# Standardized Coefficients - Unit Problem?

##### New Member
Dear all,

I am new to this forum and already wanted to thank all for reading and/or helping.
I could unfortunately not find the answer to this question anywhere, or maybe I just misread...

I have a little problem after running a multiple linear regression on XLSTAT.

I have a dependent variable ("GDP") and different independent ones ("policy 1", "policy 2", ...). All units are the same - $or any other currency. I have observations across some 25 countries. Now the results came out as: Dependent variable: GDP // Std Dev 178 Independent variable 1: policy 1 // Std Coefficient 0.984 // Std Dev 10.47 Independent variable 2: policy 2 // Std Coefficient -0.144 // Std Dev 25.42 My question is now regarding interpretation: - Can I say that for 1 standard deviation change in "policy 1" (lets say an increase of 10.47$), "GDP" will increase by (0.984 * 10.47) ? Same logic for increase/decrease of "policy 2", (std coefficient policy 2 * std deviation policy 2) = change in "GDP" ?

Or is it for 1 standard deviation change in "policy 1" and then multiplied by the (std coefficient of policy 1) * (std deviation of THE DEPENDENT VARIABLE)?

I am a bit confused... All I read in textbooks and articles is that 1 SD change in the ID variable means x SD change in the dependent one - but nowhere it says if in both cases, the SD is the SD of the respective ID variable, or the SD of the dependent one...

Hope I am making sense Thanks and warm regards
Pa