WELCOME

A PRESENTATION ON ANALYSIS OF
DATA USING ONE-WAY ANOVA

SUBMITTED BY
MARJIA RAHMAN ROLL NO.080407
AMENA BEGUM ROLL
NO. 080406 SHEULI
AFROZ ROLL NO. 080414
AFRUJA AKHTER ROLL NO.
080413
DEPT. OF BIOCHEMISTRY &
MOLECULAR BIOLOGY
JAHANGIRNAGAR UNIVERSITY . SAVAR ,DHAKA
 Contents Presenting
*ANOVA .
*ALTERNATIVE HYPOTHESES
*APPLICATION.
*ONE-WAY ANOVA .
*EXAMPLE .
*VARIABLES .
*ANOVA ASSUMPTION.
*THE MODEL
*COMPONENTS OF THE
MODEL. *ASSUMPTION OF THE MODEL.
 Contents Presenting
*THE ANOVA PROCEDURE.
*F RATIO.
*FACTORS
INFLUENCING F RATIO.
*ALTERNATIVES.
*TUKES’S HSD TEST.
*STEPS IN
PERFORMING TUKEY’S TEST.
*ANOVA CALCULATION.
*CONCLUSION.
 ANOVA
INTRODUCTION: ANOVA may be defined as
a technique where by the total variations
present in a set of data is partitioned into
two or more components . Associated with
each of these components is a specific
source of variation, so that in the analysis it
is possible to ascertain the magnitude of
the contributions of each of these sources
to the total variation.
 ANOVA
Part of the solution to this problem rests in a
statistical procedure known as the analysis of
variance or ANOVA for short
ANOVA replaces the multiple comparisons with a
single omnibus null hypothesis
Omnibus -- covers all aspects
In ANOVA, H0 is always of the form:
H0: 1 = 2 = 3 = …. = n
That is, H0 is that all the means are equal
6
ANOVA Alternative Hypothesis
The H0 and H1 must be both mutually
exclusive and exhaustive,H1 for ANOVA
H1 is
H1: 1  2  3 ….  n
In ANOVA, the alternative hypothesis is
always of the form:
H1: not H0

7
 APPLICATIONS
Analysis of variance is used for two
different purp0ses:
1.To estimate and test hypothesis
about population variances.
2.To estimate and test hypothesis
about population means.
 ONE-WAY ANOVA
One way ANOVA is the simplest type of
analysis of variance in which only one source
of variation or factor is investigated.
*It can be
thought of a generalization of the pooled t-
test.
*Instead of dealing with two
populations, we have more populations or
treatments.
Assumptions of one way
ANOVA
01.The populations from which the samples
are drawn normally distributed.
02. The populations from which the
samples are drawn have the same variance.
03. The samples drawn from
different populations are random and
independent.
 EXAMPLE
For creating an underwater sensor array, four
different alloys are tested for corrosion
resistance. Five 3 in plates of each alloy are
placed underwater for 60 days. Each plate
has the number of corrosion pits measured.
 VARIABLES
We alluded to three kinds of variable to be
present in all situations in which the use of
ANOVA is a appropriate.
1.Treatment
variable.
2.Response variable.
3.Extraneous variable.
 ANOVA Assumptions
ANOVA is fairly robust (it will give good
results even if the assumptions are violated) to
the normality assumption and the
homogeneity of variance assumption as long
as:
the number of participants in each group is
equal
the number of participants in each group is
fairly large
13
 ANOVA Assumptions
ANOVA makes certain assumptions:
Sampling error is normal, or Gaussian in
shape, and is centered around the mean
of the distribution
Homogeneity of variance -- the
variability within each group is
approximately equal
Independence of observations
14
 THE MODEL
A model is a symbolic representation of a typical value of
a data set . The symbol Xij represent this typical
value.One way analysis of variance model may be
written as follows:
Xij= µ +τj +eij ; i= 1,2,…..,nj, j= 1,2,….,k
The term in this model are defined as follows:
1.µ
represents the mean of all the k population means & is
called the grand mean.
 THE MODEL
τj represents the differences between the
mean of the jth population & the grand
mean and is called the treatment effect .
3. eij represents the amount by which an
individual measurement differs from the
mean of the population to which it belongs
and is called the error term .
COMPONENTS OF THE MODEL
A typical observation from the total set of
data under study is composed of
1.The grand mean
2.A treatment effect&
3.An error term representing the deviation
of the observation from its group mean .
 Assumptions of the Model
The assumptions for the fixed-effects
model are as follows:
(a)The sets of observed data
constitute k independent random samples
from the respective populations
(b)Each of the populations from
which the samples come is normally
distributed with mean µj & variance σj²
Assumptions of the Model
(c)Each of the populations has the same
variance . That is σ1²= σ2²=…..= σk²= σ²,the
common variance.
(d)The τj are unknown
constants & ∑ τj=0 (e) eij have a mean of 0,
since the mean of xij is µj
Assumptions of the Model

