In: Statistics and Probability
A one pound gold-plated tin of Almas caviar costs $12,000. In a blind taste test, 212 food experts tasted four premium caviars, with the most expensive caviar being Almas. Each taster was asked to select the Almas and the observed outcomes are below.
Observed outcomes: Brand A: 10, Brand B: 18, Brand C: 34, Almas: 140
Do a goodness-of-fit analysis to test the hypothesis that the distribution of the responses is not typical (that is to say, tasters could tell the difference). Ho: _________________________________________________________________ Ha: _________________________________________________________________ df = __________ X squared = _________ 0.05= significance level p = ___________
Circle one: Fail to Reject Ho OR Reject Ho
Conclusion: __________________________________________________________ ____________________________________________________________________
Category | Observed Frequency (O) | Proportion, p | Expected Frequency (E) | (O-E)²/E |
A | 10 | 0.25 | 202 * 0.25 = 50.5 | (10 - 50.5)²/50.5 = 32.4802 |
B | 18 | 0.25 | 202 * 0.25 = 50.5 | (18 - 50.5)²/50.5 = 20.9158 |
C | 34 | 0.25 | 202 * 0.25 = 50.5 | (34 - 50.5)²/50.5 = 5.3911 |
Almas | 140 | 0.25 | 202 * 0.25 = 50.5 | (140 - 50.5)²/50.5 = 158.6188 |
Total | 202 | 1.00 | 202 | 217.4059 |
Null and Alternative hypothesis:
Ho: Proportions are same.
H1: Proportions are different.
Test statistic:
χ² = ∑ ((O-E)²/E) = 217.4059
df = n-1 = 3
p-value = CHISQ.DIST.RT(217.4059, 3) = 0.000
Decision:
p-value < α, Reject the null hypothesis
There is enough evidence to conclude that the distribution of the responses is not typical that is the tasters could tell the difference at 0.05 significance level.