Re: Effect Size and Sample Size
- From: Bruce Weaver <bweaver@xxxxxxxxxxxx>
- Date: Sat, 12 Jan 2008 16:16:23 -0500
David Winsemius wrote:
Ryan <Ryan.Andrew.Black@xxxxxxxxx> wrote in
news:1c5bbd85-9681-4775-a10c-6a58a7fb3ce6@xxxxxxxxxxxxxxxxxxxxxxxxxxxx:
On Jan 10, 12:10 am, beginner1....@xxxxxxxxxxx wrote:On Jan 9, 11:26 am, Richard Ulrich <Rich.Ulr...@xxxxxxxxxxx> wrote:n
On Tue, 8 Jan 2008 23:16:26 -0800 (PST),
beginner1....@xxxxxxxxxxx wrote:
I would like to compare 2 groups (treatment and control) using
a t- test. Due to expense, I will only be able to perform a
small test ona total of 4 (2 treatment and 2 control). Would calculating
sample sizebased off of theeffectsizecalculated from this small test be
valid (difference between groups/pool sd)? Is there a rule for
theMy two cents...Can I ask, how what is the "smallest" N that can be used for aneeded to be able to calculate theeffectsizefor samplesizeRequirements on Ns are generally "rules of thumb" that
estimates?
apply in a particular area, and not hard-and-fast rules;
unless you are talking about terms that become undefined
(see below).
Doing a t-test of N=2 vs. 2 is something that I've thought
about a few times, but I don't think I've ever done it, or
even seen it done. However, I don't read papers on the
preliminary sort of laboratory experiments where that would
arise. Years ago, my lab friends who used tiny samples, and
whom I asked about it, said that they did not bother about
t-tests, like, "a hugeeffecteither exists or it doesn't, and
the replication is to make sure than you didn't screw up
the lab procedures." I don't know if that is still the same.
So - if you are working in an area when they would accept
or expect a t-test on Ns of 2, maybe folks would be happy
to see an "effectsize", too. You do want to work with
scores that are figured and written out with enough precision
that you are not screwed up by round-off error, for both the
t-test and theeffectsize. "Valid" is an interesting word to use here -- The estimate
would not be as reliable a number (precise estimate),
compared to those with much larger Ns. With 2 d.f., I *think* there is nothing to say that it is not a "valid
estimate" in the technical sense; though I believe that with 1 d.f., the
variance of the estimate itself would be undefined. That
1 d.f. test would allow for a computable "effectsize" that
would arguably be "not-valid".
I don't recall seeing these issues discussed, so I won't feel
bad if someone else chimes in. --
Rich Ulrich,
wpi...@xxxxxxxxxxxx://www.pitt.edu/~wpilib/index.html
t-test (or statistics in general) and why? Is there a specific
text that one could recommend OR online reference?- Hide quoted
text -
- Show quoted text -
The answer to your question depends upon how well the variable you
are analyzing meets the assumptions of the t test. Generally
speaking, as long as (1) the variable is normally distributed in
each group and (2) the variance of the data is reliably the same
between groups, you could have a very small sample size.
Just curious. How do you propose checking anu of your assumptions with "a very small sample size"? And aren't the "assumptions of the t-test" that you are working with a data situation that is asymptotically ....large?
Presumably one would have to know on the basis of previous data that the populations are close enough to normal and the variances close enough to homogeneous.
--
Bruce Weaver
bweaver@xxxxxxxxxxxx
www.angelfire.com/wv/bwhomedir
"When all else fails, RTFM."
.
- References:
- Effect Size and Sample Size
- From: beginner1 . mat
- Re: Effect Size and Sample Size
- From: Richard Ulrich
- Re: Effect Size and Sample Size
- From: beginner1 . mat
- Re: Effect Size and Sample Size
- From: Ryan
- Re: Effect Size and Sample Size
- From: David Winsemius
- Effect Size and Sample Size
- Prev by Date: Re: Effect Size and Sample Size
- Next by Date: Re: Presenting unorthodox factor analysis findings
- Previous by thread: Re: Effect Size and Sample Size
- Next by thread: Re: Effect Size and Sample Size
- Index(es):
Relevant Pages
|