(f) eij have a variance equal to the
variance of the xij ,since the eij & xij differ
only by a constant;
 THE ANOVA PROCEDURE
In our presentation of the analysis of variance for the
different designs, we follow the ten step procedure:
1.Description of data. 2.
Assumptions.
3.Hypotheses. 4.Test
statistic.
5.Distribution of test statistic.
6.Decision rule.
7.Calculation of test statistic. 8.
Statistical decision .
9.Conclusion.
10.Determination of P value.
 DESCRIPTION OF DATA
The measurements (or observation) resulting
from a completely randomized experimental
design, along with the means and totals that can
be computed.
Available subject
01 02 03 04 05 06 07 08 10 11 12 13 14 15
16 Random numbers
16 09 06 15 14 11 02 04 10 07 05
13 03 12 01 18
 Hypotheses.
The null hypothesis that all population or
treatment means are equal against the
alternative that the members of at least one
pair are not equal .We may state the
hypothesis as follows:
H0: 1 = 2 = 3 = …. = n
H A:Not all 1 are equal .If the population
means are equal to zero.
 Test statistic.
The test statistic for one-way analysis of
variance is a computed variance ratio
,which we designate by V.R. The two
variances from which V.R is calculated are
themselves computed from the sample
data.
 THE TOTAL SUM OF SQUARES
The total of squares is the sum of the
squares of the deviations of individual
observations from the mean of all the
observations taken together, designated as
SST.
THE WITHIN GROUP SUM OF
SQUARES
The first step is the calculations within each
group . These calculations involve computing
within each group the sum of the squared
deviations of the individual observations from
their mean . Then we get the sum of the
individual group results . This components of
variation is called the within groups sum of
squares , designated as SSW.
THE AMONG GROUPS SUM OF
SQUARES
The squared deviation of the group mean
from the grand mean and multiply the
result by the size of the group .The quantity
is a measure of the variation among groups
and is referred to as the sum of squares
among groups or SSA.
THE AMONG GROUPS SUM OF
SQUARES
In summary , it has found that the total sum
of squares is equal to the sum of the among
and the within sum of squares .
The relationship is as follow;
SST = SSA + SSW.
From the sums of squares ,it is
possible to obtain two estimates of the
common population variance , σ².
 Distribution of test statistic.
V.R is distributed as the F distribution
when H0 is true and the assumptions are
met. The variable ratio=among group
mean square\within groups mean square.
If the two estimates are about equal ,
V.R will be close to 1.
 Decision rule.
In general ,the decision rule is : reject the
null hypothesis if the computed value of
V.R is equal to or greater than the critical
value of F for the chosen α level.
 Calculation of test statistic.
We have defined analysis of variance as a
process whereby the total variation present
in a set of data is partitioned into
components that are attributable to
different sources. The term variation refers
to the sum of squared deviations of
observations from their mean, or sum of
squares or short.
 Statistical decision .
To reach a decision we must compare our
computed V.R with the critical value of F ,
which we obtain by table G with k-1
numerator degrees of freedom and N-k
denominator degrees of freedom.
 Conclusion.
When we reject H0 we conclude that not
all population means are equal . When we
fail to reject H0, we conclude that the
population means may all be equal.
 F Ratio
Fisher’s F ratio is defined as:

between  groups variance

F
within  groups variance
effect of treatment on DV variance  error variance
F
error variance

34
 F Ratio
When F equal to H0 :
When H0 is true, then there is no effect of the
treatment on the DV
Thus, when H0 is true, we have error / error which
should equal 1
When F is not equal to H0:
When H0 is not true, then there is an effect of the
treatment on the DV
Thus, when H0 is not true, F should be larger than 1

35
 Factors influencing F Ratio
Several factors influencing F ratio those reject
the null hypothesis
To reject H0, the calculated F must be
larger than the critical F which can be
found in a table
Anything that makes the critical F value
large will make it more difficult to reject
H0.

36
 ALTERNATIVES
If the data avail able for analysis do not
meet the assumptions for one-way analysis
of variance ,one may wish to consider the
use of the Kruskal - wallis procedure , a
parametric technique.
 TUKES’S HSD TEST
A multiple comparison procedure developed by
Tukey is frequently used for testing the null
hypotheses that all possible pairs of treatment
means are equal when the samples are all of same
size . When this test is employed we select an
overall significance level of α.The probability is
α,then , that one or more of the null hypotheses is
false . Tukey’s test , which is usually to as the HSD
test ,makes use of a single value against which all
differences are compared .
Steps in Performing Tukey
Tests
Write the hypotheses
H0: 1 = 2
H1: 1  2
Specify the  level
 = .05

39
Steps in Performing Tukey
Tests
Calculate the honestly significant difference (HSD)

MSwithin groups
HSD  q
 , k ,df within groups n
*q,k,dfwithin-groups = tabled q value
* =  level
*k = number of levels of the IV
*dfwithin-groups = degrees of freedom for MSwithin-groups
*MSwithin-groups = within-groups variance estimate
40
Steps in Performing Tukey
Tests
Take the difference of the means of the
conditions you are comparing
If the difference of the means is at least as
large as the HSD, you can reject H0
Repeat for whatever other comparisons
need to be made

41
ANOVA CALCULATION WITH A
DATA
Now we are going to represent an example
of one way ANOVA by collected data from
216 DATA by using SPSS.
 CONCLUSION
In our presentation we have tried to
summarize one-way ANOVA with
definition , methodology , example ,
application ,and other very related topics .
The reference book we have used is
*BIOSTATISTIC:A FOUNDAION FOR
ANALYSIS IN THE HEALTH SCIENCE BY WAYNE
W. DANIEL.7TH EDITION.
*STATISTIC BY PREM S. MANN.FIFTH EDITION.