In: Statistics and Probability
A recent census report indicated the following percentages for
methods of commuting to work for workers over 15 years of age:
76.6% drive alone, 9.7% carpool, 4.9% use public transportation,
2.8% walk, 1.7% use other forms of transportation, 4.3% work at
home. A random sample of workers yielded the following table of
observed methods of commuting to work.
Method | Alone | Carpool | Public | Walk | Other | Home |
Observed | 355 | 65 | 30 | 20 | 10 | 20 |
Is there sufficient evidence to conclude that the percentages of workers using each type of transportation differ from those in the census report? Compute the test statistic value for this test.
The expected frequency is first obtained here as:
Ei = pi* Total Frequency
These are computed here as:
Method | O_i | E_i | (E_i - O_i)^2/E_i |
Alone | 355 | 383 | 2.047 |
Carpool | 65 | 48.5 | 5.613 |
Public | 30 | 24.5 | 1.235 |
Walk | 20 | 14 | 2.571 |
Other | 10 | 8.5 | 0.265 |
Home | 20 | 21.5 | 0.105 |
500 | 500 | 11.836 |
The chi square test statistic here is computed (using the computations in the last column) here as:
Therefore 11.836 is the test statistic value for the test here.
For n - 1 = 6 - 1 = 5 degrees of freedom, the p-value here is computed from the chi square distribution tables as:
As the p-value here is 0.04 < 0.05 which is the level of significance here. Therefore the test is significant here and we can reject the null hypothesis here and conclude that he percentages of workers using each type of transportation differ from those in the census report at the 5% level of significance.