Step 7: Weighting
Observations made at the site to help develop a sampling frame and define the
sampling unit will have established times of the day when the site has most of its
visitors. Unfortunately, it may not be possible to increase the number of interviewers
to account for this. Thus, by using the next-to-pass technique there may be certain
times when many more people are passing the interview point and leaving the site
because the interviewer is occupied. For example, it could be found that half of those
interviewed on a particular day had travelled to the site by coach, whilst observations
had shown that a higher proportion of visitors had arrived by car. Some account
should be made of this to allow for possible errors in the results. Thus, some replies
in the analysis stage could be adjusted or ‘weighted’ to allow for the higher numbers
of coach trippers interviewed than were actually present on the site.
Do'stlaringiz bilan baham: |