ANOVA
Direct Method
1. Null Hypothesis: Create Null and Alternative
Hypotheses.
H0: There is NO significant difference between……..
HA: There is a significant difference between………..
Direct Method
2. Variance between samples:
i. Mean of each sample ()
ii. Combined Mean
iii. SSB (Sum of Squares between Samples)
a. Deviations of mean of each sample from Combined Mean
b. Square these deviations
c. Multiply by number of items of relevant sample
SSB = + + +………………+
iv. Mean Sum of Squares between Samples
MSB = SSB / (k-1) , where (k-1) = v1 (degrees of freedom)
Direct Method
3. Variance within Samples
i. Calculate mean of each sample ()
ii. Deviations of each item from mean
iii. Square these deviations and total up
iv. + +…………..+
v. MSW = SSW / (N – k) , where (N – k) is v2 (Degrees of
freedom)
Direct Method
4. ANOVA Table
Source of Sum of Degrees of Variance Variance Ratio
Variation Squares Freedom
Between Samples SSB V1 = k – 1 MSB
F Ratio =
Larger Variance
Within Samples SSW V2 = N – k MSW Smaller Variance
Question: In an experiment conducted on a farm in a certain village of
U.P., the following information was collected regarding the yield (in
Quintals per acre) of 6 plots of rice. Three out of these 6 plots produced
Basmati rice and rest of the three produced Australian rice. Set up an
Analysis Table and find out if the variety differences are significant.
(Use value of F at 5% LOS )
Variety Plot Yield in Quintals
Basmati 10 15 11
Australian 13 12 17
Solution:
• Calculation of Arithmetic Mean
• Combined Mean
= (12 + 14) / 2 = 13
• Variance between Samples
Sum of Squares of Deviations =
Basmati Australian
10 13
15 12
= 36 = 42
11 17