TO: Students in CJ 605

FROM: R. B. Taylor

DATE: 3/12/04

RE: 3/15/04 assignment; paper comments; ASC stuff

ASSIGNMENTS

Although it does not appear on the sequence of topics page, there IS an assignment for 3/15 on the assignments page. Basically, you want to take your L1 model with demographics, and allow the slope of one of them to vary, and see what happens, and write it up. See the assignment page for more details.

The purpose of this exercise is to give you experience with the RCR (random coefficients regression) submodel. This is an important submodel, and you need to have experience with it.

You want to do some thinking before you pick which variable you will allow to vary. This thinking is CROSS-LEVEL: you are basically saying: here is why I think this one predictor will be more important in some neighborhoods than in others.

It does not matter if the slope really does or does not vary a lot across neighborhoods; you are just trying to think about why it might, and then see what happens.

COMMENTS

By now you all should have received comments on your initial L1 paper outline. Let me know if you did NOT receive the email.

FUTURE PAPER SUBMISSIONS

Given the extremely preliminary stage of many of your models, I have opted, save for the students who also are working with Dr. Ratcliffe, to not try and put together ASC submissions.

I can guarantee you, however, that after the end of the semester, I will be willing to work with you so you can continue to improve your final papers, and submit them for ACJS 2005, which will be in Chicago. The deadline will be in mid-fall; think early October.

If you are interested in submitting to Eastern Sociological Society for 2005 (March 17-20; DC; November 1 Deadline) I will be happy to work with you on that.