Re: By trial, or by subject?

2001-01-16 Thread Elliot Cramer

In sci.stat.edu Rich Ulrich [EMAIL PROTECTED] wrote:

: If it is some other data...  When you have multiple replications,
: sometimes you don't want the *mean* -- for "best single performance"
: you might select maximum or minimum.  Or you might consider a trimmed

or in a situation where you expect an increasing trend, the final or last
few obs might be best



=
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
  http://jse.stat.ncsu.edu/
=



Re: By trial, or by subject?

2001-01-15 Thread Jeff Rasmussen
William Dunlap at Tulane has done some research on this.  Do a lit search on his name.

Jeff Rasmussen

Spirit of Tao Te Ching  Images of Taoism   http://psychology.iupui.edu/tao

TAO TE CHING... the Book, the Song, the Miniseries   http://www.symynet.com

Zen/Tao 10 Oxherding Pictures  http://psychology.iupui.edu/ox/main.htm

= Instructions for joining and leaving this list and remarks about the problem of INAPPROPRIATE MESSAGES are available at http://jse.stat.ncsu.edu/ = 

Re: By trial, or by subject?

2001-01-15 Thread Sky

Hi Jim,
Is it correct that you measured every subject 100 times (25 times in all 4
drugs/dose groups)? Then it might be a good idea to use a repeated
measurement test. This tests takes into account that people may have a
different 'base' temperature; it takes away some of the within group error
and may give your test a greater power. It's easiest to use the mean of your
25 observations in the 4 groups, also because you don't seem to be
interested in differences in temperature over time or something.
Sky


Jim Kroger heeft geschreven in bericht ...
Hello, I've received some expert help here on a couple previous occasions
(thanks). I have an issue bothering me, which I'd like to present to you.

I'm doing a two-way, 2X2 ANOVA. Suppose I have 20 subjects, and each has
25 observations of the following types:

drug1-doseA (25 for each subject)
drug1-doseB ( " )
drug2-doseA
drug2-doseB

Each observation consists of the subject's temperature. The objective is
to determine whether there is a main effect of either factor (drug type
and doseage) on temperature, and whether there is an interaction between
the two.

My question is, should I determine an average for each subject in each of
the four cells, and use this as data to put into the ANOVA, or should I
put the raw trials themselves into the ANOVA? There would be 80 datapoints
and 2000 data points respectively, 20 or 500 per cell. I have seen both
approaches taken but never heard a satisfactory justification. It would
seem they are not equivalent since the latter, having more observations,
has greater power. What are the implications of each option?

Thank you much,
Jim





=
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
  http://jse.stat.ncsu.edu/
=



Re: By trial, or by subject?

2001-01-15 Thread Elliot Cramer

In sci.stat.edu Jim Kroger [EMAIL PROTECTED] wrote:
: Hello, I've received some expert help here on a couple previous occasions
: (thanks). I have an issue bothering me, which I'd like to present to you.

: I'm doing a two-way, 2X2 ANOVA. Suppose I have 20 subjects, and each has
: 25 observations of the following types:

no matter how you do it the proper analysis will be equivalent to using
submect means.  adding a replication factor would permit other tests
(which you are not interested in) but due to their correlation structure
the tests would not be valid.


=
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
  http://jse.stat.ncsu.edu/
=



Re: By trial, or by subject?

2001-01-15 Thread Rich Ulrich

On 15 Jan 2001 17:23:11 GMT, Elliot Cramer [EMAIL PROTECTED]
wrote:

 In sci.stat.edu Jim Kroger [EMAIL PROTECTED] wrote:
 : Hello, I've received some expert help here on a couple previous occasions
 : (thanks). I have an issue bothering me, which I'd like to present to you.
 
 : I'm doing a two-way, 2X2 ANOVA. Suppose I have 20 subjects, and each has
 : 25 observations of the following types:
 
 no matter how you do it the proper analysis will be equivalent to using
 submect means.  adding a replication factor would permit other tests
 (which you are not interested in) but due to their correlation structure
 the tests would not be valid.

