Home » Countries » India » Mismatch in educational attainment and education in single years (Mismatch in educational attainment (mv/v149) and education in single years (mv/v133) in the 2019-21 recode file for India.)
| Mismatch in educational attainment and education in single years [message #28449] |
Tue, 09 January 2024 07:08  |
Navya
Messages: 8
Registered: May 2022
|
Member |
|
|
Dear all,
I am currently working with the men's and individual's recode files for the 2019-21 survey wave for India, and I noticed that there may be some inconsistency between the educational attainment variable (mv/v149) and education in single years variable (mv/v133) for both files. From what I understand, (m)v133 gives the single years of education for the individual, going from 0 years to 20 years for both files, and (m)v149 categorizes these years into "no education", "incomplete primary", "incomplete secondary", "complete secondary", and "higher". According to this, individuals who completed 11 years of education should be categorized into "incomplete secondary" and those with 12 years should be categorized into "complete secondary". However, this result seems to be flipped (individuals with 11 years are being categorized into "complete secondary" and those with 12 years are under "incomplete secondary" for both files). I have attached a screenshot of the tabulate result I get in Stata for the individual's recode file. I only see this result in the 2019-21 survey datasets for India, and not in the older datasets. Am I missing something here?
Thank you for your help.
|
|
|
|
| Re: Mismatch in educational attainment and education in single years [message #28452 is a reply to message #28449] |
Tue, 09 January 2024 09:20  |
Bridgette-DHS
Messages: 3190
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS staff member, Tom Pullum:
I'll describe the variables for men in the MR file, with the "mv" prefix. The same applies for women in the IR file, with the "v" prefix.
There are four relevant variables for education:
. describe mv106 mv107 mv133 mv149
storage display value
variable name type format label variable label
-----------------------------------------------------------------------------------------
mv106 byte %8.0g MV106 educational level
mv107 byte %8.0g MV107 highest year of education (at level in mv106)
mv133 byte %8.0g MV133 total number of years of education
mv149 byte %8.0g MV149 educational attainment
mv149 and mv107 are almost exactly the same. The only difference is that mv149 divides primary into incomplete and complete:
. tab mv149 mv106
educational | educational level
attainment | no educat primary secondary higher | Total
---------------------+--------------------------------------------+----------
no education | 12,269 0 0 0 | 12,269
incomplete primary | 0 11,710 0 0 | 11,710
incomplete secondary | 0 0 55,807 0 | 55,807
complete secondary | 0 0 4,211 0 | 4,211
higher | 0 0 0 17,842 | 17,842
---------------------+--------------------------------------------+----------
Total | 12,269 11,710 60,018 17,842 | 101,839
mv133 is a recode of mv106 and mv107. The following table gives the mean of mv133 for each combination of mv106 and mv107. (Ignore the "Total" column and row.)
. tab mv107 mv106, summarize(mv133) means
Means of total number of years of education
highest |
year of |
education |
(at level | educational level
in mv106) | primary secondary higher | Total
-----------+---------------------------------+----------
no years | 0 . . | 0
1 | 1 6 13 | 8.5814143
2 | 2 7 14 | 8.1771985
3 | 3 8 15 | 10.65447
4 | 4 9 16 | 8.6599261
5 | 5 10 17 | 9.5647399
6 | . 11 18 | 11.566783
7 | . 12 19 | 12.099424
8 | . . 20 | 20
-----------+---------------------------------+----------
Total | 3.9836038 9.5509347 15.147237 | 9.9378475
Hope this answers your question. Please let us know if it does not.
|
|
|
|
| Re: Mismatch in educational attainment and education in single years [message #28520 is a reply to message #28452] |
Wed, 24 January 2024 01:20 |
Navya
Messages: 8
Registered: May 2022
|
Member |
|
|
Hi,
Thank you so much for your quick response. I am still having some trouble reconciling the values for (m)v149 (educational attainment) with (m)v107 (highest year of education (at level in (m)v106)) and (m)v133 (total number of years of education).
. describe mv106 mv107 mv133 mv149
Variable Storage Display Value
name type format label Variable label
---------------------------------------------------------------------------------------------------------------------------------------------------
mv106 byte %8.0g MV106 educational level
mv107 byte %8.0g MV107 highest year of education (at level in mv106)
mv133 byte %8.0g MV133 total number of years of education
mv149 byte %8.0g MV149 educational attainment
. tab mv107 mv149
highest year of |
education (at level | educational attainment
in mv106) | incomplet incomplet complete higher | Total
----------------------+--------------------------------------------+----------
no years completed at | 239 0 0 0 | 239
1 | 366 2,470 0 2,071 | 4,907
2 | 1,172 4,915 0 2,237 | 8,324
3 | 1,636 10,124 0 9,066 | 20,826
4 | 2,455 11,970 0 1,004 | 15,429
5 | 5,842 13,488 0 2,797 | 22,127
6 | 0 0 4,211 371 | 4,582
7 | 0 12,840 0 185 | 13,025
8 | 0 0 0 111 | 111
----------------------+--------------------------------------------+----------
Total | 11,710 55,807 4,211 17,842 | 89,570
It seems to me that in the above table, the figure 4,211 should come under "incomplete secondary" and the figure 12,840 should come under "complete secondary". I see a similar result with mv133 and mv149 as well.
