The DHS Program User Forum
Discussions regarding The DHS Program data and results
Home » Topics » General » Computing Education Indicators using STATA
Computing Education Indicators using STATA [message #167] Wed, 20 March 2013 12:14 Go to next message
ladokofie
Messages: 1
Registered: March 2013
Location: Manchester, UK
Member
How can I compute the following education indicators from the DHS household member file using STATA?

1. Net School Attendance Rates (i.e. primary or secondary school attendance rates)
2. Out-of-school Rates
3. School completion Rates (i.e. primary or secondary school completion rates)
4. Average Years of Education.

Do-file for above calculations will be helpful and highly appreciated.
============================================================ ==========================================
I have been struggling with how to generate or compute the following education indicators from the DHS household member recode using STATA for further analysis of educational inequality in Ghana:

1. Net School Attendance Rates (i.e. primary or secondary school attendance rates) 2. Out-of-school Rates 3. School completion Rates (i.e. primary or secondary school completion rates) 4. Average Years of Education.

I know that the indicators can be computed or generated from HV106 to HV110 and HV121 to HV129 (education variables in DHS household member recode) with the household members age variable (HV105). However, I don't know how to go about it using STATA. I would want to generate these education indicators in STATA to enable me to use them for further analysis in STATA (i.e. using these indicators as dependent variables in regression analysis and Concentration Index Decomposition Analysis of Educational Inequality).

I have put this on MEASURE DHS User Forum but so far no reply.

A Do-file for generating and/or calculating the rates above will be helpful and highly appreciated.

I would be very grateful if you could assist me.

Thank you.


[Updated on: Thu, 04 April 2013 14:19] by Moderator

Report message to a moderator

Re: Computing Education Indicators using STATA [message #258 is a reply to message #167] Thu, 04 April 2013 14:18 Go to previous messageGo to next message
Bridgette-DHS is currently offline  Bridgette-DHS
Messages: 3017
Registered: February 2013
Senior Member
I am attaching a Stata syntax file that will run the tabulations you are interested in. Please note, that education is not my primary area of research. Therefore, there may be some standards in the definitions of attendance, out-of-school, and completion rates that I am unfamiliar with. For example, these rates may be restricted to a certain age range that relates to the age/number of years in the primary or secondary schooling system. I have included examples of how to restrict the age where I thought it may be relevant, but you may need to change the age cut points, as the ones in my examples are only my best guess.

I also include code to weight the data when producing the tabulations as a matter of practice. This will have minimal effect if you are simply interested in the overall rate (the unweighted and weighted proportions are nearly identical), but becomes more important if you want to examine the rates by region, urban/rural area, or other factor.

Finally, I have written this syntax assuming you are using the 2008 Ghana DHS. If you are using a different year, you may want to verify that the variable names have remained the same.

Bridgette-DHS
  • Attachment: Ghana_edu.do
    (Size: 2.48KB, Downloaded 1861 times)
Re: Computing Education Indicators using STATA [message #428 is a reply to message #258] Tue, 14 May 2013 05:17 Go to previous messageGo to next message
dhiman is currently offline  dhiman
Messages: 3
Registered: May 2013
Member

Thank you so much for the codes - I was stuck with this for a very long time (so thanks to the OP as well).
However, using Indian Data for 2005-06, I could not replicate the Table 2.9 (School attendance by state) on page 33 of

http://www.measuredhs.com/pubs/pdf/FRIND3/FRIND3-Vol1AndVol2 .pdf

Any suggestions ?

Also any idea why variables like hv121 are coded as no(0), 2 and 9?

Best Regards
Re: Computing Education Indicators using STATA [message #480 is a reply to message #428] Mon, 27 May 2013 07:34 Go to previous messageGo to next message
elan12 is currently offline  elan12
Messages: 1
Registered: May 2013
Location: United States
Member
I am in the field of research too and I have been working on a similar assignment. Thanks for sharing great insight here.

