In: Statistics and Probability
An employee of a small software company in Minneapolis bikes to
work during the summer months. He can travel to work using one of
three routes and wonders whether the average commute times (in
minutes) differ between the three routes. He obtains the following
data after traveling each route for one week.
Route 1 | 32 | 35 | 33 | 28 | 35 |
Route 2 | 22 | 24 | 25 | 24 | 22 |
Route 3 | 29 | 30 | 20 | 20 | 27 |
a-1. Construct an ANOVA table. (Round "Sum Sq" to 1 decimal place, "Mean Sq" and "F value" to 2, and round the "p-value" to 4 decimal places.)
ANOVA
Source of Variation | Df | Sum Sq | Mean Sq | F value | Pr(>F) |
---|---|---|---|---|---|
Route | |||||
Residuals |
a-2. At the 5% significance level, do the average commute times differ significantly between the three routes. Assume that commute times are normally distributed.
Yes, since the p-value is less than significance level.
Yes, since the p-value is not less than significance level.
No, since the p-value is less than significance level.
No, since the p-value is not less than significance level.
b. Use Tukey’s HSD method at the 5%
significance level to determine which routes' average times differ.
(Round difference to 1 decimal place, confidence interval
bounds to 2 decimal places, and p-values to 3.)
Population Mean Difference | diff | lwr | upr | p adj | do the average times differ? |
---|---|---|---|---|---|
Route 2 - Route 1 | |||||
Route 3 - Route 1 | |||||
Route 3 - Route 2 |
a-1)
Following table shows the calculations:
Route 1, G1 | Route 2, G2 | Route 3, G3 | |
32 | 22 | 29 | |
35 | 24 | 30 | |
33 | 25 | 20 | |
28 | 24 | 20 | |
35 | 22 | 27 | |
Total | 163 | 117 | 126 |
And
G | G^2 | |
32 | 1024 | |
35 | 1225 | |
33 | 1089 | |
28 | 784 | |
35 | 1225 | |
22 | 484 | |
24 | 576 | |
25 | 625 | |
24 | 576 | |
22 | 484 | |
29 | 841 | |
30 | 900 | |
20 | 400 | |
20 | 400 | |
27 | 729 | |
Total | 406 | 11362 |
So we have
Now
So,
Since there are 5 different groups so we have k=5. Therefore degree of freedoms are:
-------------
Now
F test statistics is
So p-value of the test using excel function "=FDIST(10.55,2,12)" is 0.0023.
a-2)
Since P-value is less than 0.05 so we reject the null
hypothesis. That is on the basis of sample evidence we can conclude
that populations are different.
Correct option:
Yes, since the p-value is less than significance level.
b)
Here we have 3 groups and total number of observations are 15. So degree of freedom is
df= 15-3 = 12
Critical value for , df=12 and k=3 is
So Tukey's HSD will be
The sample means are
The required confidence intervals are:
groups (i-j) | xbari | xbarj | ni | nj | HSD | xbari-xbarj | Lower limit | Upper limit | Significant(Yes/No) |
mu1-mu2 | 32.6 | 23.4 | 5 | 5 | 5.66 | 9.2 | 3.54 | 14.86 | Yes |
mu1-mu3 | 32.6 | 25.2 | 5 | 5 | 5.66 | 7.4 | 1.74 | 13.06 | Yes |
mu2-mu3 | 23.4 | 25.2 | 5 | 5 | 5.66 | -1.8 | -7.46 | 3.86 | No |
--------------------------------
Following is the screen shot of R script:
Following is the output: