Weighted mean for propoetions in R

1

Hi all,

I am wondering how I can compute the weighted mean for factored variables. All the packages I searched, consider a numeric variable not a factored variable.

I would appreciate it.

Farideh

2

For general R questions go to stackoverflow.com and use the r tag when posting.

3

Thanks so much. I will.

4

Factor variables are categorical variables. You can formally calculate the mean of such variables if you assign numerical values to each factor level, but the result will be meaningless.

5

Thanks for your reply.
My factor variable is smoking ( Yes=1 , No=0) and I would like to calculate the adjusted proportion of smoking versus not using weights group by City and using another medicine ( Yes=1, No=0).
All the functions and packages ( ddply, weighted.mean, survey package) I tried consider a numeric variable, so I couldn’t get the right answer.

6

0-1 variables are the only exception. you can recode the Yes/No values to 1/0 and then the mean value will be the proportion of Yes answers.

7

Please continue this on <stats.stackexchange.com>