Project

General

Profile

Support #124

Questionnaire routing of wJBSECT

Added by Karen Mackinnon over 7 years ago. Updated over 7 years ago.

Status:
Closed
Priority:
High
Assignee:
Redmine Admin
Category:
Data documentation
Target version:
Start date:
02/14/2013
Due date:
% Done:

100%

Estimated time:

Description

In the BHPS documentation [Volume A] the following is stated …

“III.9.2.2. The Employment Section
Users should be particularly aware of the structure of the employment section, and note that the
routing on the questionnaire from Wave Two onwards means that respondents interviewed at
the previous wave do not answer all questions; whether they do depends on whether they have
changed their job since September 1 of the previous year. Differences in questions asked of
respondents interviewed in earlier waves and those asked of new entrants should be
particularly noted; although the variable names are the same as at Wave One, the population
totals may differ.
This applies particularly to the following variables:
wJBSECT wJBTIME wJBONUS wJRISE wTUJBPL wTUIN1
wTUIN2 wJBOPPS wJBPEN wJBPENM wPAYLY wPAYLYG
wPAYLY wPAYLYW
As questions on union membership are not asked of employees in the same job as at Wave
One, for example, users will need to carry over the Wave One value for employees at later
waves who have not changed their job in the previous year, in order to use the variable on
union membership for all respondents.”

The code book [Volume B] details for EJBSECT to RJBSECT also states that “From Wave 2 to 4 this was not asked of employees still in the same job as last year”.

However I have been checking out the variable wJBSECT and it looks as though this variable was asked of those who had changed jobs in Wave 2 ONLY. In the Wave 3 to 15 Employment Section wJBSECT is returned to the E9 question position and is NOT routed by those who were interviewed last year and changed their jobs.
And in Wave 16-18 [P,Q,R] the question is not asked of those who have remained in the same sector but their data is fed-forward if available from a previous interview so no carry forward is required by researchers. Again please confirm that this is the case.
The frequencies of wJBSECT across the wave also suggest that this is the case - the % of ‘Inapplicable [-8]’ cases range from 47-48% for all waves except W2 where it is 81.5%.

This has been time-consuming to investigate so please can you confirm I am correct and that the only carry forwarded of data required for wJBSECT is in the W2 data.

History

#1 Updated by Redmine Admin over 7 years ago

  • Status changed from New to In Progress
  • Assignee changed from Karen Mackinnon to Redmine Admin
  • % Done changed from 0 to 50

Karen,
Just to confirm that we have come to the same conclusions as you have.
Jakob

#2 Updated by Redmine Admin over 7 years ago

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

Also available in: Atom PDF