In: Statistics and Probability
| 
 Teaching as usual  | 
 Teaching+examples  | 
 Teaching+safety training  | 
 General+Person totals  | 
| 
 T =220  | 
 T =162  | 
 T =117  | 
 Σx2 =7753  | 
| 
 SS =264.92  | 
 SS =113.23  | 
 SS =580  | 
 G =499  | 
| 
 n =13  | 
 n =13  | 
 n =13  | 
 N =39  | 
| 
 M =16.92  | 
 M =12.46  | 
 M =9  | 
 k =3 P2 Σ k =6644.33  | 
Given all of this informations, the researchers obtained the following results table:
| 
 Source  | 
 SS  | 
 df  | 
 MS  | 
 F  | 
| 
 Between treatments  | 
 410.21  | 
 2  | 
 136.74  | 
 4.7  | 
| 
 Within  | 
 958.15  | 
 36  | 
||
| 
 Between subjects  | 
 259.69  | 
 12  | 
||
| 
 Error  | 
 698.46  | 
 24  | 
 29.1  | 
|
| 
 Total  | 
 1368.36  | 
 38  | 
They also found an effect size of partial − η2 =0.37.
Null Hypothesis H0: The mean number of explosions are equal for all teaching methods.
Alternative Hypothesis Ha: Not all the mean number of explosions are equal for teaching methods.
Critical value of F at df = 2, 36 and 0.05 is 3.26
Since the observed test statistic (4.7) is greater than the critical value (3.26) , we reject H0 and conclude that there is a significant evidence that not all the mean number of explosions are equal for teaching methods.
Mean difference between Teaching as usual and Teaching+examples is 16.92 - 12.46 = 4.46
Mean difference between Teaching as usual and Teaching+safety training is 16.92 - 9 = 7.92
Mean difference between Teaching as usual and Teaching+examples is 12.46 - 9 = 3.46
Since the mean difference between Teaching as usual and Teaching+safety training is greater than Tukey0sHSD value of 5.03, there is significant difference in mean number of explosions between Teaching as usual and Teaching+safety training methods.
partial − η2 =0.37
The percentage of variance of number of explosions explained by the IV is 37%