just restarted data analysis so I've lost my edge quite a bit. Anyway, I've come across a problem and Im a bit stumped.
I have a dataset for adverts. I want to measure the uplift in conversion between two groups within the data- does advert contain X vs an absence of X.
Initially I split the data based on presence/absence of X and did a z test. The results showed a significance between the two datasets.
I was happy with this, but then remembered that there could be other factors at play between the two datasets that are driving the differences in the conversion mean.
My question is, how do I account for these potential factors in the data?