| Figure 4.4 Trends in early marriage [message #27993] |
Tue, 31 October 2023 11:17  |
sokiya
Messages: 79
Registered: May 2017
Location: Nairobi
|
Senior Member |
|
|
Hi
I am trying to replicate the analysis - Figure 4.4 Trends in early marriage in KDHS 2022 main report using the attached do file whose content is from the do files in the GitHub repository. However, I can't seem to arrive at the value of 25%. Any advice is appreciated. Thanks in advance!
|
|
|
|
| Re: Figure 4.4 Trends in early marriage [message #28018 is a reply to message #27993] |
Thu, 02 November 2023 08:43  |
Bridgette-DHS
Messages: 3199
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS staff member, Tom Pullum:
It will be easier if you refer to numbers in tables, rather than figures. I believe the number you are looking for is in Table 4.4 of the report, first panel (women), the column for age 18, the row for age range 25-49. The number in the table is 25.4%, which is rounded to 25% in Figure 4.4.
The Stata lines given below calculate the mean (for women age 25-49) of a binary variable that is 100 if the woman's age at first cohabitation (v511) was less than 18 and 0 otherwise. This mean is the percentage of women married by 18. The calculated value is 25.36749, which rounds to 25.4% in the table or 25% in the figure.
The GitHub program does basically the same calculation for all the cells of table 4.4. The calculation of median age at first union is trickier because it requires interpolation, but the GitHub program shows how to do that.
use "...KEIR8BFL.DTA", clear
gen married_by_18=0
replace married_by_18=100 if v511<18
summarize married_by_18 if v012>=25 & v012<=49 [iweight=v005/1000000]
|
|
|
|
| Re: Figure 4.4 Trends in early marriage [message #28023 is a reply to message #28018] |
Fri, 03 November 2023 05:48 |
sokiya
Messages: 79
Registered: May 2017
Location: Nairobi
|
Senior Member |
|
|
Thanks so much for the guidance. I sincerely appreciate. I realize if v012>=25 & v012<=49 is important if one were to use the syntax found on the GitHub repo
Will be referring to tables instead of figures moving forward.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
