Comparing groups with different sample sizes

Question

I have some data about network outages in two companies, by the type of machines they have:

Machine type         A     B    C    D
Company 1 outages    5%   15%   3%   7%
Company 2 outages    3%   11%   8%   9%

There are different numbers in each of the groups and I read that I can't use a chi-sq test to see if there is a difference in outages between the two companies.

Is there a test I can use to say whether there is a difference in the outages by company/machine?

Answer 1

If I am understanding the problem correctly, then you could do a logistic regression with outage as the dependent variable and company and machine type (and maybe the interaction) as independent variables.