As it was stated, Elliot nailed the first question.  
A valid ANOVA treats the average (sum) as its starting point.

If this isn't the REAL data, then something else might be true.

If those are 25 0/1  items, then you might do better to model them as
logistic, or something else other than a linear sum - as someone 
suggested.

If it is some other data...  When you have multiple replications,
sometimes you don't want the *mean* -- for "best single performance"
you might select maximum or minimum.  Or you might consider a trimmed
mean.  Or best consistency could be indicated by smallest SD, or the
best performance over a specific range (that could be, something about
the Confidence Interval).

-- 
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html


=
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
  http://jse.stat.ncsu.edu/
=



Re: By trial, or by subject?

2001-01-15 Thread Donald Burrill

Jim, a few comments in addition to those made by other respondents:

On Mon, 15 Jan 2001, Jim Kroger wrote in part:

 I'm doing a two-way, 2X2 ANOVA. Suppose I have 20 subjects, and each has
 25 observations of the following types:
 
 drug1-doseA (25 for each subject)
 drug1-doseB ( " )
 drug2-doseA
 drug2-doseB
You have not stated how many subjects (Ss) receive each 
drug/dose combination.  If all 20 Ss receive all 4 combinations, as 
implied by your assertion of 80 data points (below), then the design is 
not completely given:  you must also have a factor representing the order 
in which the combinations were encountered by each S.  If all Ss got the 
various drug/dose combinations in the same order, you have no way of 
telling whether (e.g.) the first combination affected responses to the 
later combinations, nor even in which direction.  Also in this case, as 
one other respondent pointed out, you have a repeated-measures design:
that is, instead of  S(G x E)  as in the usual 2-way ANOVA (using S for 
Subjects, G for druG, E for dosE), you have  S x G x E,  which implies 
(inter alia) different error terms for G, E, and GE.

(In that notation, using R for the raw data values within each S, the two 
designs above would be  R(S(B x E)  and  R(S) x G x E .  As remarked 
below, the same ratios of mean squares are computed, whether R is 
explicitly accounted for in the design or not.)

 ... The objective is to determine whether there is a main effect of 
 either factor (drug type and doseage) on temperature, and whether there 
 is an interaction between the two.
 
 My question is, should I determine an average for each subject in each 
 of the four cells, and use this as data to put into the ANOVA, or 
 should I put the raw trials themselves into the ANOVA?

As others have pointed out, it makes no difference.  Your 
hypotheses are tested by comparisons among the cell means, and the 
denominator mean square for each hypothesis will be the estimated 
sampling variance of the means in question.  Whether you use the mean of 
25 trials for each S, or the 25 raw trials themselves, you get the same 
numbers. 

 There would be 80 datapoints and 2000 data points respectively, 20 or 
 500 per cell.  I have seen both approaches taken but never heard a 
 satisfactory justification.  It would seem they are not equivalent 
 since the latter, having more observations, has greater power. 

As remarked above, this is not true.  One has the same numerical 
estimates, and the same numbers of degrees of freedom in numerator and 
denominator, for each hypothesis test.  As other respondents have 
mentioned, if the raw data are a 0/1 dichotomy, there may be an advantage 
in using a logistic rather than a normal model;  but if the proportions 
(the several cell means) are not very close to 0 or 1, say between .25 
and .75, there will not be much difference in the results of the 
analysis. 

 --
 Donald F. Burrill[EMAIL PROTECTED]
 348 Hyde Hall, Plymouth State College,  [EMAIL PROTECTED]
 MSC #29, Plymouth, NH 03264 (603) 535-2597
 Department of Mathematics, Boston University[EMAIL PROTECTED]
 111 Cummington Street, room 261, Boston, MA 02215   (617) 353-5288
 184 Nashua Road, Bedford, NH 03110  (603) 471-7128



=
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
  http://jse.stat.ncsu.edu/
=