Perhaps the explicit guidelines for ratings could be in the form of a
rubric. Has anyone developed a rubric specifically for TBL peer
evaluations?
Regards,
Gary
Gary A. Hunt
Department of Natural Resource Sciences
Thompson Rivers University
900 McGill Road
Kamloops, BC V2C 5N3
Phone: 250-828-5461
Fax: 250-828-5450
On Feb 23, 2006, at 9:23 AM, Smith, David W wrote:
>
> I rarely use peer evaluations any more, so take what I offer with a
> grain of salt.
>
> My scoring essentially did what Jim Sibley suggests.
>
> I put limits on the number of peers who could be rated in the top
> category. After all, if everyone is contributing reasonably well,
> there isn't much room for several people to be outstanding. In a good
> group you expect similar contributions and similar ratings over the
> long haul.
>
> Some of my groups figured out they could rotate the top category among
> all the members over the semester. A group that can figure out how to
> work in harmony on their peer evaluations and stick to their agreement
> all semester has done something well. If their reports or assignments
> are working out well, then they are achieving the substantive goals of
> the group.
>
> Try giving some explicit guidelines for ratings, eg, "did not
> participate in the discussion." If you do, then everyone in the group
> will actually speak up, or their group will ask them to. Otherwise,
> they can't legitimately hold to their agreement to fix the ratings.
> This will legitimately give everyone in the group a similar rating.
>
> Watch carefully for someone who is doing noticeably worse than their
> group on exams and other evaluations. They might be letting the rest
> of the group do all the work. You can suggest privately that they are
> passing up an opportunity to learn by not contributing in a meaningful
> way. Correspondingly, someone who has much better individual RAT
> scores than the group score is either not contributing or is not being
> listened to by the group and the group may need to be told.
>
> Finally, there shouldn't be much point in worrying about it a great
> deal. Let them have the points and move on. Use this information as a
> diagnostic, of group effectiveness. If a group can conspire to game
> your system, then they are working well on something.
>
> Regards,
>
> David Smith
>
> David W. Smith, Ph.D., M.P.H., C.Stat.
> Associate Professor, Biostatistics
> Fellow, Institute for Health Policy
> The University of Texas School of Public Health
> San Antonio Branch Campus
> voice: (210) 562-5512
> e-mail: [log in to unmask]
|