Comment 76092

By Borrelli (registered) | Posted April 20, 2012 at 15:20:38 in reply to Comment 76089

With so few data points, the leverage exerted from that single outlier is pretty huge. If you excluded it, I agree, the regression line would be almost flat.

Even without the raw data, you can see that the remaining data points (excluding the outlier) hover between 1 and 2.8ish. I calculated 2.1 as a rough estimate of the mean based on estimates of the points on the graph.

Again, using rough estimates of where those points lie, I calculated a standard deviation of 0.3, which means the outlier with a score of 4 would be more than 6 s.d.'s from the mean (a pretty good reason to exclude).

Permalink | Context

Events Calendar

There are no upcoming events right now.
Why not post one?

Recent Articles

Article Archives

Blog Archives

Site Tools

Feeds