On Thu, 19 Jul 2001 13:45:35 GMT, Jerry Dallal
<[EMAIL PROTECTED]> wrote:
>Some graduate programs in the (social) sciences require their
>students to take a minor in statistics or biostatistics. One
>purpose of such a minor is to give students the ability to analyze
>some of their own data, but at what level of understanding? Are
>such students expected to be able to synthesize the material they've
>learned, that is, rearrange it and apply it in new and possibly
>novel ways, or it is it more that they have a solid knowledge of
>standard techniques used in standard ways? (As I type this the
>answer seems self-evident!)
>
>The reason for these musings is that I'm being asked for advice on
>putting together a qualifying examination section for this type of
>minor and am finding the task daunting. I can imagine doing it by
>giving students 'typical' datasets to analyze, but that would make
>my job too easy. The committee would like it to be done orally! My
>instincts say that students exposed to such courses will invariably
>come up short when asked, "Why?" That is, their knowledge of
>underlying theory is weak by design and is perhaps unavoidable given
>the goals and time constraints of a 3 course minor.
>
>My instinct is to say the task is impossible. That is, some skills
>can be tested by having students apply them to data, but the nature
>of the 3 course minor precludes being able to judge a student's
>ability orally. I would be grateful for others' opinions on the
>subject.
>
>Thanks!
An oral qualifying examination would certainly test the students'
abilities to perform under stress. How about giving them the data
sets beforehand and enable them to prepare for any and all queries
from the student's doctoral committee members and/or other designated
questioners? Rather than having the questions come out of the blue,
they would be able to assemble notes, ideas, and a construct for
solving a research problem using the methods learned in the classes.
If the 3 course requirement is indeed a valid "minor," then why not
accept the results of the student's successful completion of same?
One must have some faith in the integrity of the faculty on this one
in so far as any grading system (or oral exam for that matter) is
concerned. Is the qualifying exam, in this instance, to gauge the
adequacy of the 3 course requirement? I'm curious what the 3 courses
are, e.g., 2 statistics courses and a research methodology class? 3
straight stat classes? Are these courses the same for all social
science degree candidates? I could come up with questions for those
with a statistics "minor" (based on 3 courses). Would these questions
measure what it is they are supposed to measure? Probably not.
Good luck.
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
