avoid reporting the R^2, or try a
different model (non-linear)
You may also look at
http://noppa5.pc.helsinki.fi/uudet/corr/avp/avp20.html
--
Juha Puranen
Department of Statistics
P.O.Box 54 (Unioninkatu 37), 00014 University of Helsinki, Finland
http://noppa5.pc.helsinki.fi
/koe/corr/cor7a/cor7a.html
and ask if it is sensible to calculate or use correlation in this case.
--
Juha Puranen
Department of Statistics
P.O.Box 54 (Unioninkatu 37), 00014 University of Helsinki, Finland
http://noppa5.pc.helsinki.fi
, what happens, if different measurers have
different scales.
Works with ns4.7, ie5 opera5
Please comments.
I am also looking someone, who is expert in making measurements and
is ready to write some text on this problem.
regards
Juha
--
Juha Puranen
Department of Statistics
P.O.Box 54
or
Element of row i and column j.
In these cases the exact Fisher test is biased.
At least in Survo (may be in some other programs too) it is possible
make the test also in these cases. Look at
http://www.helsinki.fi/survo/q/qu1_03.html
regards
Juha
--
Juha Puranen
Department of Statistics
sification
1(.2)5 or
1.05(.1)5.05
so that everyone can see: "It is discrete "
Juha
--
Juha Puranen
Department of Statistics
P.O.Box 54 (Unioninkatu 37), 00014 University of Helsinki, Finland
http://n
"it was usefull".
Regards
Juha Puranen
--
Juha Puranen
Department of Statistics
P.O.Box 54 (Unioninkatu 37), 00014 University of Helsinki, Finland
http://noppa5.pc.helsinki.fi
===
This list is open t