I need a bit help for interpreting this results…
I did a correlation analysis with R with the assocstats function.
The result is:
$summary
Call: xtabs(formula = ~MH[, i] + MH[, j], data = MH)
Number of cases in table: 2306
Number of factors: 2
Test for independence of all factors:
Chisq = 2806.6, df = 3318, p-value = 1
Chi-squared approximation may be incorrect
$object
X^2 df P(> X^2)
Likelihood Ratio 1036.1 3318 1
Pearson 2806.6 3318 1
Phi-Coefficient : 1.103
Contingency Coeff.: 0.741
Cramer's V : 0.637
$summary
Call: xtabs(formula = ~MH[, i] + MH[, j], data = MH)
Number of cases in table: 2343
Number of factors: 2
Test for independence of all factors:
Chisq = 118.46, df = 73, p-value = 0.000611
Chi-squared approximation may be incorrect
$object
X^2 df P(> X^2)
Likelihood Ratio 130.83 73 3.8115e-05
Pearson 118.46 73 6.1100e-04
Phi-Coefficient : 0.225
Contingency Coeff.: 0.219
Cramer's V : 0.225
But what does the p-value of 1 in this case mean if Cramer's V is 0.637 and therefor show a strong corelation or with 0.000611 with a Cramer's V of 0.225?
And does anyone of you know a good rule of thumb for interpreting Cramer's V?
Thanks for your help.
Best Answer
Look at the number of df. 3318 df in a crosstabs is problematic. It's especially problematic when n is 2306. But I can't think of a good reason to have a crosstabs table with so many cells. All of the results are, essentially, meaningless. If you describe what you are trying to do and why you have so many cells, and what your research question is, and so on, then perhaps someone (maybe even I) will be able to help