The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Countries » Ethiopia » Findings Inconsistency
Findings Inconsistency [message #14734] Wed, 02 May 2018 17:28 Go to next message
sko16 is currently offline  sko16
Messages: 21
Registered: June 2016
Location: Addis Ababa, Ethiopia
Dear DHS Experts,
I used the individual (women) the 2016 Ethiopian DHS data set to look at ANTENATAL CARE COVERAGE particularly, Timing of ANC Visits. It is reported on page 134 on the final report that:
Only 20% of women had their first ANC during the first trimester,
26% during their fourth to fifth month of pregnancy, and
14% during their sixth to seventh month of pregnancy.
Two percent of women did not receive any ANC until the eight month of pregnancy or later.
According to WHO classification:
1st trimester/ first ANC- is before week 12
2nd trimester close to week 26
3rd trimester around week 32
4th trimester around week 38
Inaddition, the EDHS data set values shows from 0 to 9 months. While, I categorize according to these WHO protocol, and analyse the 2016 EDHS using stata, I found that:
The first ANC visits: 32.4% (by recategorizing 0 - 3 months)
2nd trimester visits: 56.6% (by re categorizing 4 - 6 months)
3rd trimester: 9.5% (by re categorizing 7 - 8 months)
4th trimester: 1.5% (by re categorizing More than 8 months)
Would you please, help me where would be the problem for this specific variable analysis.
Thank you so much in advance for your usually expertise advises.
Re: Findings Inconsistency [message #14946 is a reply to message #14734] Fri, 18 May 2018 12:07 Go to previous message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3064
Registered: February 2013
Senior Member
Following is a response from DHS Senior Research Associate, Lindsay Mallick:

Dear user,

It does not seem that your categorization of months/trimesters aligns with the categories presented in Table in 9.2 on timing of ANC. I also wonder if your denominator matches.
You can refer to this blog post and corresponding video for more considerations when matching tables:

Here is some code to match Table 9.2:

gen wt = v005/1000000

recode m13_1 (0/3 = 1 " <4") (4/5 = 2 " 4-5") (6/7 = 3 " 6-7") (8/9 = 4 " 8+") (else = 9 "Don't know/missing") if v208>0, gen(anctime) 
replace anctime = 0 if m14_1 ==0, label 0 "None"
label define anctime 0 "No ANC", add
label var anctime "Timing of ANC"
ta anctime [iw=wt]

Previous Topic: Ethics Approval - Ethiopia 2011
Next Topic: Women Empowerment Index
Goto Forum:

Current Time: Thu May 23 10:59:44 Coordinated Universal Time 2024