Practice · Coach the Rookie

The n − 2 slip

A rookie analyst on your AFL team has submitted their working. Something's off. Spot the mistake, then fix it.

0 / 2 graded

person

Rookie's submission

Model: win % ? tackles · n = 7 teams

Fitted the line for win% on tackles. ✓ Residual sum of squares: SSE = 12.0 ✓ Residual variance: s² = SSE / n = 12.0 / 7 = 1.71 "So the residual variance is about 1.71."

1 · Where did the rookie go wrong?

2 · Fix it — what's the correct s²? (SSE = 12.0, n = 7)

s²

s² = SSE / (n − 2) = 12.0 / 5 = 2.40 (s = √2.40 = 1.55)

verified

Graded — nice coaching.

The n − 2 divisor is the single most common slip in SLR. Catching it in someone else's work is how you stop making it in your own.

↩ All case files