In: Statistics and Probability
1.Interpret the following results of analysis of variance. Interpret each statistic including post-hoc comparisons. Provide a complete interpretation using the five-step model of hypothesis testing.
Oneway: Relationship Between Lab Section and Score on Stat Final
Descriptives
Score on Stat Final
N |
Mean |
Std. Deviation |
Std. Error |
95% Confidence Interval for Mean |
Minimum |
Maximum |
||
Lower Bound |
Upper Bound |
|||||||
1 |
22 |
40.68 |
13.892 |
2.962 |
34.52 |
46.84 |
14 |
65 |
2 |
22 |
32.59 |
12.015 |
2.562 |
27.26 |
37.92 |
7 |
55 |
3 |
24 |
29.83 |
12.910 |
2.635 |
24.38 |
35.28 |
4 |
55 |
Total |
68 |
34.24 |
13.580 |
1.647 |
30.95 |
37.52 |
4 |
65 |
ANOVA
Score on Stat Final
Sum of Squares |
df |
Mean Square |
F |
Sig. |
|
Between Groups |
1438.811 |
2 |
719.406 |
4.283 |
.018 |
Within Groups |
10917.424 |
65 |
167.960 |
||
Total |
12356.235 |
67 |
Post Hoc Tests
Multiple Comparisons
Dependent Variable: Score on Stat Final
Scheffe
(I) Lab Section |
(J) Lab Section |
Mean Difference (I-J) |
Std. Error |
Sig. |
95% Confidence Interval |
|
Lower Bound |
Upper Bound |
|||||
1 |
2 |
8.09 |
3.908 |
.125 |
-1.70 |
17.88 |
3 |
10.85(*) |
3.825 |
.023 |
1.27 |
20.43 |
|
2 |
1 |
-8.09 |
3.908 |
.125 |
-17.88 |
1.70 |
3 |
2.76 |
3.825 |
.772 |
-6.83 |
12.34 |
|
3 |
1 |
-10.85(*) |
3.825 |
.023 |
-20.43 |
-1.27 |
2 |
-2.76 |
3.825 |
.772 |
-12.34 |
6.83 |
* The mean difference is significant at the .05 level.
Homogeneous Subsets
Score on Stat Final
Scheffe
Lab Section |
N |
Subset for alpha = .05 |
|
1 |
2 |
||
3 |
24 |
29.83 |
|
2 |
22 |
32.59 |
32.59 |
1 |
22 |
40.68 |
|
Sig. |
.775 |
.118 |
Means for groups in homogeneous subsets are displayed.
a Uses Harmonic Mean Sample Size = 22.629.
b The group sizes are unequal. The harmonic mean of the group sizes is used. Type I error levels are not guaranteed.
The five step model of hypothesis testing is describe below,
Step 1: The null hypothesis for the test is define as,
Null hypothesis: such that group means are all equal.
Step 2: The alternative hypothesis for the test is define as,
Alternative hypothesis, : such that not all group means are equal.
Step 3: The significance level for the test, is the maximum acceptance level at 5% risk which means null hypothesis will be rejected beyond this level ( such that if P-value is less than 0.05 null hypothesis will be rejected)
Step 4: The test statistic is the F-value which is obtained in ANOVA table, The F-value is used to calculate the P-value. (Whereas F-value determine whether the means among two or more populations are significantly different if compare against the F-critical value). Here the P-value is used to make a decision about the statistical significance of the means. From the ANOVA model summary,
Step 5: The decision for the statistical significance in mean is determine by the P-value. The P-value can be define as the probability that measures the evidence against the null hypothesis If the P-value < alpha: The differences between some of the means are statistically significant and if P-value > alpha: The differences between the means are not statistically significant. from the model summary,
Hence the differences between some of the means are statistically significant.
Post hoc test
If the null hypothesis is rejected, ANOVA only tell there is atleast one group statistically different. ANOVA doesn't compare means individually hence the post hoc test is reqired to see which means are statistically different
The Scheffe test for multiple group comparison is used here. From the test result summary, The P-value
(I) Lab Section | (J) Lab Section | Mean Difference (I-J) | Sig. | |||
Significance Level | ||||||
1 | 2 | 8.09 | 0.125 | > | 0.05 | Not Significant |
3 | 10.85(*) | 0.023 | < | 0.05 | Significant | |
2 | 1 | -8.09 | 0.125 | > | 0.05 | Not Significant |
3 | 2.76 | 3.825 | -6.83 | > | 0.05 | Not Significant |
3 | 1 | -10.85(*) | 0.023 | < | 0.05 | Significant |
2 | -2.76 | 3.825 | -12.34 | > | 0.05 | Not Significant |
The P-value for group comparison 1 and 3 (or say 3 and 1, same thing) is significant at 5% significance level.