This test is effective applied to the “before and after” designs. Nominal categorical measurement would be appropriate for assessing the “before to after” change. To test the significance of any observed changes by this method, we need to use a fourfold table of frequencies. The general features of of such a table are illustrated as follow
B and C are the frequency of individuals who responded the same both before and after. Those observations are not used in Mc Nemar test. (A + D) is the total number of people whose responses changed. So we would expect that when H0 is true, the expected frequency in each of the two cells is ½(A+D). in other words, ½(A+D) changes from (+) to (-) and ½(A+D) changes from (-) to (+).
The statistics test is chi-squares test
This approximation will be more precise if a correction for continuity is made, then the formula will become as follow
When H0 is true, it is asymptotically distributed as chi square with df = 1.
Critical region: χ2 > χ2(α ; 1)
See you in other writing sessions, have enjoying statistics.
Any questions, send to the e-mail address: firstname.lastname@example.org
*) Writer is a lecturer in The Institute of Statistics, Jakarta, Indonesia.
Bachelor of Statistics from The Institute of Statistics, Jakarta, Indonesia.
Master of Science in Experimental Statistics from NMSU, USA.