying to group them into categories for my analysis and I am having a hard time figuring out how to construct the variables.

Thanks so much.

Report message to a moderator

Re: Reasons for Non-Use [message #15021 is a reply to message #15013]

Fri, 25 May 2018 10:27

Trevor-DHS is currently offline

Trevor-DHS
Messages: 802
Registered: January 2013

Senior Member

The variables v3a08a-z are based on question 810 in the DHS-7 women's questionnaire. The base or denominator for this question is women who are not pregnant (v213 <> 1 [see filter 806 in the questionnaire]), are not currently using a method of contraception (v312 = 0 [see filter 807 in the questionnaire]), want no more children or want to wait at least two years for the next child (v605 = 5 or v605 = 2 [see filter 808], and have had sexual intercourse in the past year (100 <= v527 <= 399 [see filter 809]). In a few surveys the question 810 is also restricted just to currently married or in union women (v502 = 1).

Report message to a moderator

Re: Reasons for Non-Use [message #15026 is a reply to message #15021]

Fri, 25 May 2018 17:39

amelia.plant@berkeley.edu is currently offline

amelia.plant@berkeley.edu
Messages: 7
Registered: April 2018
Location: Tunis, Tunisia

Member

Thanks so much Trevor. When I use the following syntax, I still am a bit off. I register n = 4591 instead of n = 4567:

v213 != 1 & v312 == 0 & (v605 == 5 | v605 == 2) & & v527 >= 100 & v527 <= 399

Any idea where I am going wrong?

Report message to a moderator

Re: Reasons for Non-Use [message #15029 is a reply to message #15026]

Fri, 25 May 2018 23:25

Trevor-DHS is currently offline

Trevor-DHS
Messages: 802
Registered: January 2013

Senior Member

I have no idea which survey you are using. Can you tell me which survey and which dataset it is that you are analyzing?

Report message to a moderator

Re: Reasons for Non-Use [message #15032 is a reply to message #15029]

Sat, 26 May 2018 08:17

amelia.plant@berkeley.edu is currently offline

amelia.plant@berkeley.edu
Messages: 7
Registered: April 2018
Location: Tunis, Tunisia

Member

I am using Uganda DHS 2016, and STATA.

Report message to a moderator

Re: Reasons for Non-Use [message #15035 is a reply to message #15032]

Sat, 26 May 2018 13:37

Trevor-DHS is currently offline

Trevor-DHS
Messages: 802
Registered: January 2013

Senior Member

These difference here are cases where the answer to question 810 was missing, typically because of interviewer error or possibly refusal by the respondent. If you see the .FRQ file and look at V3A08A you will see these cases marked MISSING.

[Updated on: Sat, 26 May 2018 13:38]

Report message to a moderator

Re: Reasons for Non-Use [message #15049 is a reply to message #15035]

Mon, 28 May 2018 10:07

amelia.plant@berkeley.edu is currently offline

amelia.plant@berkeley.edu
Messages: 7
Registered: April 2018
Location: Tunis, Tunisia

Member

Got it. Thanks so much!

Report message to a moderator

Re: Reasons for Non-Use [message #25095 is a reply to message #15049]

Tue, 30 August 2022 18:52

Wityc is currently offline

Wityc
Messages: 4
Registered: June 2022

Member

Hello,

I am having some issues in creating the variable for non-use. I think the denominator could be the problem but would appreciate your help. One of the reasons for non use if infrequent sex - which I combined v3a08b (not having sex) and v3a08b (infrequent sex), as below. Could you please help if this is the correct way of defining the variable and importantly the denominator. Thanks

*infrequent sex
gen infreqsex=0
replace infreqsex=1 if (v3a08c==1|v3a08b==1)
ta infreqsex,m
svy:ta infreqsex agebin if v213 != 1 & v312 == 0 & (v605 == 5 | v605 == 2) & v527 >= 100 & v527 <= 399 & v502 == 1,percent format(%4.2f) col

Report message to a moderator

Re: Reasons for Non-Use [message #25169 is a reply to message #25095]

Thu, 08 September 2022 15:31

Janet-DHS is currently offline

Janet-DHS
Messages: 880
Registered: April 2022

Senior Member

Following is a response from DHS staff member Tom Pullum:

Are you referring to a specific survey and table?

