Similar to the chi-squared test but testing several (repeated) 2x2 contingency tables simultaneously, in order to control for a nuisance factor (Mantel & Haenszel 1959). An example could be controlling for season when testing for the effect of a medicine that was used in repeated experiments through a year. The 2x2 tables must be given in consecutive rows in the spreadsheet (first, one 2x2 table, followed below by the next table, etc.).
Our algorithm follows McDonald (2014). For details, see the Past manual.
In addition, the common odds ratio (equal to one for total independence) is calculated using the Mantel-Haenszel (1959) estimate.
Missing data not supported.
References
Mantel, N. & W. Haenszel. 1959. Statistical aspects of the analysis of data from retrospective studies of disease. Journal of the National Cancer Institute 22:719-748.
McDonald, J.H. 2014. Handbook of Biological Statistics (3rd ed.). Sparky House Publishing, Baltimore, Maryland.