In: Statistics and Probability
A police department released the numbers of calls for the different days of the week during the month ofOctober, as shown in the table to the right. Use a 0.01 significance level to test the claim that the different days of the week have the same frequencies of police calls. What is the fundamental error with this analysis?
Day | Sun | Mon | Tues | We | Thurs | Fri | Sat |
Frequency | 155 | 206 | 220 | 244 | 176 | 215 | 240 |
Calculate the test statistic,chi squaredχ2.
Calculate the P-value.
What is the conclusion for this hypothesis test?
What is the fundamental error with this analysis?
The Test Statistic is calculated as below. Since the frequencies are considered to be the same, each expected frequency = Sum of observed frequency / 7 = 1456/7 = 208
Day | Sun | Mon | Tues | We | Thurs | Fri | Sat | Total |
Frequency | 155 | 206 | 220 | 244 | 176 | 215 | 240 | 1456 |
Expected | 208 | 208 | 208 | 208 | 208 | 208 | 208 | 1456 |
(O - E) | -53 | -2 | 12 | 36 | -32 | 7 | 32 | |
(O - E)2 | 2809 | 4 | 144 | 1296 | 1024 | 49 | 1024 | |
(O - E)2/E | 13.5048 | 0.0192 | 0.6923 | 6.2308 | 4.9231 | 0.2356 | 4.9231 | 30.5289 |
The test statistic, = 30.53 (rounding to 2 decimal places)
The p value for df = n - 1 = 6 for = 30.53 is; p value = 0.0000
Since p value is < Alpha (0.01), Reject H0.
The fundamental error here us that october has 31 days, and therefore there will be 3 days of the week which will have more calls on those days as compared to the other days of the week.