Re: how to test if the variance of difference is smaller than another group?
- From: Bruce Weaver <bweaver@xxxxxxxxxxxx>
- Date: Mon, 21 Jan 2008 21:13:51 -0500
Lee Sander wrote:
Hi,
I have the following problem on which I would appreciate some
assistance.
I have two groups of people--some are Europeans while others are
Americans.
I am comparing how they perform some basic math problem during
different times of
the day (say, morning versus evening ) (i can describe their
performance by a positive real number).
On x-axis I plot their age and on y-axis I plot the similarity in
their performance during the two
times of the day (I measure similarity by computing the difference
between the scores received
during the two evaluations). I bin the ages: so I group people of age
25 to 29 in one bin, then
I bin 30 to 34 into another bin, and so forth.
I want to test if one of the groups show a larger difference in
performance than the other.
Please let me know if I can clarify the problem further.
thanks
Lee
Lee, why bin the ages? I'd look at a scatter-plot with X = actual age, and two different symbols for Europeans & Americans. The nature of that plot should be informative about what kind of model you need to create.
--
Bruce Weaver
bweaver@xxxxxxxxxxxx
www.angelfire.com/wv/bwhomedir
"When all else fails, RTFM."
.
- References:
- Prev by Date: Calculating Effect size Nonoverlap Area.
- Next by Date: Two Proportions Test vs Chi-square Test
- Previous by thread: how to test if the variance of difference is smaller than another group?
- Next by thread: Calculating Effect size Nonoverlap Area.
- Index(es):
Relevant Pages
|
|