One-Way ANOVA

One-Way Analysis of Variance

 One-Way ANOVA
 The one-way analysis of variance is used
to test the claim that three or more
population means are equal
 This is an extension of the two
independent samples t-test
 One-Way ANOVA
 The response variable is the variable
you’re comparing
 The factor variable is the categorical
variable being used to define the groups
 We will assume k samples (groups)
 The one-way is because each value is
classified in exactly one way
 Examples include comparisons by gender,
race, political party, color, etc.
 One-Way ANOVA
 Conditions or Assumptions
 The data are randomly sampled
 The variances of each sample are assumed
equal
 The residuals are normally distributed
 One-Way ANOVA
 The null hypothesis is that the means are all
equal
H :     
0 1 2 3 k

 The alternative hypothesis is that at least one

of the means is different
 Think about the Sesame Street® game where
three of these things are kind of the same, but one
of these things is not like the other. They don’t all
have to be different, just one of them.
 One-Way ANOVA
 The statistics classroom is divided into
three rows: front, middle, and back
 The instructor noticed that the further the
students were from him, the more likely
they were to miss class or use an instant
messenger during class
 He wanted to see if the students further
away did worse on the exams
 One-Way ANOVA
The ANOVA doesn’t test that one mean is less
than another, only whether they’re all equal or
at least one is different.

H :   
0 F M B
 One-Way ANOVA
 A random sample of the students in each
row was taken
 The score for those students on the
second exam was recorded
 Front: 82, 83, 97, 93, 55, 67, 53
 Middle: 83, 78, 68, 61, 77, 54, 69, 51, 63
 Back: 38, 59, 55, 66, 45, 52, 52, 61
 One-Way ANOVA
The summary statistics for the grades of each row
are shown in the table below

Row Front Middle Back

Sample size 7 9 8
Mean 75.71 67.11 53.50
St. Dev 17.63 10.95 8.96
Variance 310.90 119.86 80.29
 One-Way ANOVA
 Variation
 Variation is the sum of the squares of the
deviations between a value and the mean of
the value
 Sum of Squares is abbreviated by SS and
often followed by a variable in parentheses
such as SS(B) or SS(W) so we know which
sum of squares we’re talking about
 One-Way ANOVA
 Are all of the values identical?
 No, so there is some variation in the data
 This is called the total variation
 Denoted SS(Total) for the total Sum of
Squares (variation)
 Sum of Squares is another name for variation
 One-Way ANOVA
 Are all of the sample means identical?
 No, so there is some variation between the
groups
 This is called the between group variation
 Sometimes called the variation due to the
factor
 Denoted SS(B) for Sum of Squares (variation)
between the groups
 One-Way ANOVA
 Are each of the values within each group
identical?
 No, there is some variation within the groups
 This is called the within group variation
 Sometimes called the error variation
 Denoted SS(W) for Sum of Squares
(variation) within the groups
 One-Way ANOVA
 There are two sources of variation
 the variation between the groups, SS(B), or
the variation due to the factor
 the variation within the groups, SS(W), or the
variation that can’t be explained by the factor
so it’s called the error variation
 One-Way ANOVA
 Here is the basic one-way ANOVA table

Source SS df MS F

Between

Within

Total
 One-Way ANOVA
 Grand Mean k
 The grand mean is the average of all the n x i i
values when the factor is ignored
It is a weighted average of the individual
x i 1
k

n


sample means i
i 1

n x  n x   n x
x 1 1 2 2 k k

n  n   n 1 2 k
 One-Way ANOVA
 Grand Mean for our example is 65.08

7 75.71  9 67.11  8 53.50 

x
7 9 8
1562
x
24
x 65.08
 One-Way ANOVA
 Between Group Variation, SS(B)
 The between group variation is the variation
between each sample mean and the grand
mean
 Each individual variation is weighted by the
sample size
SS  B   n  x  x 
k 2

i i
i 1

SS  B  n  x  x   n  x  x     n  x  x 
2 2 2

1 1 2 2 k k
 One-Way ANOVA
The Between Group Variation for our example is
SS(B)=1902
SS  B  7 75.71  65.08   9 67.11  65.08   8 53.50  65.08 
2 2 2

SS  B  1900.8376 1902

I know that doesn’t round to be 1902, but if you

don’t round the intermediate steps, then it does.
My goal here is to show an ANOVA table from
MINITAB and it returns 1902.
 One-Way ANOVA
 Within Group Variation, SS(W)
 The Within Group Variation is the weighted total of
the individual variations
 The weighting is done with the degrees of freedom
 The df for each sample is one less than the sample
size for that sample.
 One-Way ANOVA
Within Group Variation

SS W   df s
k
2
i i
i 1

SS W  df s  df s    df s
2
1 1 2
2
2 k
2
k
 One-Way ANOVA
 The within group variation for our example
is 3386

SS W  6 310.90   8 119.86   7 80.29 

SS W  3386.31 3386
 One-Way ANOVA
 After filling in the sum of squares, we have …

