Negative Binomial Regression and Type I Errors

#1
Hi,

I've completed three test of negative binomial regression. One test had 14 independent variables, and the other two had 7 independent variables each. So in total I've looked at 28 IVs across the three tests.

In terms of risk of Type I errors, is this not really an issue as I've only run three tests, or is it potentially an issue as I've looked at 28 independent variables?

Any advice would be great!

Thanks
 

hlsmith

Omega Contributor
#2
A little confused, by test you mean models examining coefficients' significance?


Models all had the same DV but a mix of difference IVs with some used in multiple models?
 
#3
A little confused, by test you mean models examining coefficients' significance?


Models all had the same DV but a mix of difference IVs with some used in multiple models?
Hi, sorry stats is not my strong point. Yes I meant that I had 3 models but each IV only featured in one of the models. All had the same DV. So I had an assessment scale with a total of 28 items split across three subscales. Am I making any sense?

Thanks