In: Statistics and Probability
PRIMARY REASON FOR NOT RETURNING | Golden Palm | Palm Royal | Palm Princess |
price | 32 | 27 | 37 |
location | 39 | 13 | 18 |
Accommodations | 25 | 15 | 14 |
Other | 23 | 8 | 8 |
I wanted to know what Kind of Anova do I run for those two tables. and how my anova would look like.
Choose hotel Again | Golden Palm | Palm Royal | Palm Princess |
yes | 133 | 201 | 179 |
No | 32 | 89 | 66 |
1)
Null and Alternative Hypothesis:
H0: µGolden Palm = µPalm Royal = µPalm Princess
H1: Not all Means are equal
Alpha = 0.05
N = 16
n = 4
Degress of Freedom:
dfBetween = a – 1 = 3-1 =2
dfWithin = N-a = 16-3 = 13
dfTotal = N-1 = 16-1 = 15
Critical Values:
Time (dfBetween, dfWithin): (2,13) = 4.26
Decision Rule:
If F is greater than 4.26, reject the null hypothesis
Test Statistics:
SSBetwen = ∑(∑ai)2/n - T2/N = 424.67
SSWithin = ∑(Y)2 - ∑(∑ai)2/n = 824.25
SSTotal = SSBetwen + SSWithin = 1248.92
MS = SS/df
F = MSeffect / MSerror
Hence,
F = 212.33/91.58 = 2.32
Result:
Our F = 2.32, we fail to reject the null hypothesis
Conclusion:
All means are equal.
2)
For Second Question, Chi Square Test would be appropriate.
Null and Alternative Hypothesis:
H0: The two variables (Choice and Hotel) are independent.
Ha: The two variables are associated.
Given,
Observed Values:
Expected Values:
Expected Values are calculated as:
Eij = (Ti * Tj)/N
where,
Ti = Total in ith row
Tj = Total in jth column
N = table grand total
Alpha = 0.05
df = (r-1)*(c-1) = (2-1)*(3-1) = 1*2 = 2
Chi Square Critical = 5.99
Decision Rule:
If Chi Square> Chi Square Critical reject the null hypothesis
Test Statistic:
Chi Square = ∑(Oij – Eij)2/Eij = (133-120.92)2/120.92 + ……………….. + (66-65.45)2/65.45 = 6.86
Result:
Since, Chi Square> Chi Square Critical we reject the null hypothesis.
Conclusion:
Choice and Hotel are associated.