Borbon, Klied Erica B.

Largo, Vyankha Jhouana L.

One-Way ANOVA
1. Three sections of the same elementary mathematics courses are
taught by three teachers. The final grades were recorded as
follows:

Teacher
A B C
73 88 68
89 78 79
82 48 56
43 91 91
80 51 71
73 85 71
66 74 87
60 77 41
45 31 59
93 78 68
36 62 53
77 76 79
96 15
80
56

Test for homogeneity of variances. Is there a significant

difference in the average grades given by the three teachers?
Use a 0.05 level of significance.
Interpretation:

Residual standard Bartlett's K-

Deg. of Freedom p-value
error squared

19.23782 0.9171
0.1732 2

Sum of Squares Deg. of Freedom

A 337.598 2
Residuals 13323.376 36

ANOVA: Two – way Classification

2. The following data represent the final grades obtained by

five students in Mathematics, English, French, and Biology.

Subject
Student Mathematics English French Biology
1 68 57 73 61
2 83 94 91 86
3 72 81 63 59
4 55 73 77 66
5 92 68 75 87
Use a 0.05 level of significance to test the hypothesis that

(a) the courses are of equal difficulty;

(b) the students have equal ability;
(c) Use the multiple – range test, with a 0.05 level of
significance, to analyze the mean grades obtained by the five
students.

Interpratation:

Sum of Squares Deg. of Freedom

Subject 42.150 3

Students 46.225 1

Subject: Students 106.275 3

Residuals 2578.300 12

Residual standard error: 14.65805

 ANOVA: Two – way Classification with Interaction
3. The following data represent the results of four quizzes
obtained by five students in Mathematics, English, French, and
Biology:
Student Mathematics English French Biology
1 88 63 51 58 73 81 87 81
79 80 72 65 77 77 92 76
2 79 96 85 95 82 36 80 93
56 68 67 88 80 68 62 67
3 67 66 74 47 91 95 77 70
51 89 59 82 59 92 84 73
4 35 60 76 49 43 52 55 49
64 70 26 76 42 32 53 56
5 99 77 84 94 95 81 83 76
87 95 83 76 98 96 87 80

Use a 0.05 level of significance to test the hypothesis that

(a) The courses are of equal difficulty;
(b) The students have equal ability;
(c) The students and subjects do not interact.

Interpretation:

Degrees of Sum of Mean

F-Value Pr(>F)
freedom squares squares
Subject 3 158 52.65 0.167 0.918

Student 1 2 2.03 0.006 0.936

Subject:
3 467 155.68 0.495 0.687
Students
Residuals 72 22634 314.37
Appendix:

Part 1.
1. Homogeneity of Variances:
> t<-read.csv("D:/aaaaa.csv",header=T,sep="," )
> t
A X73
1 A 89
2 A 82
3 A 43
4 A 80
5 A 73
6 A 66
7 A 60
8 A 45
9 A 93
10 A 36
11 A 77
12 B 88
13 B 78
14 B 48
15 B 91
16 B 51
17 B 85
18 B 74
19 B 77
20 B 31
21 B 78
22 B 62
23 B 76
24 B 96
25 B 80
26 B 56
27 C 68
28 C 79
29 C 56
30 C 91
31 C 71
32 C 71
33 C 87
34 C 41
35 C 59
36 C 68
37 C 53
38 C 79
39 C 15
> bartlett.test(X73~A, data=t)

Bartlett test of homogeneity of variances

data: X73 by A
Bartlett's K-squared = 0.1732, df = 2, p-value = 0.9171

There is no significant difference in the average grades given

by the teachers since the p-value 0.92 is greater than the
alpha which is 0.05 level of significance.
> summary(t)
A X73
A:11 Min. :15.00
B:15 1st Qu.:56.00
C:13 Median :73.00
Mean :68.03
3rd Qu.:80.00
Max. :96.00
> a=aov(X73~A, data=t)
> a
Call:
aov(formula = X73 ~ A, data = t)

Terms:
A Residuals
Sum of Squares 337.598 13323.376
Deg. of Freedom 2 36

Residual standard error: 19.23782

Estimated effects may be unbalanced

2.
> a<-read.csv("D:/final1.csv",header=T,sep=",")
> a
Subject Students Grades
1 Mathematics 1 68
2 Mathematics 2 83
3 Mathematics 3 72
4 Mathematics 4 55
5 Mathematics 5 92
6 English 1 57
7 English 2 94
8 English 3 81
9 English 4 73
10 English 5 68
11 French 1 73
12 French 2 91
13 French 3 63
14 French 4 77
15 French 5 75
16 Biology 1 61
17 Biology 2 86
18 Biology 3 59
19 Biology 4 66
20 Biology 5 87
> b=aov(Grades~Subject*Students,data=a)
> b
Call:
aov(formula = Grades ~ Subject * Students, data = a)

