In: Statistics and Probability
Professor Fair believes that extra time does not improve grades on exams. He randomly divided a group of 300 students into two groups and gave them all the same test. One group had exactly 1 hour in which to finish the test, and the other group could stay as long as desired. The results are shown in the following table. Test at the 0.01 level of significance that time to complete a test and test results are independent.
Time | A | B | C | F | Row Total |
1 h | 24 | 43 | 58 | 11 | 136 |
Unlimited | 19 | 48 | 81 | 16 | 164 |
Column Total | 43 | 91 | 139 | 27 | 300 |
(i) Give the value of the level of
significance.
State the null and alternate hypotheses.
H0: Time to take a test and test score are
independent.
H1: Time to take a test and test score are not
independent.
-
H0: The distributions for a timed test and an unlimited test are the same.
H1: The distributions for a timed test and an unlimited test are different.
-
H0: The distributions for a timed test and
an unlimited test are different.
H1: The distributions for a timed test and an
unlimited test are the same.
-
H0: Time to take a test and test score are
not independent.
H1: Time to take a test and test score are
independent.
(ii) Find the sample test statistic. (Round your answer to
two decimal places.)
(iii) Find or estimate the P-value of the sample test
statistic.
P-value > 0.100
0.050 < P-value < 0.100
0.025 < P-value < 0.050
0.010 < P-value < 0.025
0.005 < P-value < 0.010
P-value < 0.005
(iv) Conclude the test.
Since the P-value < α, we reject the null hypothesis.
Since the P-value is ≥ α, we do not reject the null hypothesis.
Since the P-value ≥ α, we reject the null hypothesis.
Since the P-value < α, we do not reject the null hypothesis.
(v) Interpret the conclusion in the context of the
application.
At the 1% level of significance, there is insufficient evidence to claim that time to do a test and test results are not independent.
At the 1% level of significance, there is sufficient evidence to claim that time to do a test and test results are not independent.
Here we are using Excel for calculation:
Anova: Single Factor | ||||||
SUMMARY | ||||||
Groups | Count | Sum | Average | Variance | ||
Column 1 | 2 | 43 | 21.5 | 12.5 | ||
Column 2 | 2 | 82 | 41 | 98 | ||
Column 3 | 2 | 139 | 69.5 | 264.5 | ||
Column 4 | 2 | 27 | 13.5 | 12.5 | ||
ANOVA | ||||||
Source of Variation | SS | df | MS | F | P-value | F crit |
Between Groups | 3726 | 3 | 1242.13 | 12.82 | 0.0161 | 16.694 |
Within Groups | 387.5 | 4 | 96.875 | |||
Total | 4114 | 7 |
i) State the null and alternate hypotheses.
H0: The distributions for a timed test and an unlimited test are the same.
H1: The distributions for a timed test and an unlimited test are different.
ii) Test statistics:
F = 12.82
iii) Find or estimate the P-value of the sample test statistic.
0.010 < P-value < 0.025
iv) Conclude the test.
Since the P-value is ≥ α, we do not reject the null hypothesis.
(v) Interpret the conclusion in the context of the application.
At the 1% level of significance, there is insufficient evidence to claim that time to do a test and test results are not independent.