4.5.2 Inferential Statistics Levene's test an equal variance test-was initially employed to determine if our data sets fulfil the homogeneity of variance assumption before performing Analysis of Variance (ANOVA). The null hypothesis (Ho) and alternative hypothesis (H) are as follows: • Ho: Concentration groups have equal variances. H₁: Concentration groups have different variances. The p-values for both S. aureus and E. coli data sets are less that 0.05, indicating there is a significant difference between the variances, thus null hypothesis is rejected (Table 4.8). Since the homogeneity assumption of the variance is not violated, non-parametric test like Kruskal-Wallis, which is an alternative to one-way ANOVA, test should be used. Table 4.8: Test for equality of variances (Levene's) Bacteria F dfl df2 S. aureus 9.600 5.000 12.000 Р <.001 E. coli 6.000 5.000 12.000 < .001 Next, Kruskal-Wallis test determines whether the medians of two or more groups are different, and the hypothesis are as follows: • Ho: There is no significant difference in the median of the concentration groups. H₁: At least one concentration group has a different median compared to the other groups. If the p-value is larger than 0.05, the null hypothesis is retained, otherwise it is rejected. In this case, the p-values for both S. aureus and E. coli are 0.009 and 0.006, respectively, which are less than 0.05 (Table 4.9). The null hypothesis is thus rejected, and it is assumed that there is a significant difference between the different groups relative to its concentrations. Bacteria Factor S. aureus Concentration E. coli Concentration Table 4.9: Kruskal-Wallis test Statistic df P 15.264 5 0.009 16.399 5 0.006 The Kruskal-Wallis test does not provide an answer to the question of which of the concentration groups differed for both cultures; thus Dunn's post-hoc test is required to further address this issue for the pairwise multiple comparison. As usual, if the p-value in a pairwise comparison is less than the significance level (i.e., 0.05), the null hypothesis that there is no difference is thus rejected, and it is assumed that the respective two groups differ. For S. aureus, there is a significant difference when increasing the concentration from 50 mg/ml to 200 mg/ml, 250 mg/ml, 300 mg/ml or 400 mg/ml as well as from 100 mg/ml to 400 mg/ml (Table 4.10). On the other hand, for E. coli, increasing the concentration from 50 mg/ml to 250 mg/ml, 300 mg/ml or 400 mg/ml as well as from 100 mg/ml to 300 mg/ml or 400 mg/ml mark a significant difference (Table 4.11). Note that the p-value is unadjusted, acknowledging the increased risk of Type I errors. However, if Bonferroni adjustment is applied, the significant comparison is only seen when increasing the concentration from 50 mg/ml to 400 mg/ml for both cultures (Table 4.10 and 4.11). Table 4.10: Dunn's post hoc comparisons for S. aureus - concentration Comparison Ꮓ Wi Wj р Pbonf 50 mg/ml 100 mg/ml -1.066 2.000 6.333 0.286 1.000 50 mg/ml - 200 mg/ml -1.969 2.000 10.000 0.049 * 0.735 50 mg/ml - 250 mg/ml -1.969 2.000 10.000 0.049 * 0.735 50 mg/ml - 300 mg/ml -2.420 2.000 50 mg/ml -400 mg/ml -3.651 2.000 100 mg/ml 200 mg/ml 100 mg/ml 250 mg/ml 100 mg/ml 300 mg/ml -0.902 6.333 -0.902 6.333 -1.354 6.333 16.833 10.000 0.367 11.833 0.016 <.001 * 0.233 *** 0.004 ** 1.000 10.000 0.367 1.000 11.833 0.176 1,000

Glencoe Algebra 1, Student Edition, 9780079039897, 0079039898, 2018
18th Edition
ISBN:9780079039897
Author:Carter
Publisher:Carter
Chapter10: Statistics
Section10.4: Distributions Of Data
Problem 19PFA
icon
Related questions
Question
4.5.2 Inferential Statistics
Levene's test an equal variance test-was initially employed to determine if our data
sets fulfil the homogeneity of variance assumption before performing Analysis of Variance
(ANOVA). The null hypothesis (Ho) and alternative hypothesis (H) are as follows:
•
Ho: Concentration groups have equal variances.
