|
|
I'm using Version 13.0. I'm doing some 2x2 crosstabs and requesting FET.
Most of the time I have n > 200 < 300; and often expected cell frequencies
<5. The output does not include the FET value, but it does include the p-
value for FET. The p-values are .000. Since I'm need to use FET for a
great deal of my dissertation analysis, I want to be able to explain to my
committee why I can't use it in some situations.
CROSSTABS
/TABLES=q17E BY q17I
/FORMAT= AVALUE TABLES
/STATISTIC=CHISQ PHI CORR D BTAU CMH(1)
/CELLS= COUNT ROW COLUMN TOTAL ASRESID
/COUNT ROUND CELL
/BARCHART
/METHOD=EXACT TIMER(0).
Chi-Square Tests
Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided)
Exact Sig. (1-sided) Point Probability
Pearson Chi-Square 68.856 1 .000 .000 .000
Continuity Correction 65.655 1 .000
Likelihood Ratio 60.260 1 .000 .000 .000
Fisher's Exact Test .000 .000
Linear-by-Linear Association 68.552 1 .000 .000 .000
.000
N of Valid Cases 227
a Computed only for a 2x2 table
b 0 cells (.0%) have expected count less than 5. The minimum
expected count is 10.74.
c The standardized statistic is 8.280.
It seems that the Exact Tests module performs as expected most of the
time, but not for these tables? If its not a problem, is there an
accepted way to report the p-value without the FET value?
Thanks in advance,
Glenys Lafrance
|
Locked
