Transcript PowerPoint - Dr. Justin Bateh

••••••••••••••••••••••••••••••••••
One Way ANOVA
 In prior notes, we looked at data analysis
strategies for when the researcher needs to
compare two sample means. But, what if a
researcher designs a study in which there are
more than two samples (groups)?
 Using the Single Factor Analysis of Variance
technique in Excel, we can determine if there is a
different between averages of multiple groups or
if the averages are the same. If we have one
independent variable, then we use the One Way
Analysis of Variance.
In this example, we have data on stress
levels from patients. The patients were
randomly assigned to three different groups,
and one group was given music therapy, one
group was given relaxation therapy, and the
other was a control group.
The data is
presented below:
Music
0
6
2
4
3
Relaxation Technique
1
4
3
2
0
Control
5
6
10
8
6
Step 1
We’ve
measured
their stress
levels on a
scale from
1-10, and
we’d like to
see if the
three groups
are the same
or different.
• Put the data into Excel.
Step 2
Click on
Data
Data
Analysis
Anova:
Single
Factor.
 You are choosing Single Factor because you
only have one dependent variable (stress
level ) and three categories of independent
variables (therapy type).
Step 3
• Select the input range
and if you include the top
row then be sure to check
Labels in First Row.
• Click OK and the results
of the ANOVA analysis are
shown as below:
Step 4
Anova: Single Factor
SUMMARY
Groups
Music Therapy
Count
5
Sum
15
5
5
10
35
Relaxation Technique
Control
Average
Variance
3
5
2
7
2.5
4
ANOVA
Source of Variation
SS
df
Between Groups
70
Within Groups
46
Total
116
MS
2
F
35
12 3.833333333
14
P-value
F crit
9.130434783 0.003888652 3.885293835
 Our null hypothesis is that there is no difference
between mean scores of the three groups.
 Our alternative hypothesis is that there is a
difference between mean scores of the three
groups.
 Based upon the P-Value presented here, we
see that 0.003 is less than the alpha
significance that we chose of 0.05 when running
the ANOVA analysis. Since it is less, then we
must reject the hypothesis and conclude that
the alternative hypothesis is correct, which
means that there is a significance difference
between the groups.