Sandy:
My students often complain and resist. When I respectfully insist (and I am
indeed doing so sincerely, and for the benefit of learning), they end up
being fine with it. Of course, my Dept Chair and Dean both understand what's
going on with TBL and support it (perhaps this is the underlying problem?).
If you don't have support, or perhaps even had opposition, then that's an
interesting one. I think you need to dance that one with some subtlety...
and make buy-in from Chairs and such a real priority.
The only beef I've had with peer evaluations (a case where perhaps most of
us would have bent the rules if it had affected a letter grade) is when the
free-riders in a team colluded to give bad evaluations (and nasty comments)
to the one member of their team who I saw doing the work (easy to see
looking at I-RATs, as well). My guess is that students realize that you
may/will protect them from this kind of stuff and trust you on it. If not,
in an ugly class (which I don't think you'd have at a med school) some might
rightly get skittish.
Best,
Fritz Laux
Economics
Northeastern State University
From: Team Learning Discussion List [mailto:[log in to unmask]] On
Behalf Of Sandy Cook
Sent: Monday, October 15, 2007 6:36 PM
To: [log in to unmask]
Subject: Re: Peer Evaluations
Thank you all for your comments. I find it very interesting that those who
use the percentage/weightage method have not mentioned much push back from
the class about "lowering" some people's scores at the expense of others.
Just to let everyone know, we outlined the peer evaluation process in the
very beginning. Gave them an example to work out (sort of a TBL) on grade
setting to see the impact of both the varying percentages of IRAT,GRAT &
application as well as the peer. We also did a "trial" run of the peer for
feedback purposes. That is where we are now - the comments have been:
* I didn't realize that giving more points to one meant others get
lower
* I didn't realize the percentage impact. For example one group's
total resulted in on individual getting 108% and another 92% - that meant
that from the student's perspective, one student's grades jumped very high
by 8%, but another dropped by 8%. They really wanted to acknowledge one
person - but they did not want to penalize that much.
* Why must we penalize one in order to praise another?
* Can't we just give points - not take away points?
What I cannot seem to convey to them is that their group scores are not
theirs until it is moderated by the team contribution. They feel they have
earned all the group points - and the moderation is "subtracting" from what
is rightfully theirs.
On another note, in our first course experience there actually have been a
couple of students who's final grade was in fact brought down, ever so
slightly, by the group process. This is mainly a function of some group
processing issues they still have to work out. Those individuals actually
did better on the IRATs than the group did on the GRAT, explained by the
group that it usually was because that one person "guessed" right and was
not confident enough to convince the rest of the group that their answer was
right - thus went with majority rules. Plus, as we are new to developing
our applications, some application scores are lower than the individual's
general averages on tests and IRATs - but the difference is usually
miniscule - less than 1 percentage point - but it has had an impact.
Telling them that life is full of zero-sum experiences and that this is an
important lesson to learn and to work with - is not flying well.
Again, thanks for your input.
**************************************************
Sandy Cook, PhD
Associate Dean for Curriculum Development
Duke/NUS Graduate School of Medicine
From: Team Learning Discussion List on behalf of Molly Espey
Sent: Tue 10/16/2007 5:55 AM
To: [log in to unmask]
Subject: Re: Peer Evaluations
I use the peer evaluations to weight the group portion of the grade.
Then, I don't have to set some base for what "average" means - average
means the group's score on activites and RATs throughout the semester.
This then rewards groups that work better together and produce better work
more than groups that don't do as good work.
Some groups do quite well with nearly equal contributions from all
members. Then they are all equally rewarded by doing well as a group, as
I find that groups that really work well together end up with everyone
averaging very close to 10.
I also do the peer evaluations at least once in the middle of the
semester, as well as at the end, but the midsemester evaluation doesn't
count toward the final grade. It does, however, allow me to give feedback
to students. I remind them that mathematically there are two ways to
improve the group portion of their grade (which again is weighted for
individuals by the peer evaluations, with some subjective consideration on
my part): increase the group's scores or increase their own peer
evaluations scores. Usually, efforts to improve one's own peer evaluation
score will also increase the group's grades.
I've only had one student complain to me directly, upset that the group
activities were bringing her grade down. She was wrong, however,
mathematically.
Molly Espey
Applied Economics and Statistics
Clemson University
> There was a great thread about Peer evaluation in January, which was
> informative, but truthfully, I did not appreciate the discussion at the
> time.
>
>
>
> We have just completed our first peer evaluation process and I have some
> questions. We believe in the peer evaluation process and will not
> abandon it, but there have been some issues.
>
>
>
> In the TBL book there are two forms of peer evaluation described
> (percentage and maintenance). Several pros and cons are listed, but
> mostly ending with a suggestion of a positive learning note. Of the two
> methods described, selfishly I chose the percentage one because it made
> more sense to me and was easier to calculate. The students however, are
> incensed (well maybe too strong of a word, but upset) that it is a
> zero-sum game. They don't mind giving points to those who contribute,
> but they do not want to take points away from those who contribute less.
>
>
> * How do you rationalize the zero-sum concept?
>
> * How does one explain the value of moderating the scores?
> Maybe it is a cultural thing - being nice, but the idea of taking away
> something they believe they have earned is painful. How do you tell them
> that they have not really earned the group scores unless they
> participate in the group?
>
> * When the group size results in a proportion that is not easily
> divisible by 5 - and they want to give the team equal marks - but can't.
> For example a team of 7, with 6 ratings can only give 16.7 and 16.6 -
> someone will be a bit higher and a bit lower.
>
>
>
> Using the maintenance method might solve the logical problem by making
> the peer assessment an added component to the grade - not subtractive
> (on the surface). If I were to switch to that method,
>
> * How do you decide what % of the final grade should the peer
> assessment be?
>
> * Is it really any difference - or does it just appear that way
> to the students because they see it as adding not subtracting?
>
> * How do faculty feel about inflating grades by making portion
> of success be solely on peer points?
>
> * Will I exchange a student fight for a faculty one?
>
>
>
> This is quite a contentious topic, and I can see why people give up on
> it - or move away to more feedback rather than grade moderation - but we
> really feel that it is important to keep - so any advice on how to deal
> with student's anxiety is most welcomed.
>
>
>
> Sandy
>
>
>
>
>
>
>
> ***************************************
>
>
>
> Sandy COOK, PhD | Associate Dean, Curriculum Development | Duke-NUS
> Graduate Medical School Singapore | W: (65) 6516 8722| F: (65) 6227 2698
> |
>
>
>
>
>
>
|
