Wednesday, June 27, 2007

Statistics - YUM!

Strangely, I love this part of my job...

Colleague’s Question:
Can you add to the green and red zones for the mean, SD, and SD/mean the criteria for classifying as red or green?

My Response:
I’m not sure how to do that concisely, while being more descriptive than I’ve already been.

Within each survey (students, parents, staff), I took the mean of the means for all questions. I took the respective standard deviation for the mean of all the means. Then I highlighted all individual question means that were at least one standard deviation above or below the mean of the means – green for above and red for below.

I did a similar thing for the standard deviations and standard deviations divided by means. I took the mean of all the standard deviations, and then took the standard deviation of that mean – highlighting standard deviations that were at least a standard deviation above or below the mean of all the standard deviations...

It all sounds like double-speak, but it makes sense. The goal was to highlight any descriptive statistic that was at least a standard deviation above or below the same kind of statistic within its respective survey.

3 comments:

Jeanine said...

Hmmmmm.... at first I thought I was clever because I understood what you were saying in this post - however, having now read the statistical report - (note that I said read and not comprehended) I bow mercifully at your feet, oh Sultan of Statistics, and beg for a little enlightenment.

Anonymous said...

For some reason, in this context, my original question sounds more like a quote from "Airplane".

drh said...

Joe, surely you can't be serious.

(Hmmm...doesn't work quite as well in written form.)