The growth per day (in cm) of alfalfa sprouts was recorded for 50 sprouts kept in darkness and for 49 sprouts kept in light. The log of each observation was then taken to make the distributions reasonably normal. For the darkness sample, the average was 1.40, and the SD was 0.92. For the light sample, the average was 0.47, and the SD was 0.45.
George has all of the raw data and wants to test the null hypothesis that the two population means are equal. He wants to use ANOVA to do this test. Is this a good idea? Discuss (i) whether this is possible (saying why or why not, and if not, saying what you would do instead of ANOVA) and (ii) whether this is wise (saying why or why not, and if not, saying what you would do instead of ANOVA).