Changes

|}

|}

Applying the chi-square test to the Blount data violates the parameters under which the chi-square test produces accurate results. The number of 'expected mutants' is below one in all cells, and the chi-square test is not to be used when expected cell numbers are less than five. <ref>http://www.okstate.edu/ag/agedcm4h/academic/aged5980a/5980/newpage28.htm</ref> <ref>http://www.wellesley.edu/Psychology/Psych205/chisquareindep.html</ref><ref>http://www.graphpad.com/www/Book/Choose.htm</ref>

The chi-square test p-values are computed by comparing the test statistic to the chi-square distribution. It is generally assumed that the cell frequencies should be greater than five so that the statistic's distribution follows chi-square distribution. However, there is no consensus about what minimum cell frequency is necessary or how many expected values need to cross that threshold.

The total number of mutants, four in the first replay and eight in the third, fall below the lower cutoff for the chi-square, which is not felt to produce reliable results for n of less than twenty.<ref>http://faculty.chass.ncsu.edu/garson/PA765/chisq.htm</ref><ref>http://www.basic.northwestern.edu/statguidefiles/gf-dist_ass_viol.html</ref>

 

Under conditions of low n and low expected cell counts, the chi-square test is too conservative and results in estimated p-values that are too high.<ref>http://www.graphpad.com/www/Book/Choose.htm</ref><ref>http://mysite.du.edu/~jcalvert/econ/chisquar.htm</ref><ref>http://www.basic.northwestern.edu/statguidefiles/gf-dist_ass_viol.html</ref>

 

</ref> Thus the chi-square p-value of 0.004 calculated for replay experiment 2 cannot be interpreted as supporting Blount's conclusions, since this low p-value was derived through the inappropriate application of the test.

==References==

==References==