Need help with interpretation of regression coefficients

[EVstats]

Hi everyone,

I am currently finishing my linear regression and moderation analyses and I was wondering how to interpret the results. The coefficients are as follows:

LTO β0.141, p<0.01
IND β0.424, p<0.01,
Whereby LTO and IND reflect numerical scores ranging from 0-100, and whereby the dependent variable (tax transparency) is a percentage ranging from 0%-100%.
The way I interpreted it currently is:

"On average, an increase in the LTO score of 1 leads to an increase in the tax transparency score of 0.141%, ceteris paribus."

Is this correct?

Furthermore, I also have some significant moderating effects. These coefficients are as follows:
LTO x ANALYST β0.830, p<0.1
IND x ANALYST β0.900, p<0.05

How do I interpret these interaction effects when my dependent variable is a percentage (0-100)?

Hopefully anyone has some experience with this.

 

[Karabiner]

Is this correct?

No. Or did you mean regression weights (b's), not standardized betas? In that case it's correct.

With kind regards

Karabiner

 

[Buckeye]

If the dependent variable ranges between 0 and 100 wouldn't we use a probability distribution that accounts for the boundedness? Linear regression might produce estimates that fall outside of that range?