Related Experiment Videos
Statistical methods in epidemiology. VII. An overview of the chi2 test for 2 x 2 contingency table analysis
1Sheffield Children's Hospital, University of Sheffield, UK. a.s.rigby@sheffield.ac.uk
Purpose:
The odds ratio is an appropriate method of analysis for data in 2 x 2 contingency tables. However, other methods of analysis exist. One such method is based on the chi2 test of goodness-of-fit. Key players in the development of statistical theory include Pearson, Fisher and Yates.
Method:
Data are presented in the form of 2 x 2 contingency tables and a method of analysis based on the chi2 test is introduced. There are many variations of the basic test statistic, one of which is the chi2 test with Yates' continuity correction. The usefulness (or not) of Yates' continuity correction is discussed. Problems of interpretation when the method is applied to k x m tables are highlighted.
Results:
Some properties of the chi2 the test are illustrated by taking examples from the author's teaching experiences.
Conclusion:
Journal editors should be encouraged to give both observed and expected cell frequencies so that better information comes out of the chi2 test statistic.