Thanks to visit codestin.com
Credit goes to www.scribd.com

Scribd
Upload
12 views3 pages

Assignment 1

The document presents an experimental analysis involving four diets and their effects on bleeding time in rabbits, concluding that significant differences exist among the diets. It also includes a study on birth weights of calves from five sires, with results indicating significant differences among sires and an estimated intraclass correlation coefficient. Statistical models and tests, including ANOVA and multiple comparisons, are utilized to analyze the data.

Uploaded by

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

Assignment 1

The document presents an experimental analysis involving four diets and their effects on bleeding time in rabbits, concluding that significant differences exist among the diets. It also includes a study on birth weights of calves from five sires, with results indicating significant differences among sires and an estimated intraclass correlation coefficient. Statistical models and tests, including ANOVA and multiple comparisons, are utilized to analyze the data.

Uploaded by

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

Experimental Analysis

HW #1

9/19/2014

1. There are 4 different diets, A, B, C, D and 16 rabbits available. We are interested in how the diets
affect the time (in seconds) that it takes for a cut to stop bleeding. We randomly assign each
rabbit to one of the four diets and let it live in its own cage. The results are given below:

Diet A Diet B Diet C Diet D


63 64 67 56
60 68 66 59
62 70 72 61
59 63 68 62
a) Write the linear statistical model for this study and explain the model components
y ij =μi +e ij
e ij N ¿) normal dist.
Where [i] is the number of observations per treatment (4 rabbits) and [j]
is the number of treatments (4 diets). [E] is the error term, [y] is the
outcome, and [ μi] is the differences between the means. A normal
distribution is assumed for the error term.
b) Compute the analysis of variance table

c) Test the hypothesis of no difference among diets on the time that it takes for a cut to stop
bleeding, using F test at the 0.05 level of significance.

We reject the null hypothesis. There is a significant difference among the diets.
2. Using the data in question 1 above, control family-wise error rate at 0.05 for the following
situations and show which comparison is significant after adjusting for multiple testing.
a) compare diet B, diet C and diet D with diet A
Using Dunnett’s test, we get the following results which show that only diet’s B and C
significantly differ from Diet A.

b) Pairwise comparison of all diets A, B, C and D. (tukey)


Using Tukey’s test, we look at all diets and attain the following results, which show that
Diets C-D, C-A, and B-A are significantly different from each other. These results are
inclusive of what we found in part A, which makes sense.
3. There are 5 sires, each mated to a separate group of dams. The birth weights of eight male
calves in each of five sire group are given below:

Sire Birth weights


1 62, 99,57, 112,98,104,76,61
2 75,102,96,103,97,116,98,93
3 58,61,59,57,58,59,53,100
4 57,55,68,58,58,121,101,102
5 58,47,119,116,115,93,104,76

a) Assume a random model for this study, Write the linear statistical model for this study and
explain the model components.
y ij =μ+a i+ ε ij
a i N ¿)
ε ij N ¿)
Where [i] is the number of observations per treatment (8 calves) and [j]
is the number of treatments (5 sires). [E] is the error term, [y] is the
outcome, and [ μ+a i] is the differences between the means. Because the
study is assumed to be a random model, a normal distribution is
assumed.
b) Compute the analysis of variance table

2
c) Test the null hypothesis H 0 :σ a =0 for the sires

Reject the null hypothesis (p < 0.05)


d) Estimate the intraclass correlation coefficient (ICC).

ICC = 117.04 / (117.04 + 461.51) = 0.202

You might also like