Report message to a moderator

Re: Reasons for Non-Use [message #25172 is a reply to message #25169]

Thu, 08 September 2022 15:42

Wityc is currently offline

Wityc
Messages: 4
Registered: June 2022

Member

Thanks Tom,

Yes - I'm using Zambia DHS 2013-14. I get the results below using my syntax. But this is different from what I get from StatsCompiler i.e.
Total <30 30+
4.8 0.5 6.1

-------------
*infrequent sex
gen infreqsex=0
replace infreqsex=1 if (v3a08c==1|v3a08b==1)
ta infreqsex,m

. svy:ta infreqsex age30 if v213 != 1 & v312 == 0 & (v605 == 5 | v605 == 2) & v527 >= 100 & v527 <= 399 & v502 == 1 ,percent format(%4.2f) col
(running tabulate on estimation sample)

Number of strata = 20 Number of obs = 2,397
Number of PSUs = 674 Population size = 2,379,128,163
Design df = 654

----------------------------------
| RECODE of v012
| (respondent's current
| age)
infreqsex | <30 30+ Total
----------+-----------------------
0 | 83.53 87.13 85.66
1 | 16.47 12.87 14.34
|
Total | 100.00 100.00 100.00
----------------------------------
Key: Column percentage

Pearson:
Uncorrected chi2(1) = 6.0857
Design-based F(1, 654) = 4.0984 P = 0.0433

But this is different from what I get from StatsCompiler i.e.
Total <30 30+
4.8 0.5 6.1

Report message to a moderator

Re: Reasons for Non-Use [message #25195 is a reply to message #25172]

Mon, 12 September 2022 17:05

Janet-DHS is currently offline

Janet-DHS
Messages: 880
Registered: April 2022

Senior Member

Following is a response from DHS staff member Tom Pullum:

If you are trying to match a table in the Zambia report, please identify the table. If you are trying to match an indicator in STATcompiler, please give the full name of the indicator. You are talking about non-use but you are also talking about a very specific reason for non-use (infrequent sex) so there must be a table somewhere. We can help but we need more information.

Report message to a moderator

Re: Reasons for Non-Use [message #25270 is a reply to message #25195]

Mon, 26 September 2022 13:09

Wityc is currently offline

Wityc
Messages: 4
Registered: June 2022

Member

Thanks.

I am trying to match this indicator from STATCompiler:: "Reason for not intending to use contraception: Infrequent sex". I'm using Zambia DHS 2013-14. Based on my sytax below, the results I am getting are different from STATCompiler.

Please let me know if this is clearer

Thank you

Report message to a moderator

Re: Reasons for Non-Use [message #25289 is a reply to message #25270]

Wed, 28 September 2022 14:47

Janet-DHS is currently offline

Janet-DHS
Messages: 880
Registered: April 2022

Senior Member

Following is a response from DHS staff member Tom Pullum:

Thanks for being more specific. The binary outcome variable would be calculated as follows:

gen outcome=0 if (v501==1 | v501==2) & v313==0 & v362==5
replace outcome=1 if outcome==0 & v3a08c==1

In STATcompiler, the description of the outcome says that the denominator is all currently married women. Actually the denominator has to be further restricted to women who are not using and do not intend to use (v313==0 & v362==5). I have suggested that the online description be updated.

Report message to a moderator

Switch to threaded view of this topic

Create a new topic

Submit Reply

Previous Topic:	Matching most recent sexual activity (women) in India_2019/21
Next Topic:	Antenatal care visits by type of facilities

Goto Forum:

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

PDF

]

Current Time: Wed Nov 13 06:43:31 Coordinated Universal Time 2024