In: Statistics and Probability
The admissions office at Huge U. believes the geographical distribution of entering students may be changing. over the last 30 years, 45% of the students have come from Midwestern states, 20% from the Southwest, 20% from the coasts, 5% from the South, and 10% from the Northwest. A sample of 250 students from this year's entering class includes 96 from the Midwest, 61 from the Southwest, 45 from the coasts, 17 from the South, and 31 from the Northwest. Does this indicate that the current entering class has a different geographical distribution?
here null hypothesis:Ho: current entering class has same geographical distribution
alternate hypothesis:Ha" current entering class has a different geographical distribution
degree of freedom =categories-1= | 4 |
for 0.05 level and 4 degree of freedom :rejection region = | 9.488 |
applying chi square goodness of fit test: |
relative | observed | Expected | residual | Chi square | |
category | frequency | Oi | Ei=total*p | R2i=(Oi-Ei)/√Ei | R2i=(Oi-Ei)2/Ei |
Midwestern | 0.450 | 96 | 112.50 | -1.56 | 2.420 |
Southwest | 0.200 | 61 | 50.00 | 1.56 | 2.420 |
Coast | 0.200 | 45 | 50.00 | -0.71 | 0.500 |
South | 0.050 | 17 | 12.50 | 1.27 | 1.620 |
Northwest | 0.100 | 31 | 25.00 | 1.20 | 1.440 |
total | 1.000 | 250 | 250 | 8.400 |
as test statistic =8.40 is not in critical region ; we can not reject null hypothesis
we do not have evidence at 0.05 level to conclude that current entering class has a different geographical distribution