Small R-square, big slope, how is that possible?

Hey guys,

I'm doing a sequential linear regression. My dependent variabele has a range from 1 - 5. My independent variabele is a dummy. I found the following results:
Model 1: R-square = 0.0001 (contains only my dependent variable and two control variables)
Model 2: R-Square = 0.02 (contains the same variables as model 1 but I added 2 dummy's.) These two dummies are having ß = - 0.41 (P= 0.01) and ß -0.75 (p = 0.02).

So my question is:
The R-square is so low in model 2 (only 2% of the variance is explained by the independent variables), but these slopes of the dummies are quite big if you check the range of the dependent variabel. How is that possible and can you consider this as big?


Less is more. Stay pure. Stay poor.
Why can't these terms justexplain 2%? Can you post the outputs? Does the DV take on any value between 1-5?