Scribd
Upload
61 views3 pages

BSC 311: Design and Analysis of Experiments ANOVA, Multiple Comparisons, Contrasts and Kruskal Wallis Test

This document contains instructions for tutorial 2 on analysis of variance (ANOVA) techniques. It includes 4 problems: 1) An experiment on goat diets with 5 groups and ANOVA to test for diet effects. 2) A consumer study rating 4 pillow brands using ANOVA and planned contrasts to compare brands. 3) A randomized experiment with 5 treatments and ANOVA to test differences between treatment means. 4) A non-parametric Kruskal-Wallis test and multiple comparisons to analyze 3 treatment groups.

Uploaded by

Victor Mlongoti
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
61 views3 pages

BSC 311: Design and Analysis of Experiments ANOVA, Multiple Comparisons, Contrasts and Kruskal Wallis Test

This document contains instructions for tutorial 2 on analysis of variance (ANOVA) techniques. It includes 4 problems: 1) An experiment on goat diets with 5 groups and ANOVA to test for diet effects. 2) A consumer study rating 4 pillow brands using ANOVA and planned contrasts to compare brands. 3) A randomized experiment with 5 treatments and ANOVA to test differences between treatment means. 4) A non-parametric Kruskal-Wallis test and multiple comparisons to analyze 3 treatment groups.

Uploaded by

Victor Mlongoti
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Due: 8th November, 2021 Tutorial 2 BSC 311

BSC 311: Design and Analysis of Experiments

ANOVA, Multiple comparisons, Contrasts and Kruskal Wallis test

1. An experiment was conducted to test the effects of different diets on goats. Thirty goats were
randomly assigned to five diet groups, six per group. The measured response is weight gain (in
kilograms).

Group Weight Gains


Control diet only 4.1 3.3 3.1 4.2 3.6 4.4
Control diet + level 1 of additive A 5.2 4.8 4.5 6.8 5.5 6.2
Control diet + level 2 of additive A 6.3 6.5 7.2 7.4 7.8 6.7
Control diet + level 1 of additive B 6.5 6.8 7.3 7.5 6.9 7.0
Control diet + level 2 of additive B 9.5 9.6 9.2 9.1 9.8 9.1

a) What can you conclude from the omnibus F test?


b) Define a set of four contrasts that compare each of the last four groups to the control group.
c) Define a set of contrasts to make the following comparisons:
i) Control diet versus the average of all others;
ii) Additive A versus additive B;
iii) Level 1 versus level 2 of additive A; and
iv) Level 1 versus level 2 of additive B.
d) Test of the contrasts in (c) are orthogonal or not.
e) Test the contrasts in (c) and determine those that are significant at the 0.05 level?

Jere, W. W. L. Page 1
Due: 8th November, 2021 Tutorial 2 BSC 311

2. Four different brands of pillows were tested by a panel of consumers. Each panelist examined
one pillow and rated it on a scale of 1 (inferior) to 7 (superior). The data are summarized below.

Pillow Brand Ratings


A 1 3 5 7 2 3 4
B 7 6 7 7 6
C 1 2 3 2 3 2 1
D 4 3 4 1 5

a) Generate the ANOVA table and perform the omnibus F-test.


b) Based on this study, the manufacturers of pillow B want to claim that their pillow is
significantly better than brand C. Test this contrast in the usual manner.
c) A skeptic may claim that the test procedure in part (b) is not valid because it is a post hoc
comparison; that is, the manufacturer of B decided to compare brand B to brand C because
B happened to have the highest average rating and C happened to have the lowest. Is this
criticism valid? Explain.

3. (modified from Oehlert 2010). Consider a completely randomized design with five treatments,
four units per treatment. The treatment means are 3.2892, 10.256, 8.1157, 8.1825 and 7.5622.
The MSE is 4.012

a) Construct an ANOVA table for this experiment and test the null hypothesis that all
treatments have the same mean.
b) Test the null hypothesis that the average response in treatments 1 and 2 is not significantly
different from the average response in treatments 3, 4, and 5.
c) Use the HSD procedure to compare the means of the five treatments.

Jere, W. W. L. Page 2
Due: 8th November, 2021 Tutorial 2 BSC 311

4. (Zar, 1999). Consider the data below:

Treatment 1 Treatment 2 Treatment 3


8 10 14
4 6 13
3 9 7
5 11 12
1 2 15

a) Conduct a non-parametric Kruskal-Wallis test on the data and interpret your findings.
b) Employ a nonparametric multiple range testing to determine between which of the three
treatments population differences exist.

Jere, W. W. L. Page 3

You might also like