Terms:
Subject Students Subject:Students Residuals
Sum of Squares 42.150 46.225 106.275 2578.300
Deg. of Freedom 3 1 3 12

Residual standard error: 14.65805

Estimated effects may be unbalanced
> summary(a)
Subject Students Grades
Biology :5 Min. :1 Min. :55.00
English :5 1st Qu.:2 1st Qu.:65.25
French :5 Median :3 Median :73.00
Mathematics:5 Mean :3 Mean :74.05
3rd Qu.:4 3rd Qu.:83.75
Max. :5 Max. :94.00
> summary(a)
Subject Students Grades
Biology :5 Min. :1 Min. :55.00
English :5 1st Qu.:2 1st Qu.:65.25
French :5 Median :3 Median :73.00
Mathematics:5 Mean :3 Mean :74.05
3rd Qu.:4 3rd Qu.:83.75
Max. :5 Max. :94.00
3.
> v<-read.csv("D:/bio1.csv",header=T,sep=",")
> v
Subject Students Grades
1 Mathematics 1 88
2 Mathematics 1 79
3 Mathematics 1 63
4 Mathematics 1 80
5 Mathematics 2 79
6 Mathematics 2 56
7 Mathematics 2 96
8 Mathematics 2 68
9 Mathematics 3 67
10 Mathematics 3 51
11 Mathematics 3 66
12 Mathematics 3 89
13 Mathematics 4 35
14 Mathematics 4 64
15 Mathematics 4 70
16 Mathematics 4 60
17 Mathematics 5 99
18 Mathematics 5 87
19 Mathematics 5 77
20 Mathematics 5 95
21 English 1 51
22 English 1 72
23 English 1 65
24 English 1 58
25 English 2 85
26 English 2 67
27 English 2 95
28 English 2 88
29 English 3 74
30 English 3 59
31 English 3 47
32 English 3 82
33 English 4 76
34 English 4 26
35 English 4 49
36 English 4 76
37 English 5 84
38 English 5 83
39 English 5 94
40 English 5 76
41 French 1 73
42 French 1 77
43 French 1 81
44 French 1 77
45 French 2 82
46 French 2 80
47 French 2 36
48 French 2 68
49 French 3 91
50 French 3 59
51 French 3 95
52 French 3 92
53 French 4 43
54 French 4 42
55 French 4 52
56 French 4 32
57 French 5 95
58 French 5 98
59 French 5 81
60 French 5 96
61 Biology 1 87
62 Biology 1 92
63 Biology 1 81
64 Biology 1 76
65 Biology 2 80
66 Biology 2 62
67 Biology 2 93
68 Biology 2 67
69 Biology 3 77
70 Biology 3 84
71 Biology 3 70
72 Biology 3 73
73 Biology 4 55
74 Biology 4 53
75 Biology 4 49
76 Biology 4 56
77 Biology 5 83
78 Biology 5 87
79 Biology 5 76
80 Biology 5 80
> w=aov(Grades~Subject*Students,data=v)
> w
Call:
aov(formula = Grades ~ Subject * Students, data = v)

Terms:
Subject Students Subject:Students Residuals
Sum of Squares 157.937 2.025 467.025 22634.400
Deg. of Freedom 3 1 3 72

Residual standard error: 17.73039

Estimated effects may be unbalanced
> summary(w)
Df Sum Sq Mean Sq F value Pr(>F)
Subject 3 158 52.65 0.167 0.918
Students 1 2 2.03 0.006 0.936
Subject:Students 3 467 155.68 0.495 0.687
Residuals 72 22634 314.37
 Part 2
1.
> shapiro.test(t$X73)

Shapiro-Wilk normality test

data: t$X73
W = 0.94508, p-value = 0.05616
> bartlett.test(X73~A, data=t)

Bartlett test of homogeneity of variances

data: X73 by A
Bartlett's K-squared = 0.1732, df = 2, p-value = 0.9171
> LSD.test(model,"A",p.adj="bon",console=TRUE)

Study: model ~ "A"

LSD t Test for X73

P value adjustment method: bonferroni

Mean Square Error: 370.0938

A, means and individual ( 95 %) CI

X73 std r LCL UCL Min Max

A 67.63636 19.38181 11 55.87256 79.40016 36 93
B 71.40000 18.10998 15 61.32608 81.47392 31 96
C 64.46154 20.36179 13 53.64042 75.28266 15 91

alpha: 0.05 ; Df Error: 36

Critical Value of t: 2.51104

Minimum difference changes for each comparison

Means with the same letter are not significantly different.