. tab mv133 mv149
total number |
of years of | educational attainment
education | no educat incomplet incomplet complete higher | Total
-------------+-------------------------------------------------------+----------
0 | 12,269 239 0 0 0 | 12,508
1 | 0 366 0 0 0 | 366
2 | 0 1,172 0 0 0 | 1,172
3 | 0 1,636 0 0 0 | 1,636
4 | 0 2,455 0 0 0 | 2,455
5 | 0 5,842 0 0 0 | 5,842
6 | 0 0 2,470 0 0 | 2,470
7 | 0 0 4,915 0 0 | 4,915
8 | 0 0 10,124 0 0 | 10,124
9 | 0 0 11,970 0 0 | 11,970
10 | 0 0 13,488 0 0 | 13,488
11 | 0 0 0 4,211 0 | 4,211
12 | 0 0 12,840 0 0 | 12,840
13 | 0 0 0 0 2,071 | 2,071
14 | 0 0 0 0 2,237 | 2,237
15 | 0 0 0 0 9,066 | 9,066
16 | 0 0 0 0 1,004 | 1,004
17 | 0 0 0 0 2,797 | 2,797
18 | 0 0 0 0 371 | 371
19 | 0 0 0 0 185 | 185
20 | 0 0 0 0 111 | 111
-------------+-------------------------------------------------------+----------
Total | 12,269 11,710 55,807 4,211 17,842 | 101,839
I am seeing similar results for the individual recode file as well. Please let me know if I'm missing something.
Thank you for your help!
|
|
|
|
| Re: Mismatch in educational attainment and education in single years [message #28524 is a reply to message #28520] |
Wed, 24 January 2024 13:31 |
Bridgette-DHS
Messages: 3190
Registered: February 2013
|
Senior Member |
|
|
Following is a response from Senior DHS staff member, Tom Pullum:
You are asking about the Kenya 2022 survey. An erratum is being issued on the coding of education in this survey.
I suggest that you look at the relationship among v106, v107, v133, and v149 in the previous survey, the Kenya 2014 survey, because the relationship was ok in that survey. In the 2014 survey, you will see how v133 is supposed to be constructed as a recode of v106 and v107. In the IR file, do this:
. tab v107 v106, summarize(v133) means
Means of education in single years
highest |
year of | highest educational level
education | primary secondary higher | Total
-----------+---------------------------------+----------
no years | 0 8 12 | 8.7214076
1 | 1 9 13 | 9.2506473
2 | 2 10 14 | 10.554584
3 | 3 11 15 | 9.9238281
4 | 4 12 16 | 11.115502
5 | 5 . 17 | 5.3082022
6 | 6 14 18 | 6.2527945
7 | 7 . 19 | 7.0332318
8 | 8 . 20 | 8.0017804
-----------+---------------------------------+----------
Total | 6.6739256 10.794648 14.333705 | 8.7562835"
If you enter "tab v106 v149" you will see that v149 is a recode of v106, which just distinguishes between whether the person completed the primary or secondary levels. That's based on the value of v107 within each level of v106.
That's all there is to it. v106 and v107 are the key variables; v133 and v149 are constructed variables and they were constructed incorrectly in the Kenya 2022 survey. You can either reconstruct them yourself, for your use, or wait for the next version of the Kenya 2022 data files.
|
|
|
|
|
|
| Re: Mismatch in educational attainment and education in single years [message #28974 is a reply to message #28449] |
Sat, 06 April 2024 15:06 |
fred.arnold@icf.com
Messages: 83
Registered: May 2021
|
Senior Member |
|
|
|
I think that the problem may be that the NFHS education questions do not follow the DHS standard in that there are no questions about the level of education in NFHS surveys. The only education questions are "What is the highest grade you completed?" (Not the highest grade completed at that level, since there are no levels). I'm not sure how the NFHS-5 variables are created from the available education questions, but they were probably based on the highest grade completed, with decisions being made about what level that grade might reflect.
|
|
|
|
Goto Forum:
Current Time: Mon Nov 11 04:28:54 Coordinated Universal Time 2024
|
The DHS Program User Forum
Members
Search
Help
Register
Login
Home
