Chapter 14: Analysis of Variance

(ANOVA)
14-1 One-Way Analysis of Variance: Using the one-way
ANOVA technique to determine if there is a
significant difference among three or more means

𝑠2𝐵
𝐹= 2
𝑠𝑊
• Using the F test to compare three or more means
• The technique we will use is called an analysis of
variance or ANOVA.
• When would you use this type of test?

• A marketing analyst wants to know whether the average number of

calories for fast-food sandwiches differs among Burger King, McDonalds,
and Carl’s Jr.
• A researcher is interested in seeing if there is a difference in the average
annual child-care in three neighboring counties in his state.
• A healthcare official wants to ascertain whether blood pressure differs
among three different treatment strategies: medication, diet, and exercise.
 About the Hypothesis test for a one-way ANOVA
Assumptions for One-Way ANOVA
1. The populations from which the samples were obtained are normally
distributed
2. The dependent variable is an interval/ratio level of measurement.
3. The variances of the populations are equal
4. The samples are independent of one another

We will follow the same 5 step process we’ve become accustomed to so far.
Here are a couple of differences:

1) The hypotheses will be written as a comparison of 3 or more means:

In general,
About the F-distribution used for the ANOVA
• The F-test can be used to used to compare two
variances, but can also be used to compare three or
more means

The F Family of Curves:

• The F distribution is a distribution of
the f statistic, (the ratio of the
variances from two independent
samples).
• The F distribution is positively skewed
and the F-test to compare means is
always right-tailed
• The F distribution’s family of curves
depends on two degrees of freedom:
the d.f. of the numerator, and the d.f.
of the denominator.
 About the Hypothesis test for a one-way ANOVA
(ctnd)
2) The test value is
where
B represents “Between groups”, and
W represents “Within groups”
and where
degrees of freedom of the numerator is: d.f.N. = k – 1 and,
degrees of freedom of the denominator is: d.f.D. = N – k
k is the number of groups, and
N is the sample size of all k groups together.
The sample sizes need not be equal!

Furthermore, to compute F,

2  n Xi i  X GM 
2

and sW2 
 i
 n  1  si
2

s 
B
k 1   ni  1

where
 More about the F Test
• In the F test, two different estimates
of the population variance are made
(between group and within group)
• The first estimate is called the
between-group variance, and it
involves finding the variance of the
means. 𝑠 2
𝐵
• The second estimate, the within- 𝐹= 2
𝑠 𝑊
group variance, is made by
computing the variance using all the
data and is not affected by differences
in the means.
More about the F test (ctnd)

• If there is no difference in the

means, the between and within 𝑠2𝐵 𝑠2
group variances will be 𝐹= 2
≈ 2 ≈1
𝑠 𝑊
𝑠
approximately equal.
(F test value will be close to 1) Do
not reject .
• When the means differ
significantly, the between-group 2 𝐿𝑎𝑟𝑔𝑒
𝑠 𝐵 𝑛𝑢𝑚𝑏𝑒𝑟
variance will be much larger than𝐹 = ≈ ≈ 𝑛𝑢𝑚𝑏𝑒𝑟 >1
𝑠 𝑊 𝑠𝑚𝑎𝑙𝑙
2
the within-group variance.
𝑛𝑢𝑚𝑏𝑒𝑟
(F test will be significantly greater
than 1) Reject .
Example: Lowering Blood Pressure
A researcher wishes to try three different techniques to lower
the blood pressure of individuals diagnosed with high blood
pressure. The subjects are randomly assigned to three groups;
the first group takes medication, the second group exercises,
and the third group follows a special diet. After four weeks, the
reduction in each person’s blood pressure is recorded. At α =
0.05, test the claim that there is no difference among the
means.
Example : Lowering Blood Pressure
A researcher wishes to try three different techniques to lower
the blood pressure of individuals diagnosed with high blood
pressure. The subjects are randomly assigned to three groups;
the first group takes medication, the second group exercises,
and the third group follows a special diet. After four weeks, the
reduction in each person’s blood pressure is recorded. At α =
0.05, test the claim that there is no difference among the
means.
Step 0: Assumptions
1. The populations from which the samples were obtained are normally distributed
2. The dependent variable is an interval/ratio level of measurement.
3. The variances of the populations are equal
4. The samples are independent of one another

Step 1: State the hypotheses and identify the claim.

H0: μ1 = μ2 = μ3 (claim)
H1: At least one mean is different from the others.
Step 2: Find the critical value.

Since k = 3, N = 15,
and α = 0.05,
d.f.N. = k – 1 = 3 – 1 = 2
d.f.D. = N – k = 15 – 3 = 12

The critical value is 3.89, obtained

from Appendix E (textbook) or
from the GeoGebra Probability
Calculator.
Step 3: Compute the test value.

1. Enter the values in 3 columns

in spreadsheet view.

