TO: Students in CJ 605
FROM: R. B. Taylor
DATE: 2/28/00
RE: Explained variance; remaining variance

Click here

The above spreadsheet runs through five problems, with the z scored drug reporting outcome, and one predictor - incivilities at Level 1. It shows you how to calculate

- between variance

- total explained variance

- within explained variance

You can see that for the explained variance:
- it does NOT matter if you grand mean center vs. do not center at all (but this does affect parameter estimates)
- it DOES matter if you choose to let the level 1 slope vary or not
- it does matter if you group mean center.

Please print this out and be sure you can talk yourself through it.

The specific models that were included appear at the bottom of the calculations.