In: Statistics and Probability
Source | Df | ss | ms | f stat |
housing | 0.8088 | |||
Wtclass | 0.06 | |||
interaction | 0.3078 | |||
error total |
4.2133 |
na |
na na |
(15 points) If in the previous example, the researchers also determine the weight class (medium and large) for each egg. 3 out 6 eggs in each housing system are medium and the other 3 are large.
Source DF SS MS F-Stat
HOUSING 0.8088
WTCLASS 0.06
Interaction 0.3078
Error N.A.
Total 4.2133 N.A. N.A.
Fill in the ANOVA table (total 13 missing values)
Are there main effects here? Do hypothesis testing at alpha=0.05. Does your conclusion
consistent with question 2?
Is there evidence of interaction in the model? Do hypothesis testing at alpha=0.05. If
there is interaction, give an example of possible interaction. If no interaction, explain what “no interaction” means
Source | DF | SS | MS | F-Stat |
HOUSING | 3 | 2.4264 | 0.8088 | 16.10554 |
WTCLASS | 1 | 0.06 | 0.06 | 1.194773 |
Interaction | 3 | 0.9234 | 0.3078 | |
Error | 16 | 0.8035 | 0.05021875 | NA |
Total | 23 | 4.2133 | NA | NA |
Level of HOUSING, a = 4
Level of WTCLASS, b = 2
DF for HOUSING = a - 1 = 4 - 1 = 3
DF for WTCLASS = b-1 = 2 - 1 = 1
DF for Interaction = (a - 1) * (b-1) = (4-1) * (2 - 1) = 3
DF Total = Number of observations - 1 = 6 * 4 - 1 = 23
DF Error = DF Total - (DF HOUSING + DF WTCLASS + DF Interaction)
= 23 - (3 + 1 + 3) = 16
SS HOUSING = MS HOUSING * DF HOUSING = 0.8088 * 3 = 2.4264
SS Interaction = MS Interaction * DF Interaction = 0.3078 * 3 = 0.9234
SS Error = SS Total - (SS HOUSING + SS WTCLASS + SS Interaction)
= 4.2133 - (2.4264 + 0.06 + 0.9234)
= 0.8035
MS Error = SS Error / DF Error
= 0.8035 / 16
= 0.05021875
MS WTCLASS = SS WTCLASS / DF WTCLASS
= 0.06 / 1
= 0.06
F HOUSING = MS HOUSING / MS Error = 0.8088 / 0.05021875 = 16.10554
F WTCLASS = MS WTCLASS / MS Error = 0.06 / 0.05021875 = 1.194773
F Interaction = MS Interaction / MS Error = 0.3078 / 0.05021875 = 6.129185
Null Hypothesis H0: There is no main effect of housing on nutritional content.
Alternative Hypothesis Ha: There is main effect of housing on nutritional content.
Critical value of F at df = 3, 16 and alpha=0.05 = 3.24
Since the observed F (16.10554) is greater than the critical value, we reject the null hypothesis H0 and conclude that there is significant main effect of housing on nutritional content.
Null Hypothesis H0: There is no main effect of wtclass on nutritional content.
Alternative Hypothesis Ha: There is main effect of wtclass on nutritional content.
Critical value of F at df = 1, 16 and alpha=0.05 = 4.49
Since the observed F (1.194773) is less than the critical value, we fail to reject the null hypothesis H0 and conclude that there is no significant main effect of wtclass on nutritional content.
Null Hypothesis H0: There is no interaction between housing and wtclass on nutritional content.
Alternative Hypothesis Ha: There is an interaction between housing and wtclass on nutritional content.
Critical value of F at df = 3, 16 and alpha=0.05 = 3.24
Since the observed F (6.129185) is greater than the critical value, we reject the null hypothesis H0 and conclude that there is significant interaction between housing and wtclass on nutritional content.