Adjusting For Proportion

[EuanC]

Hey all, Anybody up for some stats ??

I am trying to adjust for proportions and will use the following example :

In Area A, 10% of tests are done on those aged 40-50 whereas in Area B 40% of tests are done on those aged 40-50. However 20% of Area A's population is aged 40-50 compared to 50% of Area B's.

If we imagine that the mean 40-50yr old population of all areas in the region [Areas A-Z] as a whole is 40% then could I use that to make an adjustment factor.
Area A 40/20=2, so 2*10=20
Giving adjusted proportion of tests in 40-50 year olds in Area A as 20

Area B 40/50= 0.8, so 0.8*40= 32
Giving adjusted proportion of tests in 40-50 year olds in Area B as 32

Does this sound fair enough or is there a better way of making this adjustment?

[square food]

Hi.

So you are giving tests to different age-groups? Use the same persentages as the overall population for breaking up your sample.

eg. 40% of Area A's population is aged 40-50. You survey 100 people, 40 of them should be in that age bracket. That way your data will accurately represent the population.

Hope that makes sence.