In: Statistics and Probability
A statistician for a drug company wishes to
determine whether short-term memory scores are affected by the type
of medication in hyperactive children. Sixty hyperactive children
(30 boys and 30 girls) were randomly assigned to receive either
Cylert or Ritalin (two kinds of amphetamines which are CNS
stimulants). Their short-term memory was tested on a 20-point scale
(where 0 = no memory and 20 = perfect memory).
Boys Cylert: 12, 10, 13, 10, 9, 12, 13, 10, 11,
12
Boys Placebo: 5, 6, 7, 6, 4, 5, 6, 7, 5, 7
Boys Ritalin: 3, 2, 5, 6, 1, 4, 2, 6, 1, 5
Girls Cylert: 5, 6, 2, 4, 3, 2, 4, 6, 2, 4
Girls Placebo: 4, 6, 7, 6, 5, 4, 5, 6, 7, 7
Girls Ritalin: 12, 9, 14, 13, 9, 12, 8, 15, 12, 11
solve for degrees of freedom, sum of squares and f values
Two factor ANOVA with replication analysis is performed here to test whether short term memory scores are affected by the type of medication in hyperactive children.
The two factor ANOVA test for three null hypothesis
H01: There is no difference in the means of factor gender
H02: There is no difference in means of factor medication type
H03: There is no interaction between factors gender and medication type
The test is performed in excel by using following steps,
Step 1: Write the data values in excel. The data values are,
Girl | Boy | |
Cylert | 5 | 12 |
6 | 10 | |
2 | 13 | |
4 | 10 | |
3 | 9 | |
2 | 12 | |
4 | 13 | |
6 | 10 | |
2 | 11 | |
4 | 12 | |
Ritalin | 12 | 3 |
9 | 2 | |
14 | 5 | |
13 | 6 | |
9 | 1 | |
12 | 4 | |
8 | 2 | |
15 | 6 | |
12 | 1 | |
11 | 5 | |
Placebo | 4 | 5 |
6 | 6 | |
7 | 7 | |
6 | 6 | |
5 | 4 | |
4 | 5 | |
5 | 6 | |
6 | 7 | |
7 | 5 | |
7 | 7 |
Step 2: DATA > Data Analysis > ANOVA: Two Factor With Replication > OK. The screenshot is shown below,
Step 3: Select Input Range: All the data values column, Rows per Sample: 10, Alpha = 0.05. The screenshot is shown below,
The result is obtained. The screenshot is shown below,
ANOVA | ||||||
Source of Variation | SS | df | MS | F | P-value | F crit |
Sample | 40.83333 | 2 | 20.41667 | 7.769556 | 0.001083 | 3.168246 |
Columns | 0.416667 | 1 | 0.416667 | 0.158562 | 0.692054 | 4.019541 |
Interaction | 593.4333 | 2 | 296.7167 | 112.9154 | 5.11E-20 | 3.168246 |
Within | 141.9 | 54 | 2.627778 | |||
Total | 776.5833 | 59 |
From the ANOVA table,
ANOVA | |||||||
Source of Variation | SS | df | F | P-value | Decision | ||
Medication type | 40.83333 | 2 | 7.769556 | 0.001083 | < | 0.05 | Significant |
Gender | 0.416667 | 1 | 0.158562 | 0.692054 | > | 0.05 | Not Significant |
Medication type*Gender | 593.4333 | 2 | 112.9154 | 5.11E-20 | < | 0.05 | Significant |
Within | 141.9 | 54 | |||||
Total | 776.5833 | 59 |
There is a significant difference in Medication type and significant interaction between medication type and gender.