You are on page 1of 25

r.

Ak ht er
ANALYSIS OF VARIANCE

QMDM

by: Dr. Akhter

Institute of Business Administration Spring 2013

r.

Ak ht er

Analysis of Variance

r.

Ak ht er

Process for comparing four population means

Hypothesis for one way ANOVA

r.

Ak ht er

r.

Ak ht er

r.

Ak ht er

r.

Ak ht er

r.

Ak ht er

Checking Assumptions of ANOVA

Checking normality

r.

Ak ht er

r.

Ak ht er

r.

Ak ht er

r.

Ak ht er

r.

Ak ht er

r.

It is suspected that higher priced automobiles are assembled with greater care than lower-priced automobiles. To investigate whether there is any basis for this feeling, a large luxury model A, a medium sedan B and subcompact C were compared for defects when they arrived at the dealers showroom. All cars were manufactured by the same company. The number of defects for several of the three models are recorded in the following table

Ak ht er

Scenario

r.

Ak ht er
Number of automobile defects A 4 7 6 6 B 5 1 3 5 3 4 C 8 6 8 9 5

r.

Check the assumption of homogeneity of variances Apply analysis of variance procedure to check the whether the average sale of all five brands are same or not If not check which means are same and which are significant

Ak ht er

Perform the followings

Samples are selected independently from the populations Samples selected from populations which are normal Variances of population must be equal

r.

Ak ht er

Assumptions of ANOVA

Total

r.

Ak ht er
Number of automobile defects A 4 7 6 6 B 5 1 3 5 3 4 C 8 6 8 9 5 23 21 36 80

si2

1.583

2.3

2.7

Checking homogeneity

As the ratio of two sample variances (highest and lowest) is less than 2 which indicates that the population variances are same and homogeneity exists

r.


2 1 2 2 2 3 2 4

Ak ht er
2 5

Accept the hypothesis and conclude that variances of all three populations are equal No transformation is needed before using analysis of variance technique

r.

Ak ht er

Decision

H 0 : 1 2 3

H1 : At least brandsare not Same

r.

Ak ht er

Analysis of variance

80 cf 426.66 15 2 2 2 SST 4 7 ... 5 cf 65.333


2 2 2

23 21 36 SSC cf 38.283 4 6 5 SSE 65.333 38.283 27.050

r.

Ak ht er
Calculations
2

S.v SSC

SSE
SST

r.

Ak ht er
ANOVA Table
S.S. d.f. 2 M.S.S F-ratio 8.49 38.283 19.142

12
14

27.050
65.333

2.254

Reject Null Hypothesis and conclude that average number of defects for the three models is not the same

r.

Ak ht er
f 0.05, 2,12 3.89

Critical value and Decision

r.
The End

Ak ht er

You might also like