Homework SourseThe following data represent the breaking strengths of sampl
The following data represent the breaking strengths of samples of 20 gauge insulated copper wire from three different suppliers. The researcher is interested in finding evidence that the distribution of breaking strength differs by supplier. Use the .10 significance level and the Kruskal-Wallis test to answer the following questions.
Supplier
A
B
C
140
156
123
144
164
127
145
166
144
151
169
154
152
180
155
Give the value of the Kruskal-Wallis test statistic for the above data.
a.
1.836
b.
7.265
c.
8.045
d.
9.395
e.
5.435
Which of the following correctly describes the p-value?
a.
p-value < .01
b.
.01 < p-value < .025
c.
.025 < p-value < .05
d.
.05 < p-value < .10
e.
p-value > .10
What is the correct critical value?
a.
7.378
b.
5.991
c.
4.605
d.
6.93
e.
9.210
What is the correct decision?
a.
Reject H0
b.
Do Not Reject H0
c.
Reject HA
d.
Do Not Reject HA
e.
Leprechaun
If the null hypothesis of identical distributions is in fact true, what would be the expected (average) value of the test statistic for this experiment?
a.
3
b.
1
c.
0
d.
2
e.
15
| Supplier | ||
| A | B | C |
| 140 | 156 | 123 |
| 144 | 164 | 127 |
| 145 | 166 | 144 |
| 151 | 169 | 154 |
| 152 | 180 | 155 |
Solution
Give the value of the Kruskal-Wallis test statistic for the above data.
Answer: d. 9.395
--------------------------------------------------------------------------------------------------------------------
Which of the following correctly describes the p-value?
Answer: a. p-value < .01
--------------------------------------------------------------------------------------------------------------------
What is the correct critical value?
Answer: b. 5.991
--------------------------------------------------------------------------------------------------------------------
What is the correct decision?
Answer: a .Reject H0
--------------------------------------------------------------------------------------------------------------------
If the null hypothesis of identical distributions is in fact true, what would be the expected (average) value of the test statistic for this experiment?
Answer: c .0
| Median | n | Avg. Rank | |
| 145.00 | 5 | 5.70 | A |
| 166.00 | 5 | 13.00 | B |
| 144.00 | 5 | 5.30 | C |
| 152.00 | 15 | Total |
