[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

census stats (was Re: Fp. oeseri )

On 2 May 2001, at 6:52, Tom Grady wrote:

> Well .. I am going to make your day ..
> 
> I have only 4 replies indicating Fp. oeseri

Thanks for making my day and now how depressing it is...:-(

> The statistic based of people to killie species is about 1:1     about 850 
> people to 800 species.
> The statistic based on people to populations is 1:4   about 850 people to 
> 3000 populations (actually more probably)

So in other words the situation is quite bleak for the dull little fish 
and quite rosy for those that stand out from the croud.
When we decide which strains make it to SMC and KCC statis we 
will have to be very selective with fish such as oeseri and walkeri 
GH 74/2 taking the No.1 spots next to say: N. orthonotus Mkuzi 
GR 99/1 which we can certainly get again but is probably just as 
rare in the hobby as the latter species.
 
> The problem with statistics with Killies is this.   I have 150 responses 
> more or less.  If we consider that an average sampling (which I do not 
> since it was volunteered) then we should be able to extrapolate (wow - big 
> words) a multiple of 5.67 x the recorded data   4 reported populations = 
> 22.6 people keeping the fish.

Look at the sample: what is the spread between the number of 
species maintained per person.
Set up a graph with number of species/strains on the x-axis and 
the number of people that maintain that number of species/strains 
on the y-axis. If you are not getting a bell curve with an even spread 
then you know the sampling is out. If the number of people keeping 
50 species is the same as the numbe rod people keeping 1 
species is the same (+/-) and the average keeper has 25 sp. then 
you will see you have a pretty even spread. Then even while you 
may of missed many species you can still derive info form the 
sampling you have. If the curve is really skewed then no real 
conclusions can be made other than a better sampling is needed.

The club census you have should give much better results as they 
would represent a good mix of advanced (50+spp) to beginers (10- 
spp). The average keeper has about say, 15-20 spp.

> Unfortunately - this was not the right type of sampling to get an active 
> reading for statistical purposes so now we have read into it what we 
> want.  My attitude is simple.  There are 4 recorded reports of the species.

This is the best we can hope for given the poor participation. Since 
only about 17% of the "species pool" was sampled we have to 
assume that +/- 15 other people have that one fish that was tallied 
only once... Of course this would be totally wrong. We can assume 
that there are about 5x as amny people who have AUS than was 
counted with some confidence but some more statistical working 
will have to be done for those species that are mentioned rarely. I 
think that 20 out of 850 people keeping Fp. oeseri is a sound 
prediction as it is an attractive fish. Perhaps you should make a 
few predictions and then ask (nicely) for people to get back to you 
and then count how many people actually have Fp. oeseri or 
mimbon...

Oh well, enough rambling... 

Bye

Tyrone Genade
tyronegenade at yahoo_com
http://www.geocities.com/tyronegenade

*************************************************************
P450 Lab, Biochemistry Department
University of Stellenbosch, 7602, South Africa
Ph: +27-021-808-5876, fax: +27-021-808-5863
**********************************************************

"Those who trust in the Lord for help will find their 
               strength renewed."
                          Isaiah 40:31
---------------
See http://www.aka.org/AKA/subkillietalk.html to unsubscribe
Follow-Ups: References: