Question 267889

Calculate if your data differ significantly from the null hypothesis that neither individual is favoured, i.e. counts should be close to 50:50.

• Use χ2 tests; degrees of freedom are the (number of classes – 1) = 1. The critical values for 2 χ1 = 3.841459 for P = 0.05, 2 χ1 = 6.634897 for P = 0.01, and 2 χ1 = 10.82757 for P = 0.001. Remember to use counts, not percentages in the test.

• For multiple tests (say n) one should adjust the p-value where you are happy to reject the null hypothesis to (p-vale)/n.

• Remember that χ2 is the sum of at least two values.

• In each case state your P-value and exactly what the P-value you obtain means in terms of the null hypothesis. If the probability is smaller than 0.05 and also smaller than 0.01, then we say p < 0.01 as this indicates a stronger rejection of the null hypothesis.