Regression coefficient B and constant

B says how much the predicted variable rises (or falls) on average when a predictor rises by 1 point - in the units of the data. The constant is the predicted value when all predictors are 0; it fixes the position of the line.

Example: Rehearsal strategies from time management and gender: B = 0.24 for time management. One point more in time management goes with 0.24 points more rehearsal on average, for the same gender. Constant 2.17.

Caution: B depends on the scale (1-5 or 10-50) and is not suited to comparing studies - use beta for that. The constant often has no meaning when 0 cannot occur (scale 1-5).