| Assessing inter-generational effect of malnutrition [message #9509] |
Wed, 06 April 2016 20:58  |
gizachew
Messages: 18
Registered: December 2015
Location: Australia
|
Member |
|
|
Dear DHS team
A group of research panel involving 2 public health specialist (one of whom is a nutritionist) and 1 Economist are seeking to conduct Ethiopian DHS data analysis using the data collected in three different periods (2000, 2005, 2011) so as to see the effects of the inter-generational effect of malnutrition. For this, the panels are proposing to compare the nutritional outcome of children born to mothers whose date of births were during 1984 and 1985 (historical known by its famine) to other group of women. In this regard, the famine was particularly concentrated in the northern part. Hence, we need to get the following information
1) We realized that the date of birth in 2011 data used Ethiopian calendar despite the first two used Gregorian calendar. Is that correct?
2) Is it possible to link the EA data to the women's data (IR file) so as to link the women to her place of residence (perhaps the challenge here is that we couldn't got assured that the current place of residence might not be their place of birth)
3) Does it sound to merge (append) the three children's data (KR file) so that we could do the analysis for large number of children.
Best regards,
G
|
|
|
|
|
|
| Re: Assessing inter-generational effect of malnutrition [message #9601 is a reply to message #9585] |
Wed, 20 April 2016 13:50  |
Trevor-DHS
Messages: 788
Registered: January 2013
|
Senior Member |
|
|
1) To my knowledge all three surveys used the Ethiopian calendar, but with minor adaptations to drop the 13th month. The years of interview in the dataset in v007 are 1992, 1997, and 2003 for the surveys in 2000, 2005, 2011 respectively.
2) They are already linked to the level available. See v024 and v025 for region and type of place of residence. If you need lower levels of information you can request to download the GPS data for each survey (see http://www.dhsprogram.com/data/available-datasets.cfm).
3) It is possible to pool the 3 surveys together, but pay attention to the weighting of data when doing that. You will need to consider the relative sample sizes at each point in time as part of the weighting needed.
|
|
|
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
