Project

General

Profile

Support #762

inconsistency between xwavedat and indresp on parental occupation?

Added by Carolina Zuccotti almost 6 years ago. Updated over 5 years ago.

Status:
Closed
Priority:
Normal
Category:
-
Start date:
04/08/2017
% Done:

100%


Description

Hello,

I am a bit puzzled with the parental information contained in the indresp file and in the xwavedat file.
The xwavedat variables are supposed to be based on indresp data, but I am not sure why I find the following inconsistencies. (Perhaps this is my mistake too)

Let me give an example, base don data I am using. I have added parental information in xwavedat to b_indresp.
I have the following variables:

. sum b_masoc90_cc b_masoc00_cc masoc00_cc masoc90_cc

Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
b_masoc90_cc | 1,734 54.04268 25.38314 10 99
b_masoc00_cc | 1,734 551.8899 260.5772 111 925
masoc00_cc | 19,497 580.8958 269.3696 111 925
masoc90_cc | 23,434 57.34847 26.26591 10 99

NOTE THAT I HAVE RECODED VALUES BELOW 0 TO MISSING.

I next do the following tests, with these results:

  • tab masoc00_cc b_hhorig = this only has values for the UKHLS sample
  • tab masoc90_cc b_hhorig = this has values for both UK and BHPS sample, but BHPS NI is missing
  • tab b_masoc00_cc b_hhorig = this has values for both UKHLS and BHPS sample
  • tab b_masoc90_cc b_hhorig = this has values for both UKHLS and BHPS sample

I would have expected that the variables from the xwavedat include more cases, and not less.

But there is even a more strange thing in this data:

tab b_masoc90_cc if masoc90_cc==.

I would have expected a "no observations" sign, assuming that masoc90_cc is based on the indresp files (as it is stated here: https://www.understandingsociety.ac.uk/documentation/mainstage/dataset-documentation/wave/xwave/datafile/xwavedat/variable/masoc00_cc); however, I get 1731 cases with a valid value. This means that there are only 3 cases that have a value in both indresp and xwavedat.

Can you explain why this is the case? It seems as if in order to get full parental data I need to merge the values in indresp with the values in xwavedat.

I then tested another variable, to double check this: b_racel and racel_dv

. sum b_racel racel_dv

Variable |        Obs        Mean    Std. Dev.       Min        Max
-------------+---------------------------------------------------------
b_racel | 3,892 4.763361 11.60202 1 97
racel_dv | 52,065 3.03609 7.31321 1 97
. tab b_racel if racel_dv==.
no observations
  • This is what I expected.

I would be very grateful if you could clarify what is included in the xwavedat file in terms of parental information, or if there is a document where I could find more information.

Many thanks in advance!!!
And please let me know if you need further clarification; or if I am doing the wrong tests :)

Carolina

Also available in: Atom PDF