Regression

I removed 53 outliers. R-squared rose from 0.818 to 0.856 and the model got worse. Every number that improved ignored the rows I deleted.

Your slide says $368. A colleague who ran the same data says $422. Four ways a regression coefficient means less than it appears to.

One broken assumption ruins your coefficients; the other three only ruin your p-values. Knowing which decides whether you scrap the model.

Everyone nods when it comes up in a meeting. What a coefficient actually is, how to read the output, and why the obvious answer was wrong.