Thanks for your reply. I hadn't thought about it from that perspective. The
structure is that these are medical students doing rotations, and evaluating
their professor for each rotation. The rotations last 6 weeks.
>From: Donald Burrill <[EMAIL PROTECTED]>
>To: Karen Scheltema <[EMAIL PROTECTED]>
>CC: [EMAIL PROTECTED]
>Subject: Re: hierarchical linear modeling?
>Date: Fri, 7 Jul 2000 12:49:01 -0400 (EDT)
>
>Hi, Karen! Interesting problem. You mention students (each of which has
>made a variable multiplicity of ratings on professors), and professors
>(each of which has received a variable multiplicity of ratings from
>students). You do not mention courses. Are all these ratings for a
>single course? (If so, how do you get up to 40 ratings from one
>student??) If not, I'd be inclined to divide the data by course:
>a professor's "teaching style/ability" might well be different, or
>perceived differently, in different courses.
> If students do not have multiple professors for a given course,
>this might simplify the structure of your data enough to make the
>problem tractable. Do there exist replicates (more than one rating of a
>given professor by a given student in a given course)?
> If students do have multiple professors in a course, can the
>course be separated into components each of which had only one professor?
>This again would simplify the analytical problem.
> Good luck! -- Don.
>
>On Thu, 6 Jul 2000, Karen Scheltema wrote:
>
> > I need some advice about a data set I've inherited. In the data,
> > students have rated professors on their teaching style/ability. The
> > problem is that students complete several evaluations of different
> > professors as they go through their rotations. A student may have
> > completed as many as 40 evaluations. In addition, each professor has
> > been evaluated by many students. The research question is looking at
> > various components of teaching ratings to predict overall satisfaction.
> > What I'm struggling with is how to account for the multiple ratings by
> > each student, as well as the fact that each professor has multiple
> > ratings. I was initially thinking of hierarchical linear modeling,
> > with student being a level of the hierarchical model. That leaves me
> > wondering how to handle the multiple ratings of each professor. Any
> > advice on how to analyze this data set would be greatly appreciated.
>
> > Karen Scheltema, M.A., M.S.
> > Statistician
> > HealthEast
> > 1700 University Ave W
> > St. Paul, MN 55104
> > (651) 232-5212 fax: (651) 641-0683
>
> ------------------------------------------------------------------------
> Donald F. Burrill [EMAIL PROTECTED]
> 348 Hyde Hall, Plymouth State College, [EMAIL PROTECTED]
> MSC #29, Plymouth, NH 03264 603-535-2597
> 184 Nashua Road, Bedford, NH 03110 603-471-7128
>
Karen Scheltema, M.A., M.S.
Statistician
HealthEast
Office of Research and Medical Education
1700 University Ave W
St. Paul, MN 55104
(651) 232-5212 fax: (651) 641-0683
________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com
===========================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
===========================================================================