Scribd
Upload
222 views10 pages

Analysis of Variance

The document discusses one-way analysis of variance (ANOVA), which tests whether the means of multiple populations are equal. It explains that ANOVA partitions the total variability into between-sample and within-sample components. If the between-sample variability is large compared to the expected within-sample variability, then the null hypothesis that the population means are equal is rejected. The document provides an example comparing the lifetimes of four electric bulb brands using a one-way ANOVA test.

Uploaded by

বিজয় সিকদার
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
222 views10 pages

Analysis of Variance

The document discusses one-way analysis of variance (ANOVA), which tests whether the means of multiple populations are equal. It explains that ANOVA partitions the total variability into between-sample and within-sample components. If the between-sample variability is large compared to the expected within-sample variability, then the null hypothesis that the population means are equal is rejected. The document provides an example comparing the lifetimes of four electric bulb brands using a one-way ANOVA test.

Uploaded by

বিজয় সিকদার
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

Analysis of Variance

The statistical method for testing the null hypothesis that the means of several
population are equal is accomplished by F-test. The F-test is based on the
technique of analysis of variance or abbreviated ANOVA. In the present context,
we discuss about one way analysis of variance, in which only one type of treatment
or factor is involved in the experiment.

Assume that we have k independent samples from k normal populations with


population mean respectively with common variance . An
appropriate for testing the hypotheses is

Model for One-Way ANOVA:

Sample 1 Sample 2 ....... Sample k


.......
.......
....... ....... ....... .......
....... ....... ....... .......
.......
Total .......
Mean .......

Grand Total

Page 1 of 10
Grand mean

Resolution of Total Variability into Components:

The Total Sum of Squares (SST) measuring the total variability in the data can be
partitioned and resolved into Sum of Squares Between samples (SSB) and Sum of
Squares Within samples (SSW).

SST = SSB + SSW

( ) ( ) ( )

i.e. Total Variance = Variance between samples + Variance within samples.

For calculation:

Variations Formulae Degree of Freedom


SSB k-1

SSW SST- SSB N-k


SST N-1

Page 2 of 10
Summary Table for Analysis of Variance:

Source of Sum of Degree of Mean Computed


Variation Squares Freedom Squares f
Between SSB k-1
Samples

Within SSW N-k


Samples

Total: SST N-1

Here the F distribution has and degrees of freedom. The


null hypothesis is to be if with and degrees of freedom.

Upper and lower percentage points of the F distribution.

Page 3 of 10
Example: As head of a department of a consumers research organization, you
have the responsibility for testing and comparing lifetimes of four brands of
electric bulbs. Suppose you test the lifetime of three electric bulbs of each of the
four brands. The data is shown below, each entry representing the lifetime of an
electric bulb, measured in hundreds of hours:
Brand
Brand A Brand B Brand C Brand D
20 25 24 23
19 23 20 20
21 21 22 20

Can we infer that the mean lifetime of the four brands of electric bulbs are equal
with 5% level of significance?

Solution: Let us take the null hypothesis that the mean lifetime of the four brands
of electric bulbs are equal, .i.e.

20 400 25 625 24 576 23 529


19 361 23 529 20 400 20 400
21 441 21 441 22 484 20 400

Page 4 of 10

( ) ( ) ( ) ( )

With the above information the ANOVA table for the given problem:

Source of Sum of Degree of Mean Squares Computed


Variation Squares Freedom f
Between
Samples

Within Samples SSW= N-k=12-4=8


39-15 = 24

Total: N-1=12-1=11

The value of with and degrees of freedom at 5% level of


significance is . Since the computed value of f = 1.67 is less than the
table value of , therefore we accept our null hypothesis. Hence, the
difference is insignificant and we can infer that the mean lifetime of the four
brands of electric bulbs are equal.

Page 5 of 10
Page 6 of 10
Page 7 of 10
Page 8 of 10
Page 9 of 10
Page 10 of 10

You might also like