You are doing interesting research, but you should stop believing that
there is a statistical "test" out there that will magically reveal the
stories hidden in your data.
Despite the hype by its proponents I have never been able to use
Correspondence Analysis successfully unless I knew in advance what the
"answer" was.
Life is hard. Tell the story as you see it.
A book that you might find useful is not a statistics book, and it is out
of print, but here is the title and author:
Casual groups of monkeys and men; stochastic models of elemental social
systems
by Joel E. Cohen
I think it was originally published by MIT Press.
[EMAIL PROTECTED] wrote:
> Hi
>
> Thanks for your comments. Here is some further explanation of my data.
> I observed a group of lemurs twice per week for about five months. I
> observed each animal of 12 for one 25 minute period over the day during
> which at every 20 second interval I recorded:
>
> 1) The behaviour the animal was doing ( e.g. supine lie=55)
>
> 2) The weather ( I catogarized this my self (e.g Sun 3 very sunny, Sun
> one no sun shining into cage_)
>
> 3) The time slot ( for example 8.00- 8.30 = Time slot 1, 8.31- 9.00
> =time slot 2)
>
> 4)The day
>
> 5) The animal ( 1-12 e.g Cetus=2)
>
> I had two groups 1-6 family group, 7-12 Control group.
>
> After 8 weeks of no manipulation, I placed small mesh between the
> family group cage whilst leaving the control group alone. After 2 weeks
> this was substituted for large mesh, then after two weeks two groups
> were formed male and female (thus animals were let into the cages with
> one another).
>
> I want to look at the influence of my manipulations to monitor the
> reintroduction process but
>
> There are however factors that influenced my data other than the
> manipulations such as sun and time. I noticed that during sun 3 the
> animals tended to sit upright and supine line, whilst during sun 1 they
> tended to curl in a ball. I also noticed that first thing on a morning
> the animals I observed were particularly active whilst after about
> 12.00 they began to just lie about and do very little.
>
> So from this I know that these variables highly influence the
> behavioral repertoire of my animals.
>
> I have been using a count of the behaviours for say each animal and
> then running cluster analysis (to group animals together in respect to
> their behavioral repertoire), then using behavior count for each sun
> category to identify associations and grouping of behaviours with each
> sun value (and then running Principal components and Correspondence
> analysis) then for each time category etc.
>
> I also know that the way behavior changes with sun depends upon the
> individual so I had hoped to include individual (animal 1-12) into my
> analysis and thus assess the effects of both sun and animal together
> hence I hoped to use caconical correspondence analysis.
>
> The counts for some behaviours (e.g prone lie, self groom are very
> high= 456) whilst the counts for others is low (e.g chatter =. I ran a
> Kolmogren Smirnov that proved the distribution is not normal. The peaks
> of the distribution are in the heads and tails.
>
> The corrlations that I ran were also on counts of the behaviour data.
>
> I apologise if I sound very confused but up until about three weeks ago
> I didnt really know what any of these things were. I tried hiring
> numerous people to help me with my data analysis and even e mailed a
> consulting group at my university to help me but nobody seems to have
> time to even give me advice. As a result I have spend three weeks night
> and day trying to teach myself statistics from books and the internet
> and then to apply it to my data. Unfortunatly after a lot of hard work
> Im beginning to understand when and why to use specific tests and
> analysis but to complicate everything further my data seems to defy the
> assumtins of every analyses I have read!!
>
> Any further advice you could give me would be a great help
>
> Thanks
>
> Claire
>
> Sent via Deja.com http://www.deja.com/
> Before you buy.
--
*************************************************
`o^o' * Neil W. Henry ([EMAIL PROTECTED]) *
-<:>- * Virginia Commonwealth University *
_/ \_ * Richmond VA 23284-2014 *
*(804)828-1301 x124 (math sciences, 2037c Oliver) *
*FAX: 828-8785 http://saturn.vcu.edu/~nhenry *
*************************************************
===========================================================================
This list is open to everyone. Occasionally, less thoughtful
people send inappropriate messages. Please DO NOT COMPLAIN TO
THE POSTMASTER about these messages because the postmaster has no
way of controlling them, and excessive complaints will result in
termination of the list.
For information about this list, including information about the
problem of inappropriate messages and information about how to
unsubscribe, please see the web page at
http://jse.stat.ncsu.edu/
===========================================================================
