Grouping of variables to conduct t-test

Hi all,

I have asked respondents in a survey to rate 10 different communication channels (e.g. email, phone, face-to-face...) from 'Not important' (1) to 'Very important' (5). Can I now split these 10 channels into two categories (e.g. offline, online), in order to see whether there is a significant difference between these two groups, by using a t-test.

I'm afraid that 1) it's not statistically acceptable to group variables like that, and 2) the t-test will not yield meaningful results. Regarding 2), I have already done the analysis and the difference of the means (of the grouped variables) is significant, but this might be just because I've grouped different variables together.

Also, I'm not 100% sure that the t-test is the right way to go.

I would love to hear your input! I've tried looking online already but to no avail...