The DHS Program User Forum       Discussions regarding The DHS Program data and results
Home » Countries » India » WEIGHTS IN NFHS1 and NFHS2
WEIGHTS IN NFHS1 and NFHS2 Fri, 16 June 2017 07:19
 gauravdhamija Messages: 6Registered: February 2017 Location: UTTAR PRADESH, INDIA Member
Hi!

I am trying to find average height of the INDIAN kids in the age of 0-2-year-olds in both NFHS 1 AND NFHS 2. In order to do this, i am using following (hv005 in NFHS1 and v005 in NFHS2) as the weight. It would be great if you could guide me on this. I am pasting my stata code also.

In NFHS 1 (1992-93)

sum height if v024 ==13 & b8 >=0 & b8 <= 2 & b4==2 [iw= hv005/1000000 ]

sum height if v024 ==6 & b8 >=0 & b8 <= 2 & b4==2 [iw= hv005/1000000 ]

In NFHS 2 (1998-99)

sum height if v024 ==13 & b8 >=0 & b8 <= 2 & b4==2 [iw= v005/1000000 ]

sum height if v024 ==6 & b8 >=0 & b8 <= 2 & b4==2 [iw= v005/1000000 ]

GAURAV DHAMIJA
Re: WEIGHTS IN NFHS1 and NFHS2 [message #12575 is a reply to message #12574] Fri, 16 June 2017 08:32
 Bridgette-DHS Messages: 1318Registered: February 2013 Senior Member
Following is a response from Senior DHS Stata Specialist, Tom Pullum:

These Stata lines will summarize the height variable for girls age 0-2 in states 6 and 13. They are correct, except that you must be sure, when you construct "height", from hw3, that you have omitted the missing value codes such as 9999. For those cases, "height" should be coded ".". I would also strongly recommend that you omit children for whom the WHO HAZ score, hw70, (height for age) has been flagged. Some values of hw3 are implausible and are flagged in the construction of the HAZ score. The flagged cases include the cases with hw3=9999, so if you omit the flagged cases, you will not have to explicitly omit the 9999 cases.

Thus I would change your first line to be the following, and would change all the others in the same way.

sum height if v024 ==13 & b8 >=0 & b8 <= 2 & b4==2 & hw70<600 [iw= hv005/1000000 ]
Re: WEIGHTS IN NFHS1 and NFHS2 [message #12576 is a reply to message #12575] Fri, 16 June 2017 08:51
 gauravdhamija Messages: 6Registered: February 2017 Location: UTTAR PRADESH, INDIA Member
Thank you so much for this suggestion.

I got all the relevant variables by appending the variables in the following file: IAIR42FL.DTA. As per your suggestion, i looked at the 'hw70' variable in this file but I could not find it.

I already converted all the values with 9999 into missing. But in order to follow your above suggestion, i need hw70 variable.

GAURAV DHAMIJA
Re: WEIGHTS IN NFHS1 and NFHS2 [message #12577 is a reply to message #12576] Fri, 16 June 2017 09:50
 Bridgette-DHS Messages: 1318Registered: February 2013 Senior Member

Following is another response from Senior DHS Stata Specialist, Tom Pullum:

To get hw70-hw72, you could merge the BR files with the HW files for those surveys. The corresponding HW files for NFHS 1 and 2 are IAHW22FL.DTA and IAHW43FL.DTA. They can be downloaded once you are logged in to your account. Do you know how to do these merges?

Alternatively, it will be easier if you just use the earlier version of the HAZ score, which is given by hw5. In terms of identifying flagged values there should be very little difference, if any, from the WHO 2006 version of the HAZ, which is given by hw70. That is, where I put "hw70<600", just use "hw5<600". If you ever need to use them, hw71 is the updated version of hw8, and hw72 is the updated version of hw11.
Re: WEIGHTS IN NFHS1 and NFHS2 [message #12578 is a reply to message #12577] Fri, 16 June 2017 11:28
 gauravdhamija Messages: 6Registered: February 2017 Location: UTTAR PRADESH, INDIA Member
Thanks !

It would be great if you could share the information via which I can merge the two files.

GAURAV DHAMIJA
Re: WEIGHTS IN NFHS1 and NFHS2 [message #12579 is a reply to message #12578] Fri, 16 June 2017 13:17
 Bridgette-DHS Messages: 1318Registered: February 2013 Senior Member
Following is another response from Senior DHS Stata Specialist, Tom Pullum:

If you don't know how to do a merge, then it will be best to use hw5 rather than hw70. There is virtually no difference in the flagging criteria for those two variables. At the time of those two surveys, hw70-hw72 did not even exist. Merging instructions are provided in several earlier forum posts, if you prefer to go that way.

Re: WEIGHTS IN NFHS1 and NFHS2 [message #12894 is a reply to message #12579] Tue, 01 August 2017 08:52
 gange026 Messages: 2Registered: January 2017 Member
To use HW70 without merging the HW files (Height and Weight Scores - WHO Child Growth Standards), you might try using IPUMS-DHS (www.idhsdata.org) to create your dataset. IPUMS-DHS has merged the HW data with the KR files if that information was not already included in the KR data.

To create a dataset in IPUMS-DHS:

2. Click SELECT DATA, and then select CHILDREN as your unit of analysis.

3. Click the SELECT SAMPLES button and choose India 1992 and India 1998. Click "SUBMIT SAMPLE SELECTIONS."

4. Under SELECT VARIABLES, select TOPICS, and click on CHILD ANTHROPOMETRY. This will bring you to a list of child anthropometric variables available in the samples you have selected. The IPUMS-DHS variable names will show up as the default. By clicking on "ORIGINAL DHS VARIABLE NAMES," you can switch to the DHS names from the DHS recode. The variable mentioned in this thread, hw70 (HWHAZWHO), is available in almost every sample in IPUMS-DHS, including all Indian samples.

5. To add a variable to your data file, click the plus sign next to it.

6. When you have added all the variables you want in your dataset, click VIEW DATA CART in the upper right of the screen. You'll see a list of the variables you've selected. There will also be some preselected variables in the list, such as sample weight.

7. Click CREATE DATA EXTRACT. The third line on the next page will read CHANGE DATA FORMAT. You will probably want to click "Change" and select "STATA." You can describe the dataset if you want, and then click SUBMIT EXTRACT. You will soon get an email saying your dataset is ready. You can then click on the link under FORMATTED DATA to download the zipped file. You will have to unzip the data to use it.

IPUMS-DHS is funded by the National Institute of Child Health and Development (Grant No. R01HD069471).

-IPUMS-DHS staff

 Previous Topic: Description of variables Next Topic: Respondents Sibling Information
Goto Forum:

Current Time: Wed Apr 25 16:16:55 Eastern Daylight Time 2018