1 Suppose you collected data from two different and independ

1. Suppose you collected data from two different and independent groups on some

continuous variable (height, weight etc) and now you want to test for a difference

between the two groups. For example, does the height of individuals in group A differ

from that of individuals in group B? However, when you plotted the data and looked at

the distributions you saw that there were some severe outliers. Of the tests listed below,

what is the most appropriate test that you should use if you have severe outliers?

Your choices

a) t-test, b) Mann-Whitney U test, c) Wilcoxon signed-rank test,

d) paired t-test

2. Suppose you collected data on the distance between the two eyes of insects
(eyespan) from two different and independent groups of individuals raised on corn and
cotton diets (see frequency distributions below). Yes, people do look at this stuff. Males with longer eyespans are apparently more attractive to
females, and so researchers have tested whether environmental conditions during growth,
particularly the diet, affects their adult eyespan. So the distributions below are based on
the eyespans for each individual. Suppose you want to test the difference between the
means of the two groups. Of the tests listed below, what is the most appropriate test to
use with these data and why?

Your choices
a) t-test because the variances are the same and

there are over 30 individuals in each group

b) Welch’s t-test because the variances look

different

c) Mann-Whitney U test because there are

severe outliers in both groups

d)Wilcoxon signed-rank test because the data are pair of eyespans, and thus the data have to be paired, rather than independent

Solution

Note: we are suppose to answer 1 question at a time and u have asked more than 1, so i will give u solution of 1st for rest repost again.

1) As we need to compare the data in to different sample so the best test will be paired t-test. A paired t-test is used to compare two population means where you have two samples in which observations in one sample can be pairedwith observations in the other sample