[Updated on: Tue, 25 June 2013 11:58] by Moderator

Report message to a moderator

Re: Computing Education Indicators using STATA [message #742 is a reply to message #428] Fri, 06 September 2013 07:33 Go to previous messageGo to next message
Liz-DHS
Messages: 1516
Registered: February 2013
Senior Member
Dear User,
HV121 in our standard recode dictionary has the following value labels:

No 0
Currently attending 1
Attended at some time 2
Missing 9
Not applicable

For additional information on how our standard country report tables are calculated, please take a look at our Guilde to DHS Statistics http://www.measuredhs.com/pubs/pdf/DHSG1/Guide_to_DHS_Statis tics_29Oct2012_DHSG1.pdf and our Standard Recode Manual(s) http://www.measuredhs.com/pubs/pdf/DHSG4/Recode6_DHS_22March 2013_DHSG4.pdf

Thank you!
Re: Computing Education Indicators using STATA [message #1248 is a reply to message #742] Mon, 27 January 2014 14:26 Go to previous messageGo to next message
Liz-DHS
Messages: 1516
Registered: February 2013
Senior Member
Dear User,
In CSPro the table definitions for Table 2.9 School attendance by state on page 33 of the final report are as follows:

crosstab float(1) t205a total+hhage6+hhage7+north+central+east+noreast+west+south
hhsex*(hv025w1+total)+hv025w1+total
exclude(rowzero,colzero,percents,totals,specval)
title( "Table 2.9 School attendance by state"," ",
"Percentage of de facto household population age 6-17 years attending",
" school in the 2005-06 school year by sex, residence, age, and state,",
" India, 2005-06" )
stub( "State" );

Row variables"
Total Total
hhage6 Age
1 6-10 years
2 11-14 years
3 15-17 years
hhage7 Age
1 6-14 years
2 6-17 years
North
Delhi 1
Haryana 2
Himachal Pradesh 3
Jammu & Kashmir 4
Punjab 5
Rajasthan 6
Uttaranchal 7

Central
Chhattisgarh 1
Madhya Pradesh 2
Uttar Pradesh 3

East
Bihar 1
Jharkhand 2
Orissa 3
West Bengal 4

Noreast
Arunachal Pradesh 1
Assam 2
Manipur 3
Meghalaya 4
Mizoram 5
Nagaland 6
Sikkim 7
Tripura 8

West
Goa 1
Gujarat 2
Maharashtra 3

South
Andhra Pradesh 1
Karnataka 2
Kerala 3
Tamil Nadu 4


Column variable:
hhsex
1 male
2 female

Layer variable:
(hv025w1+total)
1 urban
2 rural

hv025w1
1 urban
2 rural
Total


t205aw is a working table
crosstab float(1) t205aw total+hhage6+hhage7+north+central+east+noreast+west+south
hhsex*(hv025w1+total)+hv025w1+total
exclude(rowzero,colzero,percents,totals,specval)







Re: Computing Education Indicators using STATA [message #8250 is a reply to message #167] Tue, 22 September 2015 05:05 Go to previous messageGo to next message
doc14163 is currently offline  doc14163
Messages: 3
Registered: August 2015
Member
Dear Users
I am trying to compute education indicators(completion, enrollment etc) using stata and DHS cameroon.
How ever i have some difficulties. I can equally not download a sample code posted here.
I don't know if anybody can post a sample code
Re: Computing Education Indicators using STATA [message #8444 is a reply to message #8250] Wed, 28 October 2015 09:47 Go to previous message
Trevor-DHS is currently offline  Trevor-DHS
Messages: 787
Registered: January 2013
Senior Member
Doc14163 - you say you are having some difficulties, but that is very vague. Please explain the specific difficulties you are having and post your code, and then someone may be able to provide assistance.
Previous Topic: using svy function to run cross tabs
Next Topic: PSUs in earlier versions of DHS
Goto Forum:
  


Current Time: Fri Mar 29 10:03:03 Coordinated Universal Time 2024