Source SS df MS F

Between 1902

Within 3386

Total 5288
 One-Way ANOVA
 Degrees of Freedom, df
 A degree of freedom occurs for each value that can
vary before the rest of the values are predetermined
 For example, if you had six numbers that had an
average of 40, you would know that the total had to
be 240. Five of the six numbers could be anything,
but once the first five are known, the last one is fixed
so the sum is 240. The df would be 6-1=5
 The df is often one less than the number of values
 One-Way ANOVA
 The between group df is one less than the
number of groups
 We have three groups, so df(B) = 2
 The within group df is the sum of the individual
df’s of each group
 The sample sizes are 7, 9, and 8
 df(W) = 6 + 8 + 7 = 21
 The total df is one less than the sample size
 df(Total) = 24 – 1 = 23
 One-Way ANOVA
 Filling in the degrees of freedom gives this …

Source SS df MS F

Between 1902 2

Within 3386 21

Total 5288 23
 One-Way ANOVA
 Variances
 The variances are also called the Mean of the
Squares and abbreviated by MS, often with an
accompanying variable MS(B) or MS(W)
 They are an average squared deviation from the
mean and are found by dividing the variation by the
degrees of freedom
 MS = SS / df

Variation
Variance 
df
 One-Way ANOVA
 MS(B) = 1902 / 2 = 951.0
 MS(W) = 3386 / 21 = 161.2
 MS(T) = 5288 / 23 = 229.9
 Notice that the MS(Total) is NOT the sum of
MS(Between) and MS(Within).
 This works for the sum of squares SS(Total),
but not the mean square MS(Total)
 The MS(Total) isn’t usually shown
 One-Way ANOVA
 Completing the MS gives …

Source SS df MS F

Between 1902 2 951.0

Within 3386 21 161.2

Total 5288 23 229.9

 One-Way ANOVA
 Special Variances
 The MS(Within) is also known as the pooled
estimate of the variance since it is a weighted
average of the individual variances
Sometimes abbreviated s
2

p
 The MS(Total) is the variance of the response
variable.
 Not technically part of ANOVA table, but useful none the
less
 One-Way ANOVA
 F test statistic
 An F test statistic is the ratio of two sample
variances
 The MS(B) and MS(W) are two sample
variances and that’s what we divide to find F.
 F = MS(B) / MS(W)
 For our data, F = 951.0 / 161.2 = 5.9
 One-Way ANOVA
 Adding F to the table …

Source SS df MS F

Between 1902 2 951.0 5.9

Within 3386 21 161.2

Total 5288 23 229.9

 One-Way ANOVA
 The F test is a right tail test
 The F test statistic has an F distribution
with df(B) numerator df and df(W)
denominator df
 One-Way ANOVA
 Completing the table with the p-value

Source SS df MS F

Between 1902 2 951.0 5.9

Within 3386 21 161.2

Total 5288 23 229.9

 One-Way ANOVA
 The critical value is 3.47, which is less
than F=5.9,
 so we reject the null hypothesis.
 The null hypothesis is that the means of
the three rows in class were the same, but
we reject that, so at least one row has a
different mean.
 One-Way ANOVA
 There is enough evidence to support the
claim that there is a difference in the mean
scores of the front, middle, and back rows
in class.
 The ANOVA doesn’t tell which row is
different, you would need to look at
confidence intervals or run post hoc tests
to determine that
 No. of bottles sold

Coke Classic New Coke Pepsi

45 6 24
136 114 90
134 56 100
41 14 43
146 39 51
33 20 42
71 42 68
224 156 131
111 61 74
87 140 107

 Is there any difference between the average

number of bottles sold (at 5% level of
significance?)
SUMMARY
Varianc
Groups Count Sum Average e
3554.62
Coke Classic 10 1028 102.8 2
2857.28
New Coke 10 648 64.8 9
1145.55
Pepsi 10 730 73 6
ANOVA
Source of
Variation SS df MS F P-value F crit
1.58735 0.22296 3.35413
Between Groups 7997.6 2 3998.8 7 1 1
68017. 2519.15
Within Groups 2 27 6

76014.
 Five parachutes are woven using the fiber supplied by
each group—Supplier 1, Supplier 2, Supplier 3, and
Supplier 4. You perform the experiment of testing the
strength of each of the 20 parachutes by collecting the
tensile strength measurement of each parachute. The
strength of the parachutes is measured by placing
them in a testing device that pulls on both ends of a
parachute until it tears apart. The amount of force
required to tear the parachute is measured on a
tensile strength scale, where the larger the value, the
stronger the parachute.
 Test whether significant differences exist in the
average strength of parachutes woven using synthetic
fiber purchased from each of the four suppliers (5%
level of significance)
SUMMARY
Groups Count Sum Average Variance
Column 1 5 97.6 19.52 7.237
Column 2 5 121.3 24.26 3.683
Column 3 5 114.2 22.84 4.553
Column 4 5 105.8 21.16 8.903

ANOVA
Source of Variation SS df MS F P-value F crit
Between Groups 63.2855 3 21.09517 3.461628925 0.041366 3.238872
Within Groups 97.504 16 6.094

Total 160.7895 19