Extension of Fisher's exact test to 2-by-k contingency tables: a computer program in BASIC

Area of Science:

• Statistics
• Biostatistics
• Data Analysis

Background:

• Traditional chi-square tests may lack accuracy for small sample sizes in contingency table analysis.
• Fisher's exact probability test is a foundational method for analyzing categorical data.
• Accurate statistical evaluation is crucial for drawing reliable conclusions from research data.

Purpose of the Study:

• To develop a novel randomization method for analyzing 2 X k contingency table data.
• To provide a more accurate alternative to the chi-square test for specific data scenarios.
• To enhance the reliability of significance probability evaluations in statistical analysis.

Main Methods:

• Developed a randomization method for 2 X k contingency tables.
• The method is based on the principles of Fisher's exact probability test.
• A computer program in BASIC was created to implement the 'exact probability test'.

Main Results:

• The developed randomization method provides a more accurate evaluation of significance probability compared to the chi-square test.
• This improved accuracy is especially notable when analyzing data with small sample sizes.
• The 'exact probability test' offers a robust approach for 2 X k contingency table analysis.

Conclusions:

• The new randomization method is a valuable tool for analyzing 2 X k contingency tables, offering enhanced accuracy.
• Researchers can utilize this method, particularly when dealing with small sample sizes, for more precise statistical significance.
• The availability of a BASIC computer program facilitates the practical application of this advanced analytical technique.