The questionnaire had two sections. One section
contained five questions (question 1 – 5) concerning
the respondent ’ s experiences of exercise habits with the answering alternatives “ yes ” , “ no ”and in two
items also the alternative “ don ’ t know ” . This part
had a median unweighted kappa value of 0.60. There
were also fi ve questions where the respondents
answered in free text. The second part of the questionnaire
(questions 7a – 7j) contained 10 statements
about different environmental factors, where the
respondent rated the degree of importance on a
graded semantic differential scale ranging from
1 “ not at all important ”to 7 “ very important ” ,
with an extra alternative for answering “ don ’ t know ”
or “ not applicable ” . This part of the questionnaire
had a median weighted kappa value of 0.65. The
questions in the questionnaire and the kappa coef-
fi cient for each question are showed in Table I.
The questionnaires were marked with a code and
sent out by mail. The time for returning the questionnaire
was 3 weeks. A reminder including the questionnaire
was sent to the non-respondents with
another 3-week answering period.