Scribd
Upload
1K views

Analysis of Variance (ANOVA) One Way and Two Way

The document discusses analysis of variance (ANOVA). ANOVA is a statistical technique used to compare the means of two or more populations to determine if they are significantly different from each other. It can be used when there are more than two sample groups being compared. Examples provided include comparing psychiatric therapy methods, light bulb manufacturing processes, and student exam scores from different colleges. The basic principle of ANOVA is that the null hypothesis assumes there is no difference between population means, and an F-test is performed to determine if the calculated F value exceeds the critical value from F tables, which would lead to rejecting the null hypothesis. One-way and two-way ANOVA techniques are demonstrated through examples.

Uploaded by

Panma Patel
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
1K views

Analysis of Variance (ANOVA) One Way and Two Way

The document discusses analysis of variance (ANOVA). ANOVA is a statistical technique used to compare the means of two or more populations to determine if they are significantly different from each other. It can be used when there are more than two sample groups being compared. Examples provided include comparing psychiatric therapy methods, light bulb manufacturing processes, and student exam scores from different colleges. The basic principle of ANOVA is that the null hypothesis assumes there is no difference between population means, and an F-test is performed to determine if the calculated F value exceeds the critical value from F tables, which would lead to rejecting the null hypothesis. One-way and two-way ANOVA techniques are demonstrated through examples.

Uploaded by

Panma Patel
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 24

ANALYSIS OF

VARIANCE(ANOVA)

1
WHAT IS ANOVA AND IT’S APPLICATION

Using this technique , one can draw conclusion


Or inferences about whether more than two
sample drawn from populations have the same
mean or not.

The ANOVA technique is important in the context


of all those situations where we want to compare
more than two populations.( multiple sample
cases)
2
EXAMPLES
In ANOVA, basically, you’re testing groups to see if
there’s a difference between them. Examples of
when you might want to test different groups:
 A group of psychiatric patients are trying three
different therapies: counseling, medication and
biofeedback. You want to see if one therapy is
better than the others.
 A manufacturer has two different processes to
make light bulbs. They want to know if one
process is better than the other.
 Students from different colleges take the same
exam. You want to see if one college
outperforms the other.
3
THE BASIC PRINCIPLE OF ANOVA
 In short,
 The Null Hypothesis we assume that there is
no difference between the population means.
 Then we perform F test.

 The calculated value of F is compared with the


table value .

 If F value we work out is equal or exceeds the


critical value ( to be seen from the F table),
we reject null hypothesis.
4
ANOVA TECHNIQUE
One-way ( or single factor) ANOVA :
Example 1
The following table shows the per acre production
data of wheat for three varieties, each grown on 4
plots. State whether variety differences are significant
at 5% level .
Plot of land Per acre production data
Variety of Wheat
A B C
1 6 5 5
2 7 5 4
3 3 3 3
4 8 7 4

In One way ANOVA one factor influences the dependent variable ( agricultural
output) 5
SOLUTION
Null Hypothesis: we assume there is no significant
difference in wheat output due to varieties of seeds.

6
T=∑ X1 + ∑ X2 + ∑ X3 = ( 6+7+3+8) + 20 + 16
= 24 + 20+ 16
= 60
Correction factor( C.F. ) = T 2/N
= ( 60 ) 2/12
=3600/12
=300

7
SST(Total Sum of Squares)= ∑ X1 2+ ∑ X2 2 + ∑ X3 2 - (T 2/N)
= 332-300 = 32

SSB ( Sum of Squares Between ) = (∑ X 1 )2+ ( ∑ X2 )2 + ( ∑ X3 )2 - C.F


n 1 n 2 n3
= ( 24) 2 + ( 20) 2 + (16) 2 - 300
4 4 4

=144+100+ 64 =308-300= 8

SSW ( Sum of squares within )= 32-8= 24


8
STEP 3 :SET UP THE ANOVA TABLE

Source of SS d.f. MS F-ratio F table


variation value
Between 8 (c-1)=2 SS/d.f.=4 4/2.67= Check from
sample 1.498 table with
degrees of
freedom
2,9
Within 24 (n-c)=9 SS/d.f.=2.67
sample

9
 F Ratio = MS Between/ MS Within
 Table value of F v1, v2 = check

MS= mean square

10
TWO – WAY ANOVA
Illustration 2: Per Acre Production Data of Wheat
Varieties of seeds A B C

Varieties of Fertilizers

W 6 5 5

X 7 5 4

Y 3 3 3

Z 8 7 4

Also state whether variety differences are significant at 5% level.

