Hey everybody!

I am pretty new to this forum and also to statistics. In these days I try to analyze the data for my Bachelor thesis. I study psychology in the north of Germany.

Well, my questions refer to the process and the output of my multiple, hierarchal regression.
I have two directed hypotheses concerning the increase of two personality constructs from T1 to T2.

I thought of a basic linear regression where the p-value is given two-tailed and you just divide it to its half.

1. The output shows for the last step of the conducted regression in the coefficient-table for the t-value of beta a p-value of .088. As my hypothesis is directed, am I allowed to divide the p-value?

2. I'd like to include some covariates to control age, gender and also the construct's level for T1 to control the prediction for the general model. Is it important to put in those control variables in a specific order? (Andy Field writes that it would be common to do so in order of their importance...I found it more intuitive to give them a content based order)
Currently I start with the demographic stuff, second block T1 and third block the (in my theory) predicting variable. Statistically, is there any difference?!


I would be very grateful for any help or tips! (and sorry for my bad English...)