Beruflich Dokumente
Kultur Dokumente
Week 6-1
Week 6-3
Chapter Overview
Analysis of Variance (ANOVA) One-Way ANOVA F-test TukeyKramer test
Week 6-4
ANOVA
ANalysis Of VAriance
Not discussed in this chapter
Week 6-6
Week 6-8
Week 6-10
H 1 : At least one of
is not equal,
whe re i = 1,2,3,....
At least one population mean is different i.e., there is a treatment effect Does not mean that all population means are different (some pairs may be the same)
Week 6-11
Step 1 (Continue)
H0 : 1 = 2 = 3 = = k
All Means are the same: The Null Hypothesis is True (No Treatment Effect)
1 = 2 = 3
Week 6-12
Step 1 (Continue)
H 1 : At least one of
is not equal,
whe re i = 1,2,3,.... , k
At least one mean is different: The Null Hypothesis is NOT true (Treatment Effect is present)
i
or
1 = 2 3
1 2 3
Week 6-13
Significance Level =
=Type 1 error
Week 6-14
= .05
Do not reject H0
Reject H0
F , k 1 , n k
Week 6-16
Before you compute the test statistic value, you need to develop ANOVA table
Week 6-17
Step 5 (continue)
One-Way ANOVA Table
Source of Variation Among Groups Within Groups Total Sum of Square SSA SSW SST Degree of Freedom k-1 n-k n-1 Mean Square MSA = SSA k-1 SSW MSW = n-k F statistic MSA F= MSW
Test Statistic
Week 6-18
Commonly referred to as: Sum of Squares Between Sum of Squares Among Sum of Squares Explained Among Groups Variation
Step 5 (Continue)
Total variation can be split into two parts:
Step 5 (Continue)
TSS = SSA + SSW
Total Variation = the aggregate dispersion of the individual data values across the various factor levels (SST) Among-Group Variation = dispersion between the factor sample means (SSA) Within-Group Variation = dispersion that exists among the data values within a particular factor level (SSW)
Week 6-21
Step 5 (Continue)
Total Sum of Squares/ Total Variation
Where:
TSS = ( X ij X )
j =1 i =1
nj
k = number of groups (levels or treatments) nj = number of observations in group j Xij = ith observation from group j X = grand mean (mean of all data values)
Week 6-22
Step 5 (Continue)
TSS = ( X 11 X ) 2 + ( X 12 X ) 2 + ... + ( X kni X ) 2
Response, X
X
Group 1 Group 2 Group 3
Week 6-23
Step 5 (Continue)
Sum Squares Among the Groups/ AmongGroup Variation
Where:
SSA = n j ( X j X )
j =1
SSA = Sum of squares among groups k = number of groups or populations nj = sample size from group j Xj = sample mean from group j X = grand mean (mean of all data values)
Week 6-24
Step 5 (Continue)
SSA =
j =1
nj(X
X )2
j
Week 6-25
Step 5 (Continue)
SSA = n1 ( x1 x ) + n2 ( x2 x ) + ... + nk ( xk x )
Response, X
2 2 2
X3
X1
Group 1 Group 2
X2
Group 3
Week 6-26
Step 5 (Continue)
TSS = SSA + SSW
Sum Squares of Within Group/WithinGroup Variation
2
SSW =
Where:
j =1
i =1
nj
( X ij X j )
SSW = Sum of squares within groups k = number of groups nj = sample size from group j Xj = sample mean from group j Xij = ith observation in group j
Week 6-27
Step 5 (Continue)
SSW =
j =1 i =1
(X
ij
Summing the variation within each group and then adding over all groups
i
Week 6-28
Step 5 (Continue)
SSW = ( x11 X 1 ) + ( X 12 X 2 ) + ... + ( X knk X k )
Response, X
2 2 2
X3
X1
Group 1 Group 2
X2
Group 3
Week 6-29
Step 5 (Continue)
Obtaining the Mean Squares = Sum Squares / Degree of Freedom
Mean Square Among Groups
SSA MSA = k 1
SSW MSW = nk
Mean Square Within Group = SSW/degrees of freedom for sum squares within group
Week 6-30
Step 5 (Continue)
Test statistic for One-Way ANOVA is computed as
MSA F= MSW
MSA is mean squares among variances MSW is mean squares within variances
Week 6-31
Week 6-32
Week 6-33
Week 6-34
P-value is the probability that we make the type 1 error (reject Ho given that Ho is true) after conclusion is drawn.
Week 6-35
Bounded p-value
Obtain from
Actual p-value
Week 6-36
1= 2
x
Week 6-37
i = j i j
,where
i j
2. Significance level:
3. Decision rule: We reject Ho when test statistic is greater than critical value.
Week 6-39
where: QU = Value from Studentized Range Distribution with k and n - k degrees of freedom for the desired level of (see appendix E.9 table) MSW = Mean Square Within ni and nj = Sample sizes from groups j and j
Week 6-40
5.Test Statistic
Absolute Different Two Sample Means =
where:
j
Xi X
Week 6-42
Week 6-43
X1
X2
X
X3
3
Week 6-44
SSA = 5 (249.2 227)2 + 5 (226 227)2 + 5 (205.8 227)2 = 4716.4 SSW = (254 249.2)2 + (263 249.2)2 ++ (204 205.8)2 = 1119.6 MSA = 4716.4 / (3-1) = 2358.2 MSW = 1119.6 / (15-3) = 93.3
Test Statistic:
MSA 2358.2 F= = = 25.275 MSW 93.3
Decision: Reject H0 at = 0.05 Conclusion: There is evidence that at least one i differs from F = 25.275 the rest at = 0.05.
Week 6-46
Do not reject H0
Reject H0
FU = 3.89
Count 5 5 5
MS 2358.2 93.3
F 25.275
P-value 4.99E-05
F crit 3.89
2. Find the QU value from the Studentized Range Q table with k = 3 and (n k) = (15 3) = 12 degrees of freedom for the desired level of ( = .05 used here):
QU = 3.77
Week 6-48
(continued)
4. Compare: 5. All of the absolute mean differences are greater than critical range. Therefore there is a significant difference between each pair of means at 5% level of significance.
x1 x 2 = 23.2 x1 x 3 = 43.4 x 2 x 3 = 20.2
Week 6-49
Tukey Kramer Multiple Comparisons Sample Sample Mean Size 4.75 8 1 4.625 8 2 7.75 8 3 Absolute Difference Comparison 0.125 Group 1 to Group 2 3 Group 1 to Group 3 3.125 Group 2 to Group 3 Std. Error of Difference 0.71910576 0.71910576 0.71910576 Critical Range 2.5744 2.5744 2.5744
Group
Results Means are not different Means are different Means are different
Other Data Level of significance 0.05 Numerator d.f. 3 Denominator d.f. 21 MSW 4.136905 Q Statistic 3.58
Week 6-50
Summary
Described one-way analysis of variance The logic of ANOVA ANOVA assumptions F test for difference in c means The Tukey-Kramer procedure for multiple comparisons
Week 6-51
Week 6-52