Re: Fisher's Exact Test / Freeman-Halton-Test

Posted by Bruce Weaver on Jun 05, 2012; 3:28pm
URL: http://spssx-discussion.165.s1.nabble.com/Fisher-s-Exact-Test-Freeman-Halton-Test-tp5713482p5713517.html

Frank, I've just run your code using v19 running under Windoze 7, and I get the results you show for v20, including the p-value for FET.  My v19 is patched to 19.0.0.2.  Does your university have the latest patches installed?

HTH.

drfg2008 wrote

Still a problem with Fisher's Exact test that somebody asked me (maybe a bug in SPSS 19 ? ) :

The same syntax run on SPSS 19 produces different results than run from SPSS20

Here the syntax (again)

DATA LIST LIST
 / line AB count.
BEGIN DATA.
1 1 3
1 2 0
2 1 1
2 2 3
3 1 3
3 2 7
END DATA.

VALUE LABELS AB 1'A' 2'B'.

WEIGHT BY count.

CROSSTABS
  /TABLES=line BY AB
  /FORMAT=AVALUE TABLES
  /STATISTICS=CHISQ
  /CELLS=COUNT
  /COUNT ROUND CELL
  /METHOD=EXACT TIMER(1).


Here the result on SPSS 19 (wrong - University licence / run on a campus machine). Fisher's Exact test should not result 1.000 but 0.072.

Chi-Square Tests
Value    df    Asymp. Sig. (2-sided)    Exact Sig. (2-sided)    Exact Sig. (1-sided)    Point Probability
Pearson Chi-Square    5.234a    2    .073    .051        
Likelihood Ratio    6.319    2    .042    .037        
Fisher's Exact Test    4.609              1.000        
Linear-by-Linear Association    3.191b    1    .074    .119    .070    .052
N of Valid Cases    17                        
a. 5 cells (83.3%) have expected count less than 5. The minimum expected count is 1.24.
b. The standardized statistic is 1.786.



Here the result on SPSS 20  on my computer (correct)


Chi-Square Tests                  
   Value   df   Asymp. Sig. (2-sided)   Exact Sig. (2-sided)   Exact Sig. (1-sided)   Point Probability
Pearson Chi-Square   5.234a   2   .073   .072      
Likelihood Ratio   6.319   2   .042   .072      
Fisher's Exact Test   4.609         .072      
Linear-by-Linear Association   3.191b   1   .074   .119   .070   .052
N of Valid Cases   17              
a 5 cells (83.3%) have expected count less than 5. The minimum expected count is 1.24.                  
b The standardized statistic is 1.786.