2. Select all values and click on

the “Multiple Variable
Analysis” button.

3. Click Analyze
Step 3: Compute the test value (continued).

4. Click the Show

Statistics button.

5. In the statistics drop

down list, select ANOVA.
Step 3: Compute the test value (continued).

6. Record your F test value.

Notice that the d.f. are computed

for you?

** Be efficient:
you could run the ANOVA test in
GeoGebra first, then use the
results to find your critical values
based on the d.f. that are given.
Step 4: Make the decision.
Reject the null hypothesis, since the critical value is
3.89 and the test value f=9.17 is in the critical region
for a right-tailed test.

Step 5: Summarize the results.

There is enough evidence to reject the claim that the

means are the same.

Note that this suggests that at least one mean is different from the others.
Another statistical test can be used to determine which one is different.
Some important things to note:
• For three groups, the F test can only show whether or not a difference exists
among the three means, not where the difference lies.
• Other statistical tests, Scheffé test and the Tukey test, are used to find where
the difference exists.
• You might be wondering why we don’t just use the t test. The t test was used to
compare two means. It should not be used to compare three or more because
• it would require you to do comparisons of two groups at a time. When you compare 2
means, the other means are ignored. The ANOVA tests all means simultaneously.
• If you are testing 3 means with a t-test, you’d need to run the test 3 times. If you were
testing 5 means, you would need to run the t-test 10 times. You only need to run an F-test
once.
• When you run multiple t-tests instead of one F –test you are increasing your probability of
making a type one error.

15
Example: Toll Road Employees
A state employee wishes to see if there is a significant
difference in the number of employees at the interchanges of
three state toll roads. The data are shown. At α = 0.05, can it be
concluded that there is a significant difference in the average
number of employees the interchanges?
Example: Toll Road Employees (continued)
Step 1: Assumptions:
1. The populations from which the samples were obtained are normally distribu
2. The dependent variable is an interval/ratio level of measurement.
3. The variances of the populations are equal
4. The samples are independent of one another

Step 1: State the hypotheses and identify the claim.

H0: μ1 = μ2 = μ3
H1: At least one mean is different from the others
(claim).
 Example : Toll Road Employees
Let’s do this together. Run the ANOVA test and we will find our test value and
critical value from the results.

Step 2: Find the critical value.

Step 3: Compute the test value.

Example: Toll Road Employees

Step 4: Make the decision.

Step 5: Summarize the results.

ANOVA Summary Tables
• The between-group variance is sometimes called the mean square, MSB.
• The within-group variance is sometimes called the mean square, MSW.
If you are calculating the test value by hand:
• The numerator of the formula to compute MSB is called the sum of squares
between groups, SSB.
• The numerator of the formula to compute MSW is called the sum of squares
within groups, SSW.
ANOVA Summary table from Toll Road Employees example

20
Note Regarding Example 1
• The following slides demonstrate hand calculations
required to do an ANOVA by hand.
• Geogebra will run the ANOVA test for you.
However, it may be useful for you to view the next
four slides to that you can appreciate what
technology does for you in terms of running the
tests and obtaining the results from a software
program.
Computing the test value (by hand) Example: Lowering Blood Pressure

Step 3: Compute the test value.

a. Find the mean and variance of each sample
(these were provided with the data).

b. Find the grand mean, the mean of all

values in the samples.

X GM 
 X

10  12  9    4 116
  7.73
N 15 15

2
c. Find the between-group variance, . sB
n X  X GM 
2

s 2

 i i
B
k 1
Step 3: Compute the test value. (continued)
2
s
c. Find the between-group variance, B .
5 11.8  7.73  5 3.8  7.73  5 7.6  7.73 
2 2 2
2
sB 
3 1
160.13
  80.07
2
2
s
d. Find the within-group variance, W .

s2

 i
 n  1  si
2


4 5.7   4 10.2   4 10.3  104.80
  8.73
  ni  1
W
444 12

e. Compute the F value.

sB2 80.07
F 2   9.17
sW 8.73
Computing the Test Value (by hand) Example: Toll Road Employees

Step 3: Compute the test value.

a. Find the mean and variance of each sample
(these were provided with the data).

b. Find the grand mean, the mean of all

values in the samples.

X GM 
 X

7  14  32    11 152
  8.4
N 15 18
2
c. Find the between-group variance, sB

n X  X GM 
2

s 2

 i i
B
k 1
Step 3: Compute the test value. (continued)
c. Find the between-group variance, sB2
6 15.5  8.4   6  4  8.4   6 5.8  8.4 
2 2 2
2
s 
B
3 1
459.18
  229.59
2
2
s
d. Find the within-group variance, W
sW2 
  ni  1 si
2

  ni  1
5 81.9   5  25.6   5  29.0  682.5
   45.5
555 15
e. Compute the F value.
sB2 229.59
F 2   5.05
sW 45.5