In: Statistics and Probability
A? government's economic analysis branch recently released figures that they claimed showed a growth spurt in the western region of a certain country that is divided into 51
states. Using the accompanying? table, where the cell values are numbers of? states, determine if the percent change in real GDP by state was independent of region of the country. Use alpha?=0.05.
GDP? % |
Change |
|||||
Region |
Top? 40% |
Bottom? 60% |
Total |
|||
West |
11 |
4 |
15 |
|||
Midwest |
4 |
8 |
12 |
|||
Southeast |
4 |
8 |
12 |
|||
Northeast |
2 |
10 |
12 |
|||
Total |
21 |
30 |
51 |
What is the test? statistic?
What is p value
H0 : The percent change in real GDP by state was independent of region of the country
H1: The percent change in real GDP by state was not independent of region of the country
Let the los be alpha = 5%
The expected frequencies are
Gender | |||
Top 40% | Bottom 60% | Total | |
West | 6.176 | 8.824 | 15 |
Midwest | 4.9412 | 7.0588 | 12 |
Southeast | 4.941 | 7.059 | 12 |
Northeast | 4.941 | 7.059 | 12 |
Total | 20.9992 | 30.0008 | 51 |
The chisquare contribution values are
Contribution | ||
Oi | Ei | (Oi-Ei)^2 /Ei |
11 | 6.176 | 3.768 |
4 | 8.824 | 2.6372 |
4 | 4.9412 | 0.1793 |
8 | 7.0588 | 0.1255 |
4 | 4.941 | 0.1792 |
8 | 7.059 | 0.1254 |
2 | 4.941 | 1.7506 |
10 | 7.059 | 1.2253 |
Total: | 9.9905 |
Test Statistic, X^2: 9.99
Critical X^2: 7.814736
P-Value: 0.0187
Here chsiquare value > chsiquare critical value and P-value < alpha 0.05 so we reject H0
Thus we conclude that the percent change in real GDP by state was not independent of region of the country