t4_2 [compatibility mode].pdf
TRANSCRIPT
-
7/27/2019 T4_2 [Compatibility Mode].pdf
1/4
1
ANALYSIS OF VARIANCE
(ANOVA)
Wah Mong Weh
Jabatan Matematik
IPG KSAH
2
Test Statistic for One-Way ANOVA
A excessively large F test statistic isevidence against equal population means.Thus, the null will be rejected.
F = variance between samplesvariance within samples
3
The variance is calculated in
two different ways and the
ratio of the two values is
formed.
W
B
MSMSF=
4
MSB
, Mean Square Between, the variance
between samples, measures the differences
related to the treatment given to each sample.
2.MSW
Mean Square Within, the variance
within samples, measures the differences
related to entries within the same sample. The
variance within samples is due to sampling
error.
5
Example 1 (Equal sample sizes)
A researcher wishes to try three different
techniques to lower the blood pressure of
individuals diagnosed with high blood
pressure. The subjects are randomly assigned
to three groups; the first group takes
medication, the second group exercises, and
the third group follows a special diet. After
four weeks, the reduction in each person's
blood pressure is recorded. At =0.05, test the
claim that there is no difference among the
means. The data are shown6
Medication
Exercise Diet
10 6 5
12 8 9
9 3 12
15 0 8
13 2 4
Sample means 11.8 3.8 7.6
sample sd 2.3874 3.193 3.209
sample var 5.7 10.2 10.3
-
7/27/2019 T4_2 [Compatibility Mode].pdf
2/4
7
The hypotheses
H0: 1=2=3
H1: At least one mean is different from theothers
Find the critical value for F =0.05,
df (2, 12) = 3.89
2
2
p
x
s
nsF =
8
(i) Compute the variance of the sample means ie
compute the variance of the three values: 11.8, 3.8,
7.6 with a calculator
std dev = 4.00 var = 16
(ii) Compute the mean of the sample variances ie
find the mean of the three values: 5.7, 10.2, 10.3
mean = 8.73
F = 5(16)/8.73
= 9.1638 = 9.16
(iii) Since 9.16 > 3.89, reject the null hypothesis
There is enough evidence to support the claim that
there is a difference among the means
9
Example 2
A PLUS employee wishes to see if there is a
significant difference in the number of
employees at the PLUS offices in three states.
The data are shown. At =0.05, can it be
concluded that there is a significant difference
in the average number of employees at each
state?
10
Kedah Penang Perak7 10 1
14 1 1232 1 119 0 910 11 111 1 11
mean 15.5 4 5.8std dev 9.05 5.06 5.38
ar 81.9 25.6 29.0
11
The hypotheses are
H0: 1=2=3
H1: At least one mean is different from the other
Find the critical value for F based on =0.05,
df (2, 15) = 3.68
(i) Compute the variance of the sample means
compute the variance of the three values: 15.5,
4.0, 5.8 with a calculator
(ii) Compute the mean of the sample variances ie
find the mean of the three values: 81.9, 25.6,
29.0
12
F = 6(38.32)/45.5
= 5.05319= 5.05
(iii) Since 5.05 > 3.68, rejectthe null hypothesis
There is enough evidence to support the claim thatthere is a difference among the means
-
7/27/2019 T4_2 [Compatibility Mode].pdf
3/4
13
Example 1 (Equal sample sizes)Medica
tionExercis
e Diet
10 6 512 8 99 3 12
15 0 813 2 4
Sample totals 59 19 38
Grand mean = 7.73
x = 116x2 = 1162
14
The hypotheses are
H0: 1=2=3
H1: At least one mean is different from theothers
Find the critical value for F =0.05, df (2, 12)
3.89
(i) Compute the total sum of squares SST
(ii) Compute the between sum of squares SSB
15
Lets try solve example 2 with thismethod
16
Example 1 : Output ExcelAnova: Single Factor
SUMMARY
Groups Count Sum Average Variance
Column 1 5 59 11.8 5.7
Column 2 5 19 3.8 10.2
Column 3 5 38 7.6 10.3
ANOVA
Source of
Variation SS df MS F P-value F crit Between
Groups 160.13 2.00 80.07 9.17 0.00 3.89
WithinGroups 104.80 12.00 8.73
Total 264.9333 14
17
Q5AMei2012
Sebanyak 15 orang telah diagih secara rawakkepada 3 kumpulan untuk menjalani satukajian dengan 3 jenis kaedah pengajaranMatematik yang berbeza.
Pada penghujung semester, ujian diberikepada 15 pelajar itu. Jadual di bawahmenunjukkan markah yang diperolehi olehpelajar-pelajar dalam ketiga-tiga kumpulanitu.
18
KAEDAH A KAEDAH B KAEDAH C
50 85 95
70 55 75
50 70 70
65 90 85
90 70 70
N1 = 5 N2 = 5 N3 = 5
T1 = 325 T2 = 370 T3 = 395
(a) Jalan ujian ANOVA pada aras keertian 0.01 dengan
(1) nyatakan hipotesis nol dan hipotesis alternatif yangberkaitan.
-
7/27/2019 T4_2 [Compatibility Mode].pdf
4/4
19
(ii) tentukan nilai kritikal bagi F
(iii) Kirakan nilai bagi ujian statistik
(b) Apakah keputusan ujian tersebut ?
20
Q5AMei2011
Bahagian pentadbiran akademik Universiti
Putra (UPM) berminat untuk mengkajipencapaian akademik bagi tiga programIjazah Sarjana Muda (BS). Data bagi puratamata gred terkumpul (PMGK) daripadaprogram BS Pertanian, BS Ekonomi dan BSEkologi telah dipungut dan direkod sepertidalam jadual dibawah.
21
BS Pertanian BS Ekologi BS Ekonomi
2.3 2.5 3.1
2.2 2.7 3.3
2.4 2.7 3.2
2.3 2.9 2.9
2.5 2.6 3.0
(a) Nyatakan hipotesis alternatif dan hipotesis nol.
(b) Lengkapkan jadual bagi ringkasan ANOVA
22
PuncaPuncaVariansVarians
Darjahkebebasan
(d,f)
Jumlahkuasa Dua
(SS)
Varians(MS) F
AntaraKumpulan
2
Dalamkumpulan
12
Jumlah 14
Gunakan = 0.05
( c) Apakah kesimpulan yang dapat dibuat oleh bahagian
pentadbiran pada aras = 0.05 ?