Re: Grade inflation? Test/examinations

In a message dated 04/16/2002 5:10:07 PM Eastern Daylight Time,
[email protected] writes:

<< One explicit premise underlying my previous posts on grade inflation is
 that the students be fully informed, when a course begins, on what they are
 expected to learn as they take the course. >>

I agree.  The expectation, as such, can be presented as a list of
behavioral objectives (AKA, performance objectives).  The best method
for writing these was developed in the 1960s by Robert Mager.
Julie Vargas also has a highly useful book on writing "worthwhile"
behavioral objectives.

<< That requires that the instructors have responsibly analyzed and then set
 forth the subject matter of that course. >>

Yep.  And this activity should be carried out BEFORE writing the objectives.
This includes (1) conducting either a topic analysis or a task analysis,
or both as necessary, (2) specifying the target audience, and (3)
identifying the assumptions and limits for both the course and of
the learners comprising the target audience.  One may also want
to take the behavior of previous classes of learners into account
when making these determinations.

<< That is a matter of "curriculum."  >>

Yes.

<< The exchange of posts on this topic have been most instructive,
 demonstrating a variety of discordant views on _how_ to teach, _what_ to
 teach, how to examine (a single exam), and how to reach a grade for a
 course from a number of exams, term papers, lab exercises, or
 whatever.  One or two posts suggest that there's some information out
 there!  And all suggest that Mickey Keenan's post on those five monkeys is
 exactly on target. >>

<< So, first, what behaviors are we trying to enable students to
 acquire?  What are we seeking to add to their behavioral repertory?  What
 do we want to be sure they _know_ after we have 'taught them?'  _Know_, a
 bad "ontological," even "metaphysical," word--bait for philosophers--but a
 word we "know" how to use in conversing. >>

To a radical behaviorist a term such as "to know" may seem anathema.
To me, while it ultimately is anathema, I don't mind is as a loose
goal, provided that it is then broken down further into whatever
set of behaviors are said to exhibit knowing.

<< Given the great variety of issues and practices raised by respondents, it
 seems advisable to narrow this post to a single problem--that is, to the
 single "tests" or "examinations" given to determine how much verbal
 material that an instructor has taught can be reproduced by the student. >>

IMHO, there are issues above and beyond whether or not a learner's
demonstration of the quantity of learning produced should rest on
a single test or examination.  There may well be -- I contend there are --
better ways of indicating what has been learned, and to what degree
it has been learned.  But, for now, that's a side issue in the context
of your overall message.

<< Stated differently, how much has taking the course added to the student's
 repertory of "intraverbals."  Further posts may deal with 'oral' and
 'essay' samples of behavior that do not restrict either examiner or student
 to spoken or written, heard or read verbal behavior. >>

Intraverbals comprise a common element in much instruction. However,
there are also subject matters where you would want a tact repertoire
demonstrated.  For instance, in a biology class you would want people
to be able to look through a microscope and identify the nucleus of
an actual cell, not simply to read about it or see pictures or drawings
of it.  In teaching a botany class, being able to identify the common
or scientific names of trees, what classification they are in, etc., out
in the wilds might be of use, and that would imply teaching a tact
repertoire of some sort.  These would be see-say or perhaps
see-write learning channels, but in the tact category, not the intraverbal.

Other classes of verbal behavior, whether of those identified by Skinner
or other ones, might be shaped and/or strengthened as well.  Steve
Graf in his university courses, for instance, teaches pronunciation
of scientific terms as one of the low-level objectives.  Graf found
several years ago that college students anymore may not even be
able to correctly pronounce terms.  Give them the word "reification,"
and they may not be able to pronounce it correctly, or fluently.  Given
that that is a deficit exhibited by the learners, and not something
that would ordinarily be uncovered by a topic analysis, or even by
a task analysis, this also suggests that when specifying the
target audience that one observe their behavior.

<< Nor need we consider how to instruct anyone how to be a "genius," someone
 who develops something altogether new and different--what the instructor
 could _not_ teach (This rules out the "contributions to knowledge" made by
 the cookie-cutter procedures taught by the requirements for the Ph.D.).   >>

True.

  -- JE