How would I combine these results from multiple studies to do a pooled comparison.

Question

I am working on a meta-analysis. The first part of this analysis involved looking at the prevalence of a certain condition in specific populations.

For this, I used MetaXL's pooled prevalence function (random effects) to do all the calculations.

However, now I need to do something different for the second half.

Each of the studies measured several variables (let's call them A, B, C)

They measured them in 3 different population types (i, ii, iii)

So right now, my table looks like this:

             A            B           C
          i ii iii  |  i ii iii  |  i ii iii  
study 1:  .  .  .      .  .  .      .  .  .

study 2:  .  .  .      .  .  .      .  .  .

study 3:  .  .  .      .  .  .      .  .  .

...

total:    -  -  -      -  -  -      -  -  -

Each of the studies reported a mean and standard error value within most of the categories (some studies did not have data for specific categories).

What I want to do is pool the results of all the studies in the total row and then use a t-test to identify which categories had significant differences. The problem is, I'm not sure how to correctly do this pooling. For prevalence, I used MetaXL's built in example demo but I'm not even sure what the name of the statistic I want is called.