Stats Lecture. 13. Chi Square Test
Stats Lecture. 13. Chi Square Test
(a)
H0: There is no difference in the true distributions of response to this question when asked by a male
interviewer and when asked by a female interviewer for subjects like these.
Ha: There is a difference in the true distributions of response to this question when asked by a male
interviewer and when asked by a female interviewer for subjects like these.
Expected Counts and the Chi-Square Test Statistic
Problem:
The table shows the results.
(a) State the appropriate null
and alternative hypotheses.
(b) Show the calculation for the
expected count in the
Male/Yes cell. Then provide
a complete table of expected counts.
(c) Calculate the value of the chi-square test statistic.
(b)
The expected count
for the Male/Yes cell
is (69 · 50)/100 = 34.5.
The rest of the expected
counts are shown in the table.
Expected Counts and the Chi-Square Test Statistic
(c)
(𝟑𝟎 − 𝟑𝟒. 𝟓) 𝟐 (𝟑𝟗 − 𝟑𝟒. 𝟓)𝟐 (𝟖 − 𝟏𝟎)𝟐
𝝌𝟐 = + +⋯+
𝟑𝟒. 𝟓 𝟑𝟒. 𝟓 𝟏𝟎
𝝌𝟐 = 𝟒. 𝟐𝟓
Expected Counts and the Chi-Square Test Statistic
Problem:
(b)
df = (3 – 1)(2 – 1) = 2, 0.05 = 5.991
Using Table C:
The P-value is between 0.10 and 0.15.
Using technology:
2 cal is 4.25 < 2 5.991, it doesn’t fall in critical region area. So, we fail to reject the H0
Expected Counts and the Chi-Square Test Statistic
Problem:
Ha: Anger and heart-disease status are not independent in the population
of people with normal blood pressure.
Tests for Independence: Expected Counts
STATE
H0: There is no difference in the true distributions of opinion about speaking English for residents
of Australia, the U.K., and the U.S.
Ha: There is a difference in the true distributions of opinion about speaking English for residents of
Australia, the U.K., and the U.S.
We’ll use α = 0.05.
The Chi-Square Test for Independence
Problem:
Do these data provide convincing evidence at the α = 0.05 level that the distributions of
opinion about speaking English differ for residents of Australia, the U.K., and the U.S.?
The Chi-Square Test for Independence
Test Statistic
𝟐 𝟐
(𝟔𝟗𝟎 − 𝟕𝟒𝟏. 𝟖) (𝟏𝟏𝟕𝟕 − 𝟏𝟎𝟖𝟑. 𝟏)
𝝌𝟐 = + + ⋯ = 𝟔𝟖. 𝟓𝟕
𝟕𝟒𝟏. 𝟖 𝟏𝟎𝟖𝟑. 𝟏
df = (4 – 1)(3 – 1) = 6
Using Table C: 𝝌𝟐 , df 6, 0.05 = 12.592
Using technology: 2 cal is 68.57 > 𝛘2 tab is 12.592. It fall in critical region area. So, we
Reject the H0.
The Chi-Square Test for Independence