The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » Child Health » stunting variable
stunting variable [message #3682] Mon, 26 January 2015 09:06 Go to next message
adotei is currently offline  adotei
Messages: 3
Registered: January 2015
Member
Hi,
I am working on child malnutrition and need a measure for stunting. I need clarification as to the differences between hc70_1/8 in the GHHR5AFL data file (i.e. the 2008 GDHS household dataset). Whereas the recode file describes only HC70, the actual data in stata shows hc70_1, hc70_2 ... to hc70_8. I would be glad if you could explain or describe why these are the case. Thank you.

[Updated on: Mon, 26 January 2015 09:58]

Report message to a moderator

Re: stunting variable [message #3685 is a reply to message #3682] Mon, 26 January 2015 22:13 Go to previous messageGo to next message
Reduced-For(u)m
Messages: 292
Registered: March 2013
Senior Member

I believe you are using the wrong (read: less optimal) dataset. The "_01" variables usually refer to the birth order when you have multiple children per observation. My guess is you are using one of the household-level recodes, where a household (or a woman interviewed) is the unit of observation (the row in your spreadsheet). Better would be to use the Child recode which has a child as the unit of observation. In that recode there should only be 1 value of HC70 - the one for that particular child.

That help?
Re: stunting variable [message #3688 is a reply to message #3685] Tue, 27 January 2015 10:44 Go to previous messageGo to next message
adotei is currently offline  adotei
Messages: 3
Registered: January 2015
Member
I am truly grateful for that. I thought with the child recode file, I would not be able to access as much data points as with the household recode file. That is why I resorted to the Household file. But I think that is not the case. Could you kindly confirm if indeed using the household file allows for more data points than using the child file?

Secondly, I wish to replicate the procedure used in arriving at the height-for-age Z score (HAZ). I tried the WHO Anthro Software and then with Stata without success. Are you able to help with some suggestions? I need to compute HAZ from a dataset for comparative analysis.

Thank you.
Re: stunting variable [message #3689 is a reply to message #3685] Tue, 27 January 2015 10:53 Go to previous messageGo to next message
adotei is currently offline  adotei
Messages: 3
Registered: January 2015
Member
Again, I would love for you to throw more light on the "_01" variables. To illustrate, how do we interpret hc70_10 or how do we distinguish between say hc70_1 and hc70_2 (etc)?

Thank you.
Re: stunting variable [message #3693 is a reply to message #3689] Wed, 28 January 2015 12:36 Go to previous messageGo to next message
Liz-DHS
Messages: 1516
Registered: February 2013
Senior Member
Dear User,
Below is a response from another post which should help clarify multiple occurrence variables in packages such as SPSS. This was previously posted by Bridgette James, our archive specialist.
Quote:
Multiple or repeating records are placed one after the other on the record, with the maximum number of occurrences of each section being represented in every case. Each variable in a repeating section is placed immediately after the preceding variable of the same occurrence, such that all variables for occurrence 1 precede all variables for occurrence 2 of a section.

Multiple occurring variables and sections of data represent the main disadvantage to flat files. Each occurrence of every such variable must have its own name because statistical packages do not generally support the use of arrays or subscripts. For example, the third occurrence of the variable named MM11 would be named MM11$03 in SPSS, or MM11_03 in SAS and STATA.



In addition to this, you may want to access our website and read the section on "Using Datasets for Analysis" http://dhsprogram.com/data/Using-Datasets-for-Analysis.cfm. Here you will find a lot of useful information including several instructional videos.
Re: stunting variable [message #3782 is a reply to message #3688] Tue, 10 February 2015 14:48 Go to previous message
Reduced-For(u)m
Messages: 292
Registered: March 2013
Senior Member


For calculating z-scores in Stata I like the package "zscore06". You just put in height (cm), age (months, with decimals ok), and gender and it spits out WHO z-scores.

And in case you still weren't clear on the "_01" the var X_01 would refer to the first child born (or last, I am not sure, you could check by birthdate), the next child would be X_02, etc. It is just indexing the child (or household member) number, since some households have heights for multiple children. The "child recode" would have each child as its own observation, so there is no need for the "_01" or "_02" indexing.
Previous Topic: Distances to Health Facilities
Next Topic: How to transform children's height into standardized height
Goto Forum:
  


Current Time: Thu Mar 28 13:36:24 Coordinated Universal Time 2024