Hi Allstat,
I have a problem using Chi-Square Test when I compare observed and
estimated frequencies. This test is not independent from the unit of scale.
The frequencies represent the interregional migration flow and the values
are big. For example, when I obtain a difference between observed and
estimated of 1000 persons at numerator I have (1000)^2!
This implies that the test is always significant (this means that I can not
accept the model under Hypothesis).
Is there another test? What can I do?
Thanks in advance,
Nicla
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|