Strange data from DR Congo survey2013/2014 [message #8317] |
Tue, 06 October 2015 13:20 ![Go to next message Go to next message](/theme/default/images/down.png) |
DHS user
Messages: 111 Registered: February 2013
|
Senior Member |
|
|
I am currently working with the DR Congo survey data from 2013/2014, and especially the domestic violence module. When I look at the data on the question asking respondent if she has ever been forced to perform unwanted sexual acts (variable coded d125), I find that 690 of totaly 6492 answer yes. Then I look at the variable on age at first forced sexual act (variable d126), and I find that 1685 respondents (out of a total 6501 respondents) provide an answer to that question. I find this strange. When I look at those two variables together, many of those who answered no to the question represented in variable d125 still reports an age for first time they were forced.
I also detect that only those who answered yes on "ever been forced to perform unwanted sexual acts", are the only ones who are also asked who the perpetrator was the first time she was forced (variable d127).
I just want to make sure I use the right variable when I count how many respondents have experienced forced sexual acts, so if you know why this seemingly irregularity between the variables happen, or have any advice on whether I should count the number of respondent providing an age in d126, or just count the number of "yes"-answers in d125.
|
|
|
Re: Strange data from DR Congo survey2013/2014 [message #8318 is a reply to message #8317] |
Tue, 06 October 2015 13:23 ![Go to previous message Go to previous message](/theme/default/images/up.png) |
Bridgette-DHS
Messages: 3151 Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS Stata Specialist, Kerry MacQuarrie:
It is most helpful to examine the variables in conjunction with reading the questionnaire. There are, in fact, three relevant sexual violence variables, as follows:
D108 (itself an aggregation of d105h, d105i, and d105k)--this is from questions 1305h-j, which asks only ever-married women in the domestic violence module about their experience with sexual violence perpetrated by the current or most recent husband/partner.
D125 (an aggregation of s1323 and s1324)--this if from questions 1323 and 1324 which is asked of all women in the domestic violence module about their experience with sexual violence perpetrated by someone other than a husband/partner (s1323 is the question for ever married women, s1324 is the same question for never married women)
D130b--this is from question 1315b which is asked of ever-married women in the domestic violence module about their experience with sexual violence perpetrated by a previous husband/partner.
Anyone with a positive response to any of these three variables has a response for d126, age at first forced sexual act. Conversely, there are no responses on d126 if there is no response to all three variables d108, d125, and d130b. This is why there are more responses for d126 than 'yes' responses for d125.
I would also check your numbers, as mine are slightly different (using file CDIR61FL.DTA and d005 as the weight). You may have restricted your sample in some way. For example, I get:
766 of 6800 answering yes to d125;
1,295 of 5,078 answering yes to d108;
183 of 770 answering yes to d130b;
and 1,716 numerical answers and 200 'don't know' answers (total 1,916) on d126.
I hope this is helpful.
|
|
|