In: Statistics and Probability
Situation:
Family transportation costs are usually higher than most people believe because those costs include car payments, insurance, fuel costs, repairs, parking, and public transportation. We asked twenty randomly selected families in four major cities to use their records to estimate a monthly figure for transportation costs.
Action:
Use the following data and ANOVA to test whether there is a significant difference in monthly transportation costs for families living in these cities. Use significance level of 5%. Discuss the business implications of the findings.
Atlanta | New York | Los Angeles | Chicago |
$850 | $450 | $1050 | $740 |
$680 | $725 | $900 | $650 |
$750 | $500 | $1150 | $875 |
$800 | $375 | $980 | $750 |
$875 | $700 | $800 | $800 |
If you obtain a significant F statistic, run a Post-Hoc Comparison's test on the four means to answer this question. You can select whichever Post-Hoc Comparison procedure you desire. If none exists on your software, do individual t-tests on all pairs of means.
Following is the output of one way ANOVA generated by excel:
The p-value of F test is 0.0004. Since p-value of F test is less than 0.05 so we can conclude that there is a significant difference in monthly transportation costs for families living in these cities.
---------------------------------------------
The Tukey HSD is 2.86 for alpha= 0.05. It is less than for each pair of difference in mean so each pair is significantly different from other.