I used to give a short "math test" at the beginning of stats. It was
very simple and designed to reduce math anxiety. It had problems in
simple mathematical operations; I think that the hardest problems were
exponents. I would make the point that if you can do these problems,
you can do stats. I stopped giving the quiz when I realized that it was
creating anxiety instead of reducing it. I had many students who
couldn't even do those problems. (For example, one student answered
that 5-squared = 10.)
Now (or the last time I taught stats), I assigned a take-home quiz on
math operations (maybe from the back of the text?) and emphasized that
anyone who had problems with those topics should seek a tutor
immediately.
Students in my classes must know all the formulas through the t. I
emphasize *know* and not memorize, though. I believe that if they
really understand what the test measures, that they should be able to
generate the formula on their own. (So, obviously I am teaching
definitional formulas.) They must show all the work, using the
definitional formulas, and have points taken off if they don't. My
personal view is that they will probably never use either the
definitional or computational formulas again (b/c they will use SPSS or
calculators) so I would rather have them practice the definitional
formulas, to reinforce the theory.
Once we are out of the t's, then I give them the formulas. On the
(cumulative) final, I give them a list of formulas but they have to be
able to pick the right one.
I have done little with SPSS b/c at my former school, that was done by
the TAs. They did have some limited exposure. Now, if I teach stats, I
would do both the lecture and the lab, and I would incorporate SPSS into
the lab.
I feel that stats is somewhat "special," in that it seems to evoke more
anxiety than other classes. So, I do a number of things in stats that I
don't do in other classes, such as:
- giving re-tests (usually on the Sat. following the exam). I make
sure that I grade the first exam early and e-mail grades to students so
that they can decide early on if they want to retake the exam. I use
the higher of the grades toward their final course grade. I think that
knowing they have a second chance reduces their anxiety and makes them
feel that the grades are very fair.
- I also put all my notes, old stats tests, and sample answers, up on
my website. They report using them often, although I don't have a
counter to track that. I have taken that info down for the moment, but I
could put it back up if anyone is interested.
Good luck planning. Stats is my favorite course to teach!
Marcia
Marcia J. McKinley-Pace, J.D., Ph.D.
Assistant Professor of Psychology
Mount St. Mary's College
Emmitsburg, MD 21727
[EMAIL PROTECTED]
www.erols.com/mcpace
____________
Marc wrote:
>I've been deebating how much hand calculation I should include in the
>course, and tend to agree that some hand calculation is a good thing.
But,
>I see little to no reason to have people hand calculate ANOVAs by hand.
>Right now my leaning is towards bringing SPSS into the course early so
I
>can make sure that they understand how to interpret the output when
they
>move on to the methods course, along with knowing which test(s) are
>appropriate for different situations. My biggest problem in teaching
the
>methods course now is having to "re-teach" stats to the students, and I
>often find myself with students who have never used SPSS in their stats
>course which forces me to deal with topics that I feel they should know
>prior to the methods course. (But that's another topic...)
>I'd love to hear responses from others on this topic as I try to refine
my
>plan for the coming semester.
- Marc
G. Marc Turner, MEd
Lecturer & Head of Computer Operations
Department of Psychology
Southwest Texas State University
San Marcos, TX 78666
phone: (512)245-2526
email: [EMAIL PROTECTED]
winmail.dat
