Hi all
This is probably a very silly question, but please answer anyway.
I have read tons of material describing how the degrees of freedom is
calculated when doing a chi2 analysis for at simple crosstabulation, so
I think I know the formula.
But could anyone please tell me, WHY the degrees of freedom is
important, when comparing the observed distribution to the expected
distribution.
Could it be somthing like: The more cells that are free to vary, the
bigger the uncertenty of the observede values being equal to the
expected?? This would account for the chi-value getting bigger with
increasing degrees of freedom??
Or how would you explain it??
Please help me out
Thanks in advance
Peter L�vgreen
Denmark
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
