Project

General

Profile

Support #480

Analysis by highest educational or vocational qualification

Added by Veronique Siegler about 8 years ago. Updated about 8 years ago.

Status:
Closed
Priority:
Urgent
Category:
Data analysis
Start date:
01/07/2016
% Done:

100%


Description

We are planning to analyse some of the Understanding Society variables by highest educational or vocational qualification in waves 1 and 3.
• My understanding is that I should be using: c_hiqual_dv for wave 3 and a_hiqual_dv, and this should give me the following categories: degree; other higher degree; A-level; GSCE level; other qualification; no qualification. Or shall I use c_qfhigh_dv?
• I would like to know what qualifications are included in each category, please? I presume that c_hiqual_dv is derived from c_qfhigh_dv?
• Also, why is there missing variables in c_hiqual_dv and why is there even more missing variables (and not the same number of missing variables) in c_qfhigh_dv?
• I also understand that not all respondents were ever asked the highest educational qualification question, and I would like to know how I should be dealing with this in both waves? Still unclear to me after reading documentation on website.

#1

Updated by Victoria Nolan about 8 years ago

  • Status changed from New to In Progress
  • Assignee changed from Alita Nandi to Veronique Siegler
  • % Done changed from 0 to 50

Dear Veronique,

Many thanks for your enquiry.

qfhigh_dv includes vocational qualifications which could not be fed forward from the BHPS sample because the code frame for collecting vocational qualifications was different. If you use both BHPS and Understanding Society you should stick to hiqual_dv as this is less detailed and the code frames on the BHPS and Understanding Society were similar enough.

With regards to how to deal with ‘missings’, just like in any analysis you can go ahead using the valid responses to the variable. With regard to dealing with the fact that some respondents were never asked the initial _qfhigh question, our suggestion would be to either drop those respondents (the flag _qfhighfl_dv tags these) from the analysis in W3 or to include a robustness check for this.

hiqual_dv is the better choice and this uses the last known w_hiqual_dv updated by the current wave w_nhiqual_dv as appropriate.

Hope this helps,

Best wishes, Victoria

On behalf of the Understanding Society User Support Team.

#2

Updated by Victoria Nolan about 8 years ago

  • Status changed from In Progress to Closed
  • % Done changed from 50 to 100

Also available in: Atom PDF