Fisher's Exact Test / Freeman-Halton-Test
Locked
 
|
This post was updated on .
I always thought, SPSS would compute "Fisher's Exact Test" equivalant to the Freeman-Halton-Test. But this doesn't seem to be so.
Following data example: line A B Sum 1 3 0 3 2 1 3 4 3 3 7 10 Sum 7 10 17 SPSS result: p=0,152 ------------------------------------------------- Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Point Probability Pearson Chi-Square 4.635a 2 .099 .152 Likelihood Ratio 5.783 2 .055 .152 Fisher's Exact Test 4.194 .152 Linear-by-Linear Association 3.602b 1 .058 .070 .055 .043 N of Valid Cases 18 a 5 cells (83.3%) have expected count less than 5. The minimum expected count is 1.33. b The standardized statistic is 1.898. ------------------------------------------------- But the Freeman-Halton p-Value should be p(A) = p(B) = 0.07229535170711625 according to: -> http://vassarstats.net/fisher2x3.html
Dr. Frank Gaeth
|
Re: Fisher's Exact Test / Freeman-Halton-Test
|
|
What version are you running?
Here is what SPSS 20 produces when you use the Exact button: 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. The "gold standard" would be to run in statxact but I don't have it on my machine. Tony Babinec [hidden email] -----Original Message----- From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of drfg2008 Sent: Saturday, June 02, 2012 2:08 AM To: [hidden email] Subject: Fisher's Exact Test / Freeman-Halton-Test I always thought, SPSS would compute "Fisher's Exact Test" equivilant to the Freeman-Halton-Test. But this doesn't seem to be so. Following data example: A B Sum 1 3 0 3 2 1 3 4 3 3 7 10 Sum 7 10 17 SPSS result: p=0,152 ------------------------------------------------- Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Point Probability Pearson Chi-Square 4.635a 2 .099 .152 Likelihood Ratio 5.783 2 .055 .152 Fisher's Exact Test 4.194 .152 Linear-by-Linear Association 3.602b 1 .058 .070 .055 .043 N of Valid Cases 18 a 5 cells (83.3%) have expected count less than 5. The minimum expected count is 1.33. b The standardized statistic is 1.898. ------------------------------------------------- But the Freeman-Halton p-Value should be p(A) = p(B) = 0.07229535170711625 according to: -> http://vassarstats.net/fisher2x3.html ----- Dr. Frank Gaeth FU-Berlin -- View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Fisher-s-Exact-Test-Freeman-Ha lton-Test-tp5713482.html Sent from the SPSSX Discussion mailing list archive at Nabble.com. ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD ===================== To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD |
Re: Fisher's Exact Test / Freeman-Halton-Test
Administrator
|
In reply to this post by drfg2008
Maybe you are running it incorrectly?
*PLEASE* post your syntax in the future. It is difficult to try to guess what you are fumbling around with. Not even your other stats are correct from that table of counts. ------------------------------------------------------------- My results using the following syntax: data list free / x y w. begin data 1 1 3 1 2 1 1 3 3 2 1 0 2 2 3 2 3 7 end data. weight by w. CROSSTABS/TABLES=y BY x /STATISTIC=CHISQ/METHOD=EXACT TIMER(5). Chi-Square Tests Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) Point Probability Pearson Chi-Square 5.234 2 .073 .072 Likelihood Ratio 6.319 2 .042 .072 Fisher's Exact Test 4.609 .072 Linear-by-Linear Association 3.191 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.
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me. --- "Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis." Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?" |
Re: Fisher's Exact Test / Freeman-Halton-Test
|
|
In reply to this post by drfg2008
Your summary report says --
N of Valid Cases 18 whereas the count in the 2x3 table is 17. So your statistics are not a report on the table shown. -- Rich Ulrich > Date: Sat, 2 Jun 2012 00:08:14 -0700 > From: [hidden email] > Subject: Fisher's Exact Test / Freeman-Halton-Test > To: [hidden email] > > I always thought, SPSS would compute "Fisher's Exact Test" equivilant to the > Freeman-Halton-Test. But this doesn't seem to be so. > > Following data example: > > A B Sum > 1 3 0 3 > 2 1 3 4 > 3 3 7 10 > Sum 7 10 17 > > > SPSS result: p=0,152 > > ------------------------------------------------- > Chi-Square Tests > Value df Asymp. Sig. (2-sided) Exact Sig. (2-sided) Exact Sig. (1-sided) > Point Probability > Pearson Chi-Square 4.635a 2 .099 .152 > Likelihood Ratio 5.783 2 .055 .152 > Fisher's Exact Test 4.194 .152 > Linear-by-Linear Association 3.602b 1 .058 .070 .055 .043 > N of Valid Cases 18 > a 5 cells (83.3%) have expected count less than 5. The minimum expected > count is 1.33. > b The standardized statistic is 1.898. > ------------------------------------------------- > > But the Freeman-Halton p-Value should be p(A) = p(B) = 0.07229535170711625 > according to: > > -> http://vassarstats.net/fisher2x3.html > |
|
|
This post was updated on .
yes, it was simply a programming problem.
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).
Dr. Frank Gaeth
|
|
|
In reply to this post by drfg2008
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.
Dr. Frank Gaeth
|
Re: Fisher's Exact Test / Freeman-Halton-Test
Administrator
|
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.
--
Bruce Weaver bweaver@lakeheadu.ca http://sites.google.com/a/lakeheadu.ca/bweaver/ "When all else fails, RTFM." PLEASE NOTE THE FOLLOWING: 1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above. 2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/). |
|
|
it seems to be a known problem with SPSS19 and has been solved already by a patch
http://gruener.userpage.fu-berlin.de/IBM_SPSS_Statistics_19_0_Fix_Pack_2_Fix_List.pdf Thanks
Dr. Frank Gaeth
|
Automatic reply: Fisher's Exact Test / Freeman-Halton-Test
|
|
Hello, I will be out of the office on Friday, June 6, 2012. If you are a graduate applicant and have a question about the status of your submitted application, please contact the department to which you applied. They have your application file and know the status of your
application. If you are having an issue with the online application system, please contact our Help Desk ([hidden email]). Best, Michael Toner |
«
Return to SPSSX Discussion
|
1 view|%1 views
| Free forum by Nabble | Edit this page |