How to create a new variable that contains the total frequencies of only specific values of another variable?

[William Wong]

Hello community,

I'm trying to create a new variable "ER" (employment rate) from another variable LFACT (labor force activities) which has indicator values from 1 to 14. Values of 1 and 2 indicate "employed" and 3-14 indicate "unemployment" for various reasons.

The way I calculate the ER is: total frequencies of 1 and 2 appearing in LFACT divided by the total frequencies of 1 to 14 appearing in LFACT.

So, how to create a new variable that would list the total frequencies of only values 1 and 2 in LFACT? as well as a second new variable that lists the total frequencies of all values from 1 to 14... ?

Many thanks,
William

[Maarten Buis]

It would be helpful if you gave us some example data to work with. Now we have to guess what your datastructure is. For example I guess you have not told us all we need, especially I think you have a third variable (maybe sector, maybe region, maybe year) over which you want to compute those rates, otherwise a simple tab LFACT would do (Just look at the cumulative percentage at category 2, which will give you the percentage employed, and hundred minus that percentage gives you the percentage unemployed). In the FAQ there is a description on how to give us example data (lookfor dataex).