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: "non-parametric" OR  "parametric"   

Test The alternative hypothesis for the above test is

Select: "p1=p2=p3" OR "D1 not equal D2" OR "mean 1 = mean 2"

The value of the statistics for a 2-sided test here would be T =

Select: "83" OR  "63"

Giving that the critical values from the appropriate tables is equal to: Select: "62" OR "66"

we can conclude that there : select "IS" OR  "ISN'T" 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" OR  "43" OR "72"  

Answer 1

Trt A	Trt B	rank for sample 1	rank for sample 2
2	1	2	1
3	4	3	4
5	6	5	6
9	7	9	7
10	8	10	8
12	11	12	11
13	16	13	16
14	17	14	17
15	18	15	18

Trt A      
sample size ,    n1 =    9
sum of ranks ,    R1 =    83
      
Trt B      
sample size ,    n2 =    9
sum of ranks ,   R2 =    88
      
W=sum of ranks for smaller sample size =       83

=====================

1) non parametric

2)  D1 not equal D2

3) 83

4) critical value= 62

5) is not

6) 43