Need help with paired t-test?

Hello forum,

It has been years since I took stats in college and I am having difficulty determining which test is most appropriate.

Here is the situation.

I have multiple locations that were sampled for population size before and after a treatment was applied. However, I have an uneven number of samples i.e. for site 1 I have 3 measurements before and 5 after, and for site 2 I have 5 before and 10 after, etc.

My gut is telling me this is a paired t-test but I am not sure how to deal with the unequal sampling.

Any help would be appreciated,

Thanks J.

p.s. I typically use sysstat and this is post-hoc, I know the design scheme isn't good.


Omega Contributor
So you have multiple observations per person that are not even between sites or you have uneven number of people between sites?