Class: DH19TP (Advanced program)

Course: Applied Statistics & Design of Experiments

Name: Nguyễn Thị Ngọc Huyền

ID: 19125516

HOMEWORK ASSIGNMENT # 3
Question # 1: Table 1 & 2 are results of one-way ANOVA analysis. Find the missing values in ANOVA

Table 1 & 2 at question marks “?”

Table 1:

Source df SS MS = SS/df F – statistic

Treatment 3 2.124 0.708 0.75

Error 20 18.88 0.944

Total 23 21.004

𝑆𝑆𝑇𝑅 2.124
𝑀𝑆𝑇𝑅 = => 0.708 = => 𝑘 = 4
𝑘−1 𝑘−1

𝑀𝑆𝑇𝑅 𝑀𝑆𝑇𝑅 0.708

𝐹= => 𝑀𝑆𝐸 = = = 0.944
𝑀𝑆𝐸 𝐹 0.75

𝑛 − 𝑘 = 20 => 𝑛 = 24

𝑆𝑆𝐸
𝑀𝑆𝐸 = => 𝑆𝑆𝐸 = 18.88
𝑛−𝑘

Table 2:

Source df SS MS = SS/df F – statistic

Treatment 4 37.84 9.46 1.399

Error 20 135.2 6.76

Total 24 173.04

Treatment: df = 𝑘 − 1 = 4 => 𝑘 = 5

Error: 𝑛 − 𝑘 = 20 => 𝑛 = 25

𝑆𝑆𝐸
𝑀𝑆𝐸 = => 𝑆𝑆𝐸 = 135.2
𝑛−𝑘
SSE = SST – SSTR => SSTR = 173.04 – 135.2 = 37.84

𝑆𝑆𝑇𝑅 37.84
MSTR = = = 9.46
𝑘−1 5−1

Question # 2

Staphylococcus aureus is causing staph Infections. A researcher conducted a study on culture five strains
of Staphylococcus aureus in for 24 hours at 27◦C. The following table reported bacteria counts (CFU)
for different cases from each of the five strains.

Questions:

a) At the 5% significance level, do the data provide sufficient evidence to conclude that a difference
exists in mean bacteria counts among the five strains of S. aureus?

b) Use Table and Chart to present results of the study

Solution

Hypothesis of this research: a study on culture five strains of Staphylococcus aureus in for 24 hours at
27◦C

Type of experiment: one- factor experiment with CRD design

Response variable: culture five strains of Staphylococcus aureus

Name of factor: strains Staphylococcus aureus.

Number of factors: one factor

Number of treatment: 5 treatments (t=5) or 5 level of rates between content of strains Staphylococcus
Aureus (Strain A, Strain B, Strain C, Strain D, Strain E)

Number of replicates: five replicate in each treatment (r=5)

Number of experimental units: 25 experomental units (n = r x t = 5 x 5 = 25)

a.
- The null and alternative hypotheses are, respectively

𝐻0 : 𝜇1 = 𝜇2 = 𝜇3 = 𝜇4 = 𝜇5

Ha: Not all the means are equal

 - The value of 𝛼 = 0.05
- Compute the value of the test statistic

Analysis of Variance
Source DF Adj SS Adj MS F-Value P-Value
Strain 4 0.4325 0.10813 1.33 0.294
Error 20 1.6288 0.08144
Total 24 2.0613
- The F-statistic has df = (4;20). Use table A-4, we have the value of critical value is 2.87
- Here the test statistic < critical value
- It usually dose not reject Ho

At the 5% level, the data on bacteria number among the 5 distinct staphylococcus aureus strains does
not provide enough information to draw the conclusion that there is a difference in mean bacteria number
among the 5 different staphylococcus aureus strains.

b. Table and Chart to present results of the study

Result of ANOVA table

Analysis of Variance
Source DF Adj SS Adj MS F-Value P-Value
Strain 4 0.4325 0.10813 1.33 0.294
Error 20 1.6288 0.08144
Total 24 2.0613

Boxplot of LOG10(CFU)

7.75

7.50

7.25
LOG10(CFU)

7.00

6.75

6.50

A B C D E
Strain

Tukey Pairwise Comparisons

Grouping Information Using the Tukey Method and 95% Confidence

Strain N Mean Grouping

E 5 7.5723 A
C 5 7.500 A
A 5 7.2819 A
D 5 7.2791 A
B 5 7.257 A

Means that do not share a letter are significantly different.

