In: Statistics and Probability
During the first few weeks of the new television season, the evening news audience proportions were recorded as: ABC = 31 5, CBS = 34%, NBC = 35%. A sample of 600 homes yielded the following viewing audience data. Channel Number of Homes ABC 150 CBS 200 NBC 250 We want to determine whether or not there has been a significant change in the percentage of viewing audience of the three networks. Conduct the appropriate hypothesis test at an alpha level of .01.
Solution:
Given: During the first few weeks of the new television season, the evening news audience proportions were recorded as:
ABC = 31%,
CBS = 34%,
NBC = 35%.
Sample size = N =600
Channel Number of Homes
ABC : 150
CBS : 200
NBC : 250
We have to test whether or not there has been a significant change in the percentage of viewing audience of the three networks.
Level of significance = 0.01
Step 1) State H0 and H1:
H0: there has not been a significant change in the percentage of viewing audience of the three networks.
Vs
H1: there has been a significant change in the percentage of viewing audience of the three networks.
Step 2) Find test statistic:
We use Chi-square test statistic:
Where
Oi = Observed Counts
Ei =Expected Counts
Thus we need to make following table:
Channel | Oi | Expected Proportions | Ei | Oi2/Ei |
ABC | 150 | 31% | 186 | 120.968 |
CBS | 200 | 34% | 204 | 196.078 |
NBC | 250 | 35% | 210 | 297.619 |
N =600 |
To get Ei, we multiply each Expected Proportions by N =600.
Thus we get:
Step 3) Find Chi-square critical value:
df = k - 1 = 3 - 1 = 2
Level of significance = 0.01
Chi-square critical value = 9.210
Step 4) Decision Rule:
Reject H0, if Chi-square test statistic value > Chi-square critical value = 9.210, otherwise we fail to reject H0.
Since Chi-square test statistic value = > Chi-square critical value = 9.210, we reject H0.
Step 5) Conclusion:
Since we have rejected H0, there has been a significant change in the percentage of viewing audience of the three networks