Re: Trying to find significant factors in experimental results

Rich,

Thanks for you response. I've interspersed my comments below.

Best wishes,


Rob

On Apr 11, 1:22 am, Richard Ulrich <Rich.Ulr...@xxxxxxxxxxx> wrote:
On Thu, 10 Apr 2008 14:27:21 -0700 (PDT), Rob <rtshils...@xxxxxxxxx>
wrote:

Hi,

I've asked a 100 subjects to assess a quantity.  For simplicity,
assume I've asked them to estimate the length of a piece of wood.
I've also asked each subject a number of questions (sex, age, whether
they're short sighted or long sighted, if they were glasses, contact
lenses or nothing).

My hypothesis is that the results are independant of all question
answers.

Sometimes, these examples "for simplicity" turn
out to be misleading.  The standard hypothesis
has to do with an interesting *difference* or effect,
so there is an extra problem of trying to figure if
you really want to reverse things, or if you just asked
the question in a clumsy way.


In this case, the actual task is to judge distance based on looking at
a photograph. My hypothesis is that subjects ability to judge
distance is not related to their vision.

But I will take the question as it is asked.


Thanks!

Entertaining the hypothesis of no-difference usually
requires much larger samples than other experiments
with the same factors and outcomes.  That is because
you can *never*  establish that there is no difference
at all - which is almost impossible, philosophically,
for observational data -  You can only establish that the
difference is too small to be interesting.

This is what is done in bio-equivalence studies, which
you might want to look up, for comparison.

With multiple factors, you are stuck with providing
confidence limits for each of the observed differences,
and then arguing that they are each small.  If there is
any suggestion that these might combine ("bad vision,
no glasses"), you also need to discredit those things
that anyone else might argue for.


Very clear, and thanks for pointing me towards some more reading.


How can I prove this?  As a first attempt, I've partitioned the
subjects into every possible partition, based on the question answers,
where I've still got a minimum of ten subjects in the smallest
population.  I then ran ANOVA for the two formed populations, and
repeated for every possible partition.

This strikes me as being a bit clumsy, but I'm not sure how else I can
do this.  My reading of Factorial ANOVA suggests that every population
needs to be of the same size, and so this isn't possible with my data.

That is surely not the case.  The most powerful
tests have equal group sizes;  and cross-classifications
need cell sizes that are proportionate if the tests are
to remain independent and "unconfounded" with each
other.  There still can be tests.



I see. So fringe populations (eg short sighted, colour blind, wearing
contact lenses against those who don't) are less powerful as the
population is enormously mis-blanced? I presume this is because there
might be a larger inter quartile range (or similar) within the small
population because of the lower quantity of results?


Can anyone point me in the direction of a technique that could be used
to analyse my data?

If you don't want separate confidence intervals for
every difference that anyone might argue for, then
you may need to be more explicit about the variables
and the problem.


I'd be incredibly greatful if you would look closer, but I'd like to
take such discussion out of the news group. If you're happy, I will
email details of the experiment directly.

--
Rich Ulrich

http://www.pitt.edu/~wpilib/index.html

.

Relevant Pages

  • Re: How do I transfer speech profiles to other PCs?
    ... > different systems at a mouse-click and I want to use the same speech ... > by using shortcuts to a separate partition. ... >> personal account will be deleted without reading. ...
    (microsoft.public.office.misc)
  • Re: How do I transfer speech profiles to other PCs?
    ... mouse-click and I want to use the same speech profile in each one, ... OS's by using shortcuts to a separate partition. ... > account will be deleted without reading. ... personal account will be deleted without reading. ...
    (microsoft.public.office.misc)
  • Re: swap and /tmp
    ... During that time I remember reading that some programs that write to ... matter weather it is stored on a disk partition or a ramfs filesystem. ... I think you might be misunderstanding what ramfs does. ...
    (Debian-User)
  • Re: [SLE] Problems mounting Windows drive/partition
    ... There is a known flaw in the version of parted shipped with suse 9.1, ... which leads to strangeness reading LBA drives. ... excitement in a partition table" you'll find what I did to fix it. ...
    (SuSE)
  • Re: The Tolkien Universe
    ... GMT, Derek Broughton wrote: ... argue that there's practically no such thing ... that's better than not reading. ... Heinlein or Asimov were Fantasy writers (though Heinlein's /Glory ...
    (rec.arts.books.tolkien)