Reporting Fisher's exact test (chi squared test)

[Christy127]

Hello,

I have conducted some chi squared tests but because the assumption for "minimum expected cell frequency" was violated, I'm using the value for Fisher's exact. However, I don't know how I can report the output.
For example, the normal chi squared would be reported like this: χ2 (1, N=51) = 14.88, p = .002.
But how do I report the Fisher's exact test?

Thank you very much!

 

[Karabiner]

It isn't called exact test for nothing. https://tinyurl.com/skw73aq

HTH

Karabiner

 

[Christy127]

I found this link (http://www.pmean.com/00/fishers.html) which states the following:

*******

"I did get an inquiry from someone about how to report Fisher's Exact Test in a publication. This is tricky, because this test only provides a p-value. Fisher's Exact Test has no formal test statistic and no critical value. Also, Fisher's Exact Test does not produce a confidence interval. [Actually, someone pointed out that you can get a confidence interval from Fisher's Exact Test, but it turns out that most software programs do not include this confidence interval. I have used StatXact software when I needed a confidence interval in this situation.]

Here's of abstracts on Medline that show how different people have reported results from Fisher's Exact test.

Follow-up examination at 7 to 10 days showed negative urine cultures in 76% of patients from the single-dose group and 89% from the multiple-dose group, a difference that was not statistically significant (P = 0.665, Fisher's exact test). [Medline]
However, endometrial carcinoma was significantly more likely to stain positively with C/EPB beta than benign endometrial glands (92% versus 3% respectively, p < 0.0001, Fisher's exact test). [Medline]
Statistical analysis (two-sided Fisher's exact test) confirmed that mutations in K-ras and p53 co-occurred less frequently than expected by chance (P < 0.01, Fisher's exact test). [Medline]
Nineteen of 99 de novo (19%) and 1 of 30 treatment-related (3%) leukemias carried the CYP3A4-V (P = 0.026; Fisher's Exact Test, FET). Nine of 42 de novo leukemias with MLL gene translocations (21%), and 0 of 22 treatment-related leukemias with MLL gene translocations carried the CYP3A4-V (P = 0. 016, FET). This relationship remained significant when 19 treatment-related leukemias with MLL gene translocations that followed epipodophyllotoxin exposure were compared with the same 42 de novo cases (P = 0.026, FET). [Medline]
Of the patients with pre-existing coronary heart disease, 1 out of 15 in the intervention group (6.7%) and 1 out of 23 in the control group (4.3%) were tested during follow up (P=0.851, one tailed Fisher's exact test). [Medline]
Along with an effective reduction of pathogenic bacteria colonizing the oropharynx, the rate of anastomotic insufficiency could be reduced significantly, to 6% in decontaminated animals compared with 80% in controls (P < 0.001 by Fisher's exact test). [Medline]​

Notice that most of these write-ups include a listing of the two proportions. Some of them also specify whether the test was one-sided or two-sided".

********

So, I understand that in a Fisher's exact test we only report the p value and then state that it is a Fisher exact test.

 

[Karabiner]

You are correct.

With kind regards

Karabiner