Why doesn't Prism report R² for linear regression when I force the line through the origin (or some other point)?

When you constrain a line to go through a point, there would be two ways to compute R²:

• Compare the fit of the best-fit line with the fit of a horizontal line at the mean Y value. But that null hypothesis (horizontal line through the Ymean) doesn't obey the constraint that it go through the origin.
• Compare the best-fit line with a horizontal line at Y=0. This obeys the constraint, but often fits the data really badly, pushing up the R² value.

Since R² is ambiguous when you constrain linear regression, we chose to simply not report it with Prism. But you can get the value if you want:

1. Choose nonlinear (rather than linear) regression.
2. Choose a straight line model
3. Set the constraint  (intercept=0).

The nonlinear regression results include R², using the first definition above (the null hypothesis therefore does not follow the constraint).