H₁: Concentration groups have different variances.
The p-values for both S. aureus and E. coli data sets are less that 0.05, indicating there
is a significant difference between the variances, thus null hypothesis is rejected (Table 4.8).
Since the homogeneity assumption of the variance is not violated, non-parametric test like
Kruskal-Wallis, which is an alternative to one-way ANOVA, test should be used.
Table 4.8: Test for equality of variances (Levene's)
Bacteria
F
dfl
df2
S. aureus
9.600
5.000
12.000
Р
<.001
E. coli
6.000
5.000
12.000
< .001
Next, Kruskal-Wallis test determines whether the medians of two or more groups are
different, and the hypothesis are as follows:
•
Ho: There is no significant difference in the median of the concentration groups.
H₁: At least one concentration group has a different median compared to the other
groups.
If the p-value is larger than 0.05, the null hypothesis is retained, otherwise it is
rejected. In this case, the p-values for both S. aureus and E. coli are 0.009 and 0.006,
respectively, which are less than 0.05 (Table 4.9). The null hypothesis is thus rejected, and it
is assumed that there is a significant difference between the different groups relative to its
concentrations.
Transcribed Image Text:4.5.2 Inferential Statistics Levene's test an equal variance test-was initially employed to determine if our data sets fulfil the homogeneity of variance assumption before performing Analysis of Variance (ANOVA). The null hypothesis (Ho) and alternative hypothesis (H) are as follows: • Ho: Concentration groups have equal variances. H₁: Concentration groups have different variances. The p-values for both S. aureus and E. coli data sets are less that 0.05, indicating there is a significant difference between the variances, thus null hypothesis is rejected (Table 4.8). Since the homogeneity assumption of the variance is not violated, non-parametric test like Kruskal-Wallis, which is an alternative to one-way ANOVA, test should be used. Table 4.8: Test for equality of variances (Levene's) Bacteria F dfl df2 S. aureus 9.600 5.000 12.000 Р <.001 E. coli 6.000 5.000 12.000 < .001 Next, Kruskal-Wallis test determines whether the medians of two or more groups are different, and the hypothesis are as follows: • Ho: There is no significant difference in the median of the concentration groups. H₁: At least one concentration group has a different median compared to the other groups. If the p-value is larger than 0.05, the null hypothesis is retained, otherwise it is rejected. In this case, the p-values for both S. aureus and E. coli are 0.009 and 0.006, respectively, which are less than 0.05 (Table 4.9). The null hypothesis is thus rejected, and it is assumed that there is a significant difference between the different groups relative to its concentrations.
Bacteria
Factor
S. aureus
Concentration
E. coli
Concentration
Table 4.9: Kruskal-Wallis test
Statistic
df
P
15.264
5
0.009
16.399
5
0.006
The Kruskal-Wallis test does not provide an answer to the question of which of the
concentration groups differed for both cultures; thus Dunn's post-hoc test is required to further
address this issue for the pairwise multiple comparison. As usual, if the p-value in a pairwise
comparison is less than the significance level (i.e., 0.05), the null hypothesis that there is no
difference is thus rejected, and it is assumed that the respective two groups differ. For S. aureus,
there is a significant difference when increasing the concentration from 50 mg/ml to 200
mg/ml, 250 mg/ml, 300 mg/ml or 400 mg/ml as well as from 100 mg/ml to 400 mg/ml (Table
4.10). On the other hand, for E. coli, increasing the concentration from 50 mg/ml to 250 mg/ml,
300 mg/ml or 400 mg/ml as well as from 100 mg/ml to 300 mg/ml or 400 mg/ml mark a
significant difference (Table 4.11). Note that the p-value is unadjusted, acknowledging the
increased risk of Type I errors. However, if Bonferroni adjustment is applied, the significant
comparison is only seen when increasing the concentration from 50 mg/ml to 400 mg/ml for
both cultures (Table 4.10 and 4.11).
