Hello
I would appreciate confirmation that it is legitimate to perform the chi-square test of association with mortality frequency data which have been calculated in the form of deaths per 100,1000 population per year for a particular disease.
The idea is to test for an association between level of deprivation (two categories) and year of study (6 categories) but in the first instance the mortality data have been standardized within the individual deprivation categories according to the size of the populations within these categories. Of course, as is typically the case for standardized frequencies, the values of the frequency data are not integers.
Strictly speaking, I intend to perform a chi-square test of linear trend. However, I suspect that the answer to my first question also applies here.
I look forward to receiving constructive replies.
Many thanks
Best wishes
Margaret
---------------------------------
For ideas on reducing your carbon footprint visit Yahoo! For Good this month.
|