Marken's spreadsheet II

[From Bruce Abbott (971204.1210 EST)]

Rick Marken (971204.08250) --

Bruce Abbott (971204.0945 EST)

A repeated measures design is an individual subject design replicated
as many times as there are subjects. Nothing precludes the
investigator from examining the data for each individual separately,
and large differences in the trends of individual subjects would
show up as a large error term within the ANOVA, as well as big
standard errors around at least some of the individual means.

Everything you say here is wrong.

So you wish to claim that a repeated measures design is NOT an individual
subject design replicated as many times as there are subjects? I'm sure
that any competent statistician would be very surprised to learn of this.
Why don't you ask one and see?

You also wish to claim that something precludes the investigator from
examining the data for each individual separately. What, pray tell,
prevents him or her from doing so? They are right there in the data matrix.

Large differences in the trends of individual subjects would NOT show up as
a large error term in the ANOVA? I'd like to see a demonstration of this,
and you don't need a spreadsheet. I'd be happy if you would provide a table
showing the individual subject values at each level of the IV and the
one-factor repeated measures ANOVA results. Just make up some data that
support your case.

There will be NO big standard errors around at least some of the treatment
means? O.K., don't forget to provide the standard errors for each mean, too.

If you want, I can run the ANOVA for you. Then you only have to provide the
table of data. Fair enough?

Regarding individual trends, I consider it a "large difference
in trend" if the effect of A on B is positive linear for one
individual, negative linear for another, U shaped for another
and inverted U for still another. This is how the individuals
responded in my "oh god, not again" spreadsheet.

Fine, just copy the individual data into a table and post it.

Regards,

Bruce

[From Rick Marken (971204.1100)]

Bruce Abbott (971204.1210 EST) --

If you want, I can run the ANOVA for you. Then you only have to
provide the table of data. Fair enough?

OK. I had to try to recreate it in a rush here. I don't know
if this will show up as significant (.01). While playing with
the data I realized that it's quite easy to get a significant
result when all the subjects have _very_ different functions
relating IV to DV. It all depends on the range of IV values
you use. For some IV values, you only pick up the increasing
part of the range for a non-monotonic subject. But I figure
you are intersted in seeing if it is possible to get a
significant group effect when the individual subject effects
are visibly different within the range of the IV used. So
here's some data that _might_ be statistically significant

IV 0.00 1.00 2.00

S1 0.01 0.43 0.46
S2 0.11 0.20 0.11
S3 -0.04 0.00 0.01
S4 0.02 0.12 0.22
S5 0.03 0.53 0.74
S6 0.00 0.14 0.57
S7 0.00 0.08 -0.11
S8 0.01 0.06 0.41
S9 0.00 -0.05 -0.10
S10 0.10 0.09 0.08

XBar0.02 0.16 0.24

If this data doesn't produce a significant F I'll try adjusting
it so that it does (I'll be out of town for a couple of days so
this will have to happen on the weekend), Incidentally, the actual
relationship between IV and DV is different (a different mathematical
function relates IV to DV) for every subject in the experiment.

Best

Rick

ยทยทยท

--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken