In: Statistics and Probability
You found that three grocery stores having the following market shares: A=.35, B=.50, C=.15. You survey 500 people to see if the market shares have changed and you find the data below. Use the Chi-squared goodness of fit test and ?=.05. A—150 B—250 C—100.
(1) The Hypothesis
H0: The market shares is distributed as follows A - 35%, B = 50% and C = 15%.
Ha: The distribution differs from that stated in the null hypothesis.
The Test Statistic
Each Expected value = (% / 100) * N. N = 500
Observed | Expected % | Expected | (O-E)2 | (O-E)2/E | |
Share A | 150 | 35 | 175 | 625 | 3.57140 |
Share B | 250 | 50 | 250 | 0 | 0.00000 |
Share C | 100 | 15 | 75 | 625 | 8.33330 |
Total | 500.00 | 100.00 | 500.00 | 11.905 |
Test = 11.91
The Critical value at = 0.05 (default level), df = n – 1 = 2 is 5.992
The p value at Test = 11.91 is 0.0026
(4) The Decision Rule: If test is > critical, then Reject H0.
If p value is < , Then Reject H0.
(5) The Decision: Since test (11.91) is <> critical (5.992), We Reject H0.
Since p value (0.0026) is < (0.05), We Reject H0.
(6) The Conclusion: There is sufficient evidence at the 95% significance level to conclude that the distribution is different from that as stated (A = 35%, B = 50%, C = 15%).