[EB] "Extra" variables in Euro-Barometer 44.0 (ICPSR 6721)

[James Cassell]

A client turned up something odd in Euro-Barometer 44.0 (ICPSR 6721).

Question 8a in the survey asks for reasons why the respondent favors a 
common currency (or something like that--I don't have the codebook 
handy). This is a multi-response question--interviewers apparently were 
instructed to mark all of the responses given by the respondent. For 
question 8a, there are 26 responses listed in the codebook. As usual for 
this this type of question, a variable is created for each response, 
coded 1 if the response was mentioned, 0 if it was not.

Trouble is, the codebook lists 26 responses, but the spss cards indicate 
there are 29 variables associated with Q8a. The variable labels for these 
are useless, as they are ALL THE SAME for this group of variables.

Question 8b has the same problem, with 23 responses listed in the 
codebook and 30 variables in the file.

Does anyone have a clue what's going on here? I suppose we could *assume* 
that the first 26 variables for 8a correspond to the 26 responses listed 
in the codebook...but that seem a bit imprecise. <grin>

Thanks in advance,
James
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
James Cassell                                cassell at irss.unc.edu              
Institute for Research in Social Science     http://www.irss.unc.edu/cassell/ 
University of North Carolina at Chapel Hill  Phone: 919/962-0782
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~