7. Which multiple comparison method to use Suppose you have conducted a study in which you have five random samples from five populations. Assume that each of the populations are normally distributed with the same variance. You collect samples of sizes 16, 22, 21, 18, and 18 from each of the five populations, respectively. You carefully consider your research objectives prior to performing any hypothesis tests. You decide you want to compare the means of every pair of populations. Given your research objectives, which of the following is your best strategy? Use Tukey's HSD test to compare the means of all pairs of populations. Perform an analysis of variance. If it is significant, use Fisher's LSD method with the Bonferroni adjustment to compare the means of all pairs of populations. Perform 10 t-tests for independent samples among all pairs of populations. Perform an analysis of variance. If it is significant, use Fisher's LSD method (no adjustment necessary) to compare the means of all pairs of populations. Suppose you plan to use Tukey's HSD test. You should use ng = as the number of observations in each of the samples