The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Data » Dataset use in SPSS » Analyzing amenorrhea and contraceptive use patterns in SPSS
Analyzing amenorrhea and contraceptive use patterns in SPSS [message #77] Wed, 20 February 2013 12:01 Go to next message
DHS user is currently offline  DHS user
Messages: 95
Registered: February 2013
Senior Member
I want to replicate Tables 7.8 and 7.9 and I am using the UGIR52FL dataset. This is ideally to help me understand how this variable (postpartum amenorrhoeic) was computed. After this, am interested in getting the pattern of amenorrhoeic women in the last three years which I would like to cross tab with their contraceptive use.... to understand when these women resumed contraceptive use if at all they did.
Re: Analyzing amenorrhea and contraceptive use patterns in SPSS [message #78 is a reply to message #77] Wed, 20 February 2013 12:03 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 1626
Registered: February 2013
Senior Member
Here is a response from one of our DHS experts Sarah Bradley, that should answer your question.

If you are analyzing women's amenorrhea and contraceptive use patterns, you are using the correct file, the women's Individual Recode file.   But the tables you're trying to replicate are all Births in the last 3 years, which is difficult to calculate from the women's file.  Using the woman-based (rather than birth-based, or M$ variables) for postpartum amenorrhea is fairly straightforward, so I'll explain that first.  Trying to match DHS tables is a great thing to do, though, so I'll explain how to match table 7.8 from the births file next and attach some SPSS syntax.

From the IR file, if you want the percentage of women who are amenorrheic among those women who gave birth in the last 36 months, you will need to

1. select those for whom v222<=35 (months since last birth begin from 0, so 0-35 = 36 months). 
2. Weight the data by v005/1000000.
3. You can then use v405, currently amenorrheic, to capture the percentage of women who are currently amenorrheic. 

From there, you can crosstab v405 with contraceptive use and continue on your analysis!

If you do want to match tables 7.8 and 7.9, The easiest thing to do is use the UGKR52 or UGIR52 files, which are respectively the Kids Recode (births in the 5 years prior to survey) or Births Recode (all births) files.  From the KR file, the steps to replicate the percentages amenorrheic, abstaining, and insusceptible are:

1. Weight the data by v005/1000000.
2. Calculate the age of children in months, and select only births in the last 36 months.
3. Exclude multiple births by selecting only singleton births or the last of multiple births (b0<>1).
4. Calculate the percentage amenorrheic among only most recent births (MIDX=1).

I'm attaching some code that shows how to do this.  You can read more about these calculations, and the description of smoothed averages used to calculate the median duration of amenorrhea, in the Guide to Statistics: -questionnaires-and-manuals.cfm.

I hope this helps.


[Updated on: Mon, 18 March 2013 09:09]

Report message to a moderator

Next Topic: Transforming variables to combine responses in SPSS
Goto Forum:

Current Time: Wed Jun 19 11:30:30 Eastern Daylight Time 2019