My dissertation is due in the next 4 weeks, and what with it being distance learning and having other exams that I needed to do this year, I have just started it, and have come across a big problem.

I am doing a meta-analysis on two modes of insulin delivery in children with type 1 diabetes, which leaves me with results of how well controlled their diabetes is - including a pre-intervention, and a post-intervention result (mean +- SD).

Now, to calculate the difference between the means, I just take them away from each other - thats easy, but how do i calculate the new SD?

The reason I ask this, is because I will need to do this for both groups (insulin pumps vs insulin injections) - and THEN ill be comparing the pumps vs the groups - and put this info into my meta-analysis. But I struggle to see how I am to do it.


Thanks !