ÖйúAƬ

Reliable information is unlikely to be gleaned from student responses to the sample survey questions you print. A question about libraries asks for strength of agreement with a positive statement; the adjacent question gives a negative statement about computers. Respondents are likely be confused and think that if 5 means good in the library context then 5 also means good in the computer context.

Another question asks for strength of agreement with the statement that "it was often hard to discover what was expected of me".

Disagreement could indicate that it was "never" hard, but it could also indicate that it was "always" hard. How will the pollsters interpret the answers to this question?

R. A. Bailey
School of Mathematical Sciences
Queen Mary, University of London