Chi-square / Fisher’s Exact Test

 

Independence Testing

 

        A test of independence is a statistical test which determines whether two categorical variables associate with each other. Chi-square test and Fisher’s exact test, which apply to contingency tables, are common approaches of independence testing.

 

        Fisher’s exact test is one of exact tests, while chi-squared test is based on approximation. When there are more than 20% of cells with  expected frequencies, Fisher’s exact test is preferrable than chi-squared test because applying approximation is inadequate.

 

        If the corresponding p-value of the test statistic is less than the chosen significance level, then the association between the two variables is statistically significant.

 

Chi-squared Test

 

        Chi-squared test is a non-parametric statistical hypothesis test with null hypothesis that the observed frequency is consistent with the expected frequency of certain events in a sample. If the frequency distribution of a categorical variable does not differ across groups from another categorical variable, the two variables can be concluded as independent.

 

        The test statistics is:

 ,  following  distribution with degrees of freedom .

        Where   is the observed frequency,   is the expected count,  is the number of rows of table and  is the number of columns.

 

Fisher’s Exact Test

 

         Fisher’s Exact Test is based on hypergeometric distribution of the counts in cells of the contingency table. For a 2 x 2 contingency table as shown below,

 

 

A

Not A

Total

B

Not B

Total

 

 

         the probability of obtaining such frequency distribution is

 

         Some statistical analysis software and packages, for example, SAS, supports Fisher’s exact test on general  x  tables.

 

 

SAS example codes

 

DATA PERSONS ; INPUT GROUP $ SUCCESS $ @@;

DATALINES ;

DRUG NO DRUG NO DRUG NO DRUG YES

DRUG YES DRUG YES DRUG YES DRUG YES

DRUG YES DRUG YES

PLACEBO NO PLACEBO NO PLACEBO YES PLACEBO YES

PLACEBO YES PLACEBO YES PLACEBO YES PLACEBO YES

PLACEBO YES PLACEBO YES

RUN ;

 

PROC FREQ DATA = PERSONS ;

TABLES GROUP * SUCCESS/ NOPERCENT NOCOL NOROW

CHISQ FISHER EXPECTED ;

RUN ;

 

Reference

1.     Kim H. Y. (2017). Statistical notes for clinical researchers: Chi-squared test and Fisher's exact test. Restorative dentistry & endodontics, 42(2), 152–155. https://doi.org/10.5395/rde.2017.42.2.152

2.     Hoffman, J. I. E. (2015). Biostatistics for Medical and Biomedical Practitioners. Academia Press. https://doi.org/10.1016/B978-0-12-802387-7.00013-5.