Dennis Roberts writes:
>>Which type of sample survey would provide the most reliable data on the
>>frequency of cinema-going of the population?
>
>>3) Interviewing outside the cinema
>
>i still like this but, where are these cinemas? how many? what particular
>kinds of movies?
And when you find out that almost 100% of your sample attends movies, what do you do then? Unless you are willing to exclude non movie goers from your population, this is the worst possible sample. It's like estimating how many sick people there are using a sample of patients in a hospital.
Even if you excluded non movie goers, it's still a bad sample since people are represented in the sample proportionate to the amount of movies they attend. Someone who goes to a movie once a week is 52 times more likely to be included in the sample than someone who attends once a year.
Steve Simon, [EMAIL PROTECTED], Standard Disclaimer.
The STATS web page has moved to
http://www.childrens-mercy.org/stats
