The incorrect use of Chi-square analysis for paired data.
Julien I.E. Hoffman · PubMed · 1976
When results are classified in categories (for example, improved and unimproved, reactive and non-reactive) and the same patients are examined by two different treatments or two different tests, then the correct statistical analysis is by McNemars test. More often a Chi-square analysis is used; this is not only incorrect but also leads to erroneous conclusions.