11
12
TWO – WAY ANOVA
Illustration 2: Per Acre Production Data of Wheat
A B C Total
Varieties
of seeds

Varieties
of
Fertilizers
W 6 5 5 16
X 7 5 4 16
Y 3 3 3 9
Z 8 7 4 19
Total 24 20 16 Grand Total

(i)Test whether variety differences are significant at 5% level.


(ii) i)Test whether Fertilizer differences are significant at 5%
13
level.
Step 1
T = Take the total of all values of individual items= 60
N= no. of observations. = 12.
Correction factor = T 2 = 300
N
Step 2
Total SS = ( 36+25+ 25+ 49+ 25+ 16 + 9+9+9+64+49+16) - T 2
N
= 32

14
Step 3
SS between columns treatment = ∑( T j)2 T2
nj n

= 24x 24 + 20x20 + 16x16 - 300


4 4 4
=8

SS between rows treatment = ∑( T i) 2 T2


ni - n
= 18

15
 Step 5
SS residual or error = Total SS – (SS between columns + SS between rows)
=6

16
THE ANOVA TABLE

Source of SS d. f. MS F-ratio F table


variation value(5
%)
between columns 8 (c-1)=(3- 8/2=4 MS between columns F(2,6)=
( i.e between 1)=2 MS residual 5.14
varieties of
seeds) = 4/1 =4
between rows( i.e 18 (r-1) =(4- 18/3= MS between rows F(3,6)=
between varieties 1)=3
6 MS residual 4.76
of fertilizers) = 6/1 = 6

residual or error 6 (c-1) (r-1) 6/6=1


=(3-1). (4-1)
=6

17
CONCLUSION
Column wise F Ratio is significant which mean
production varies with different varieties of
fertilizers.

18
PRACTICE QUESTION
A company wants to test whether its three salesmen A, B
and C have the same selling ability. Their records of sales
(in Rs ‘000) during various weeks of the last month are
given in the following table:

Sales men 1st week 2nd week 3rd week 4th week

A 16 21 18 25

B 22 20 15 26

C 25 24 16 20

Prepare an analysis of variance table and test the


hypothesis that the mean sales per week of all the
salesmen are equal.
19
CASE STUDY 1
One of the questions in the mind of the students while
pursuing their MBA is which area of specialization to
choose. Some of the students do choose their
specialization based on their interest and the future
career they wish to continue. However many are guided
by the prospects of the salary they are capable of
earning. Therefore the question uppermost in their mind
is company’s salaries depend on the nature of
specialization. A student in order to understand the same
independently selects students from three different
specializations: Finance, Marketing and Operations.
These students are both independently and randomly
selected from each of the areas of specializations. The
salary package (CTC) offered to them are gathered and
tabulated below:

20
Finance Marketing Operations
7 5 4
5 6 5
10 7 4
8 4 6
6 4 5
8 6 5
9 4 4

21
CASE LETS 2
The salaries are in lakh of rupees per annum. In
the final analysis if there is no difference in the
salaries in the population, the student would opt
for a choice of specialization of their own interest
but if the sample gives the impression that the
population average salary for the different
specialization is different then he would choose the
specialization which gives him the highest package.
(a) Indicate the null and alternative hypothesis for the
above situation.
(b) Develop an ANOVA table and interpret the solution.
Show each step clearly. (Table value= 3.46 at 5 % level
of significance)
22
3
 The manager in a branch of a major bank is
worried about increasing number of complaints
from retail customers about delays in getting
services. The branch has four executives who
work at the retail customer desks. The manager
would like to know if one or more among the
executives are slower in their work than others.
The following data show the time taken by each
executive in completing the most typical service
request from retail customers. Test the
hypothesis that the average times taken by all
four executives are same. Use the 5% level of
significance.
23
Executive 1 5 7 4 8 6 - -

Executive 2 4 10 6 7 7 - -

Executive 3 9 4 8 6 7 10 6

Executive 4 6 5 7 8 8 6 7

24

You might also like