By Ruben Geert van den Berg on October 2, 2013 under Other Recoding Tutorials.

SPSS Recode – Cautionary Note

Summary

When variables are recoded, it is important to adjust their value labels as well. Otherwise, misleading data may result.

SPSS Recode Resulting in Misleading Data

SPSS Recode Syntax Example

*1. Create test variable.

data list free/v1.
begin data
1 2 2 3 3 3 4 4 4 4
end data.

value labels v1 1'Very bad' 2'Rather bad' 3'Rather good' 4'Very good'.

*2. Frequencies. Note we have a happy sample (mostly 'Very good').

set tnumbers both.

frequencies v1.

*3. Reverse coding for v1. Alternative: compute v1 = 5 - v1.

recode v1 (1=4)(2=3)(3=2)(4=1).

*4. If value labels are left unaltered, we now have an angry sample (mostly 'Very bad').

frequencies v1.

Solutions

Comment on this Tutorial

*Required field. Your comment will show up after approval from a moderator.