What about the multiple and adjusted R-squared? First of all, you should know that the multiple R-squared is simply the R-squared computed in the presence of more than one explanatory variable. That said, the adjusted R-squared is a kind of rectified version of the multiple R-squared, defined to correct this number from the effect of increasing the number of predictors.

It turns out that a very simple way to increase the R-squared is to increase the number of predictors employed: The R-squared will never decrease, and most of the time it will increase.

The adjusted R-squared is therefore defined to consider this effect and remove it from the final value. This means that through the adjusted R-squared you can compare models that have a different number of predictors.