Question # 3: In a study of heat resistance of potato varieties, six plantlets of four varieties of potatoes
were randomly assigned to each of four temperature regimes. Weights of tubers were recorded after
45 days. The resulting experiment is a 4 × 4 factorial. The data are shown in Table 9.30. Perform the
analysis to determine the nature of differences in heat resistance among the varieties. Make
recommendations indicated by the results.
Solution
Type of experiment: A two factor factorial experiment with CRD design

Response variable: heat resistance of potato

Number of factor: Varieties of potato (BUR, KEN, NOR, RLS) and Temperature regimes ( 15, 20, 25,30)

Number of treatment:16 treatments (4 varieties of potato, 4 temperature regimes)

Number of experiments unit: 96 unit (16 treatment x 6 replicates)

Result of ANOVA analysis

Factor Information
Factor Type Levels Values
Temperature Fixed 4 15, 20, 25, 30
Varieties of potato Fixed 4 BUR, KEN, NOR, RLS

Analysis of Variance
Source DF Adj SS Adj MS F-Value P-Value
Temperature 3 1.1841 0.39469 13.64 0.000
Varieties of potato 3 0.2923 0.09744 3.37 0.023
Temperature*Varieties of potato 9 0.3283 0.03648 1.26 0.271
Error 80 2.3141 0.02893
Total 95 4.1188

Model Summary
S R-sq R-sq(adj) R-sq(pred)
0.170076 43.82% 33.28% 19.10%

Fits and Diagnostics for Unusual Observations

Obs Weights Fit Resid Std Resid
21 0.7000 0.2833 0.4167 2.68 R
45 0.0000 0.4733 -0.4733 -3.05 R
47 1.2500 0.4733 0.7767 5.00 R
54 0.4100 0.0917 0.3183 2.05 R
63 0.0000 0.3850 -0.3850 -2.48 R

R Large residual

Result on multiple comparisons

Tukey Pairwise Comparisons: Response = Weights, Term = Temperature

Grouping Information Using the Tukey Method and 95% Confidence
Temperature N Mean Grouping
20 24 0.354167 A
15 24 0.289583 A
25 24 0.269583 A
30 24 0.058333 B
Means that do not share a letter are significantly different.

Tukey Pairwise Comparisons: Response = Weights, Term = Varieties of potato

Grouping Information Using the Tukey Method and 95% Confidence
Varieties
of potato N Mean Grouping
NOR 24 0.277917 A
KEN 24 0.276667 A
RLS 24 0.269583 A B
BUR 24 0.147500 B

Means that do not share a letter are significantly different.

Tukey Pairwise Comparisons: Response = Weights, Term = Temperature*Varieties of

potato
Grouping Information Using the Tukey Method and 95% Confidence
Temperature*Varieties
of potato N Mean Grouping
20 RLS 6 0.473333 A
15 KEN 6 0.391667 A B
25 NOR 6 0.385000 A B
25 KEN 6 0.348333 A B C
15 NOR 6 0.330000 A B C
20 BUR 6 0.325000 A B C
20 NOR 6 0.313333 A B C
20 KEN 6 0.305000 A B C
15 RLS 6 0.283333 A B C
25 RLS 6 0.253333 A B C
15 BUR 6 0.153333 A B C
25 BUR 6 0.091667 B C
30 NOR 6 0.083333 B C
30 RLS 6 0.068333 B C
30 KEN 6 0.061667 B C
30 BUR 6 0.020000 C

Means that do not share a letter are significantly different.

Question 4:
Number of “two-factor factorial” treatment: 2𝑘 = 24 = 16

Number of “Star” treatment: 2𝑘 = 2 × 4 = 8

Number “Center” treatment: n = 7

Total number of treatment: 2𝑘 + 2𝑘 + 1 = 16 + 8 + 1 = 25

Total experimental units (or run): 2𝑘 + 2𝑘 + 𝑛 = 16 + 8 + 7 = 31

Coded
Uncoded
-2 -1 0 +1 +2
Lighting intensity 100 200 300 400 500
Cultivation 2.5 10 17.5 25 32.5
Salinity 2 3 4 5 6
pH 2.5 5 7.5 10 12.5