Print

Print


We have always allowed developmental instructors in writing and reading
to recommend a retest or placement correction during the drop/add period
for students whose early class work shows that they are misplaced. We
are starting to allow instructors to recommend a retest before the end
of the semester to allow students to finish their developmental course
early.

Retesting is a sticky issue. We use COMPASS (plus a writing sample to
confirm whether students need developmental composition--we don't rely
just on COMPASS for that). When we researched the performance of repeat
testers, we found that a surprising proportion of students improved so
much on their retests, even a few days after their first test, that they
qualified for a higher placement level. In math, this was true for those
who placed in the lowest level (arithmetic) but rare for those whose
initial placements were for the higher developmental levels (algebra).
When I showed these findings to our COMPASS rep, he suggested that the
gains were within the normal variance for repeated trials, but that
didn't explain to me why the retest scores were almost always higher,
not lower. I think that test-wiseness and attitude have a lot more to do
with the changes. That's why I favor retesting: I think the function of
developmental courses should be to establish and strengthen skills in
math, reading, and composing, not just to teach skills in taking
Accuplacer or COMPASS (also, not just to teach students that bad things
can happen when they blow off tests).

It's highly unlikely that the students are seeing the same items on
repeated tests. The item banks are huge. Even if they did, they don't
get specific feedback about their answers on the first test, so how
would they know which questions they need to answer differently?


-----Original Message-----
From: Open Forum for Learning Assistance Professionals
[mailto:[log in to unmask]] On Behalf Of Alan Craig
Sent: Tuesday, May 20, 2008 12:05 PM
To: [log in to unmask]
Subject: Re: Retesting for higher placement - developmental

Deborah,

At Georgia Perimeter College, a two-year unit of the University System
of Georgia, students must earn at least a 'C' in the exit-level
developmental course in a discipline (math, writing, reading) in order
to be eligible to take the exit test. We use COMPASS. 

I wonder whether the frequent testing/retesting with Accuplacer is
exhausting the pool of questions so that students get sufficiently
familiar with them to pass the test but not the course?


Thanks!

Alan Craig

Coordinator-Dunwoody Learning & Tutoring Center    
             
Recording Secretary-NCLCA
           
Georgia Perimeter College     
2101 Womack Road    
Dunwoody, GA 30338    
770-274-5242    
[log in to unmask]     

Visit our award-winning website:
http://www.gpc.edu/~duniss
     
The union of the mathematician with the poet, fervor with measure,
passion with correctness, this surely is the ideal --William James

-----Original Message-----
From: Open Forum for Learning Assistance Professionals
[mailto:[log in to unmask]] On Behalf Of Deborah Pratt
Sent: Tuesday, May 20, 2008 11:57 AM
To: [log in to unmask]
Subject: Retesting for higher placement - developmental

Hi all,
  We use Accuplacer to determine placement into courses, developmental
or otherwise.  In developmental courses, faculty also test/retest using
Accuplacer to see how students' scores change.  Some students who have
performed poorly in class have taken the retest and scored high enough
to be placed into the next level course.  Do any of you have a policy on
whether you allow developmental students to do this?
  Regards,
  Deborah Pratt
  Ancilla College
  Donaldson, IN 46513

~~~~~~~~~~~~~~~
To access the LRNASST-L archives or User Guide, or to change your
subscription options (including subscribe/unsubscribe), point your web
browser to http://www.lists.ufl.edu/archives/lrnasst-l.html

To contact the LRNASST-L owner, email [log in to unmask]

~~~~~~~~~~~~~~~
To access the LRNASST-L archives or User Guide, or to change your
subscription options (including subscribe/unsubscribe), point your web browser to
http://www.lists.ufl.edu/archives/lrnasst-l.html

To contact the LRNASST-L owner, email [log in to unmask]