Professor: Howard B. Lee
Lecture Notes
Week 11: Chapter 10

ANOVA Summary Table:
| source | degrees of freedom | sum of squares | mean of squares | F |
|---|---|---|---|---|
| between | 2 | 16.3334 | 8.1667 | .5334 |
| within | 21 | 321.5000 | 15.3095 | |
| Total | 23 | 337.8334 |
SS within/ df within = MS within
MS between/ MS within = F
ho: u1 = u2 = u3
h1: u1 not = u2 or, u1 not = u3 or, u2 not = u3,
and at least one is true.
You must find out how many group means you have by looking at the df between. This is the number of levels of IV - 1, and can be found under column "df" row "between" on the ANOVA summary table. For the above example, the number of group means = 3.
Test statistic : F = .5334
Decision rule : alpha = .05
Always treat it as a right-tailed test .
Use table E. (The top value = .05, bottom value (in bold) = alpha = .01)
df between = 2
df within = 21
Reject ho if the calculated F value is greater than or equal to the
critical value found on table E.
Decision rule: Since F = .5334 < 3.47,
do not reject ho.
Conclusion: (This relates back to the original problem)
There is insufficient evidence of a difference between therapies.
If there are many groups:
ho : u1 = u2 = u3 = u4 = u5
h1 : at least one pair of group means are statistically different.