Groups, Treatments and means

a B 71.4
a A 67.64
a C 64.46
> comparison<-duncan.test(model,"A",main="X73 with different A")
> duncan.test(model,"A",alpha=0.01,console=TRUE)

Study: model ~ "A"

Duncan's new multiple range test

for X73

Mean Square Error: 370.0938

A, means

X73 std r Min Max

A 67.63636 19.38181 11 36 93
B 71.40000 18.10998 15 31 96
C 64.46154 20.36179 13 15 91

alpha: 0.01 ; Df Error: 36

Critical Range
2 3
20.68556 21.57125

Harmonic Mean of Cell Sizes 12.79324

Different value for each comparison

Means with the same letter are not significantly different.
Groups, Treatments and means
a B 71.4
a A 67.64
a C 64.46

2.
> shapiro.test(a$Grades)

Shapiro-Wilk normality test

data: a$Grades
W = 0.95858, p-value = 0.5159
> bartlett.test(Students~Subject, data=a)

Bartlett test of homogeneity of variances

data: Students by Subject

Bartlett's K-squared = 0, df = 3, p-value = 1
> bartlett.test(Grades~Students, data=a)

Bartlett test of homogeneity of variances

data: Grades by Students

Bartlett's K-squared = 1.8657, df = 4, p-value = 0.7604
> library(agricolae)
> model<-aov(Students~Subject, data=a)
> comparison<-duncan.test(model,"Subject", main="Students with
different Subject")
> duncan.test(model,"Subject",alpha=0.01,console=TRUE)

Study: model ~ "Subject"

Duncan's new multiple range test

for Students
Mean Square Error: 2.5

Subject, means

Students std r Min Max

Biology 3 1.581139 5 1 5
English 3 1.581139 5 1 5
French 3 1.581139 5 1 5
Mathematics 3 1.581139 5 1 5

alpha: 0.01 ; Df Error: 16

Critical Range
2 3 4
2.920782 3.046418 3.128607

Means with the same letter are not significantly different.

Groups, Treatments and means

a Biology 3
a English 3
a French 3
a Mathematics 3

> library(agricolae)
> model<-aov(Students~Subject, data=a)
> out<-LSD.test(model,"virus", p.adj="bonferroni")
> attach(a)
> df<-df.residual(model)
> MSerror<-deviance(model)/df
> out<-LSD.test(model,"Subject", p.adj="bonferroni")
> out
$statistics
Mean CV MSerror LSD
3 52.70463 2.5 3.008334

$parameters
Df ntr bonferroni
16 4 3.008334

$means
Students std r LCL UCL Min Max
Biology 3 1.581139 5 1.501001 4.498999 1 5
English 3 1.581139 5 1.501001 4.498999 1 5
French 3 1.581139 5 1.501001 4.498999 1 5
Mathematics 3 1.581139 5 1.501001 4.498999 1 5

$comparison
NULL

$groups
trt means M
1 Biology 3 a
2 English 3 a
3 French 3 a
4 Mathematics 3 a
3.> shapiro.test(v$Grades)

Shapiro-Wilk normality test

data: v$Grades
W = 0.95499, p-value = 0.006728
> bartlett.test(Students~Subject, data=v)

Bartlett test of homogeneity of variances

data: Students by Subject

Bartlett's K-squared = 0, df = 3, p-value = 1

> bartlett.test(Grades~Students, data=v)

Bartlett test of homogeneity of variances

data: Grades by Students

Bartlett's K-squared = 7.9037, df = 4, p-value = 0.09517
> library(agricolae)
> model<-aov(Students~Subject, data=v)
> comparison<-duncan.test(model,"Subject", main="Students with
different Subject")
> duncan.test(model,"Subject",alpha=0.01,console=TRUE)

Study: model ~ "Subject"

Duncan's new multiple range test

for Students

Mean Square Error: 2.105263

Subject, means

Students std r Min Max

Biology 3 1.450953 20 1 5
English 3 1.450953 20 1 5
French 3 1.450953 20 1 5
Mathematics 3 1.450953 20 1 5

alpha: 0.01 ; Df Error: 76

Critical Range
2 3 4
1.212269 1.263747 1.298570

Means with the same letter are not significantly different.

Groups, Treatments and means

a Biology 3
a English 3
a French 3
a Mathematics 3
>
> library(agricolae)
> model<-aov(Students~Subject, data=v)
> out<-LSD.test(model,"Subject", p.adj="bonferroni")
> attach(v)
The following objects are masked from a (pos = 3):

Grades, Students, Subject

The following objects are masked from a (pos = 4):

Grades, Students, Subject

> df<-df.residual(model)
> MSerror<-deviance(model)/df
> out<-LSD.test(model,"Subject", p.adj="bonferroni")
> out
$statistics
Mean CV MSerror LSD
3 48.36508 2.105263 1.243007

$parameters
Df ntr bonferroni
76 4 2.70907

$means
Students std r LCL UCL Min Max
Biology 3 1.450953 20 2.353816 3.646184 1 5
English 3 1.450953 20 2.353816 3.646184 1 5
French 3 1.450953 20 2.353816 3.646184 1 5
Mathematics 3 1.450953 20 2.353816 3.646184 1 5

$comparison
NULL

$groups
trt means M
1 Biology 3 a
2 English 3 a
3 French 3 a
4 Mathematics 3 a