Table 4.10: Dunn's post hoc comparisons for S. aureus - concentration
Comparison
Ꮓ
Wi
Wj
р
Pbonf
50 mg/ml 100 mg/ml
-1.066
2.000
6.333
0.286
1.000
50 mg/ml - 200 mg/ml
-1.969
2.000
10.000
0.049
*
0.735
50 mg/ml - 250 mg/ml
-1.969
2.000
10.000
0.049
*
0.735
50 mg/ml - 300 mg/ml
-2.420
2.000
50 mg/ml -400 mg/ml
-3.651 2.000
100 mg/ml 200 mg/ml
100 mg/ml 250 mg/ml
100 mg/ml 300 mg/ml
-0.902 6.333
-0.902
6.333
-1.354 6.333
16.833
10.000 0.367
11.833 0.016
<.001
*
0.233
*** 0.004
**
1.000
10.000
0.367
1.000
11.833
0.176
1,000
Transcribed Image Text:Bacteria Factor S. aureus Concentration E. coli Concentration Table 4.9: Kruskal-Wallis test Statistic df P 15.264 5 0.009 16.399 5 0.006 The Kruskal-Wallis test does not provide an answer to the question of which of the concentration groups differed for both cultures; thus Dunn's post-hoc test is required to further address this issue for the pairwise multiple comparison. As usual, if the p-value in a pairwise comparison is less than the significance level (i.e., 0.05), the null hypothesis that there is no difference is thus rejected, and it is assumed that the respective two groups differ. For S. aureus, there is a significant difference when increasing the concentration from 50 mg/ml to 200 mg/ml, 250 mg/ml, 300 mg/ml or 400 mg/ml as well as from 100 mg/ml to 400 mg/ml (Table 4.10). On the other hand, for E. coli, increasing the concentration from 50 mg/ml to 250 mg/ml, 300 mg/ml or 400 mg/ml as well as from 100 mg/ml to 300 mg/ml or 400 mg/ml mark a significant difference (Table 4.11). Note that the p-value is unadjusted, acknowledging the increased risk of Type I errors. However, if Bonferroni adjustment is applied, the significant comparison is only seen when increasing the concentration from 50 mg/ml to 400 mg/ml for both cultures (Table 4.10 and 4.11). Table 4.10: Dunn's post hoc comparisons for S. aureus - concentration Comparison Ꮓ Wi Wj р Pbonf 50 mg/ml 100 mg/ml -1.066 2.000 6.333 0.286 1.000 50 mg/ml - 200 mg/ml -1.969 2.000 10.000 0.049 * 0.735 50 mg/ml - 250 mg/ml -1.969 2.000 10.000 0.049 * 0.735 50 mg/ml - 300 mg/ml -2.420 2.000 50 mg/ml -400 mg/ml -3.651 2.000 100 mg/ml 200 mg/ml 100 mg/ml 250 mg/ml 100 mg/ml 300 mg/ml -0.902 6.333 -0.902 6.333 -1.354 6.333 16.833 10.000 0.367 11.833 0.016 <.001 * 0.233 *** 0.004 ** 1.000 10.000 0.367 1.000 11.833 0.176 1,000
Expert Solution
steps

Step by step

Solved in 2 steps

Blurred answer
Recommended textbooks for you
Glencoe Algebra 1, Student Edition, 9780079039897…
Glencoe Algebra 1, Student Edition, 9780079039897…
Algebra
ISBN:
9780079039897
Author:
Carter
Publisher:
McGraw Hill
College Algebra (MindTap Course List)
College Algebra (MindTap Course List)
Algebra
ISBN:
9781305652231
Author:
R. David Gustafson, Jeff Hughes
Publisher:
Cengage Learning