Question

A high school principal recruits 18 instructors for teaching training. Each of the 18 individuals is randomly assigned to one of two groups (T= traditional methods, E=experimental methods). After the instructors spend 3 months on the job, the principal ranks them on the basis of their performance, from 1 (worst) to 18 (best), with the following results.

T: 2   3   5   9   10   12   13   14   15

E: 1   4   6   7   8   11   16   17    18  

The principal wants to test if there evidence of a difference in the median performance between the two methods. The best technique to use in this case to answer the research question is a                            [ Select ]                       ["parametric", "non-parametric"]           test.

The alternative hypothesis for the above test is                            [ Select ]                       ["p1=p2=p3", "mean 1 = mean 2", "D1 not equal D2"]      

The value of the statistics for a 2-sided test here would be T =                            [ Select ]                       ["63", "83"]         . Giving that the critical values from the appropriate tables is equal to                            [ Select ]                       ["62", "66"]           we can conclude that there                            [ Select ]                       ["ISN'T", "IS"]         statistical evidence to support the hypothesis that the medians are not equal.

Before the study ends, the researchers realizes that two individuals in the experimental group didn’t provide a written consent, therefore he cannot use their data. He also decides to run a non-parametric lower test at a 95% confidence level to see if one group performs better than the other. The new critical value from the appropriate table will be equal to                            [ Select ]                       ["66", "43", "72"]      

Answer 1