05-01-2017 01:33 AM
I have the following results and I'm not sure about the conclusions.
I am checking the impact of CEO in three stages. Stage 1,2 and 3. The results are as follows:
Stage 1 - 0.073 (p=0.05)
Stage 2 - 0.02 (p=0.01)
Stage 3 - 0.058 (p<0.0001)
All are positive impact, but I am trying to understand the difference between stage 1 and 3. Stage 1 is 0.073 but p=0.05 and stage 3 is with less impact but with higher significent. Can I say that the CEO impact is higher in stage 1 even if it is in p=0.05?
Thanks a lot!
05-01-2017 01:43 AM
Yes. The impact inside your model of Stage 1 is larger than Stage 3 (given alpha = 0.05). I am guessing that the significance in this case reffers to t-testing your coefficients=0, in which case for all three stages, you reject the null hypothesis of the three stages having no impact on your response variable.