Difference in hypothesis testing simple lineair regression or hierachical lineair regression

[Naomiii]

Dear Reader(s),

I am studying the relationship between burnout on GPA.

I wrote down my hypothesis as follows:
H1: Exhaustion will negatively affect GPA
H2: Distance will negatively affect GPA
H3: Competence will positively affect GPA

However, I have also two control variables and now I have conducted a Hierachical regression. However, I have the feeling I have to do 3 simple lineair regression, or is it no problem that I have conducted a multiple regrssion? , my supervisor wanted that, namely. I really don't understand the differnce between one or the other model. Could someone explain which how I have to state my hypothesis?

 

[obh]

Hi Naomiii

Generally multiple regression sound as the best way to show the connection between GPA and the dependent variables.
Only If you want to show how each variable influence GPA independent on other variables you will do a simple regression, and I can't think why would you want this.

When using multiple you need to be careful to insert only variable which makes sense and not to have too many variables to avoid overfitting (the number of observations should be at least 10-15 times the number of the variables)

In the following example, X2 is significant only if you add it to X1,

Y = -0.09843 + 1.0063 X1 + 1.0189 X2 p-val (x1,x2): (3.073e-9 , 0.000005077)

Y = 2.4458 + 0.9697 X1 pval(x1): 0.00004885
Y = 5.5294 + 0.5717 X2 pval(x2): 0.6799

x1
1
2
3
4
1
6
7
8
9

x2
1
2
3
3
3
3
3
2
1

Y
2
4.1
5.8
7.1
3.76
9.11
10.12
9.91
9.87