Project

General

Profile

Support #56

BHPS Variable problem

Added by IAN ALCOCK over 10 years ago. Updated about 10 years ago.

Status:
Closed
Priority:
Normal
Assignee:
Redmine Admin
Category:
Data analysis
Start date:
05/25/2012
% Done:

100%


Description

I have a problem understanding the data for the BHPS variable on asthma, wHLPRXE, which is available for waves K and P only. This variable is routed through the all waves variable on Health problems: Chest/breathing wHLPRBE.

At wave K the asthma variable KHLPRXE is routed: if KMEMORIG==1-4 & KHLPRBE==1; if from the BHPS/ECHP sample and R has chest/breathing problems. KHLPRXE has 1404 valid yes/no response (+ 3 cases missing). BUT, KHLPRBE has 1464 cases which meet the condition for being routed to KHLPRXE; i.e. there are 1464 cases KMEMORIG=1-4 & KHLPRBE=1. (A cross tab of KMEMORIG and KHLPRBE if KHLPRBE==1 shows that of the 2542 cases KHLPRBE=1, OS=1172, and ECHP categories=143, 126 and 23; total=1464.) So where are the other 60 (or perhaps 57) who do not have a response to KHLPRBE?

The numbers do not seem to add up logically at wave P either. At wave P the asthma variable is routed: if PHLPRBE==1; if R has chest/breathing problems. KHLPRXE has 1969 valid yes/no response (+ 2 cases missing / don’t know). BUT PHLPRBE has 2010 cases which meet the condition for being routed to PHLPRXE; i.e. there are 2010 cases PHLPRBE=1. So where are the other 41 (or perhaps 39) who do not have a response to PHLPRXE?

Is this a routing statment problem, or have I missed something? Thanks in advance

Also available in: Atom PDF