In: Statistics and Probability
It is a general belief that in the United States there is a strong relationship between education and smoking: well-educated people are less likely to smoke. Does a similar relationship hold in France? To find out, researchers recorded the level of education and status of a random sample 459 French men aged 20 to 60. The two-way table below displays the data.
Smoking Status |
||||
Education |
Nonsmoker |
Former |
Moderate |
Heavy |
Primary school Secondary school University |
56 37 53 |
54 43 28 |
41 27 36 |
36 32 16 |
a. State the test you plan to use and determine whether the
conditions for test are satisfied.
b. State the hypotheses for this test.
c. At alpha = 0.05, determine whether there is a significant relationship between smoking status and educational level.
a)here as we are testing if smoking status and educational level have any relatioship based on given contingency table
therefore we will apply chi square test of independence.
b)
null hypothesis:Ho: smoking status and educational level are independent
alternate hypothesis: Ha:smoking status and educational level are dependent
c)
degree of freedom =(row-1)*(column-1)=(3-1)*(4-1)=6
for 6 df and 0.05 level of significance rejection region >=12.592
applying chi square test:
Observed | OBS | nonsmoker | former | mderate | heavy | Total |
primary | 56 | 54 | 41 | 36 | 187 | |
secondary | 37 | 43 | 27 | 32 | 139 | |
university | 53 | 28 | 36 | 16 | 133 | |
Total | 146 | 125 | 104 | 84 | 459 | |
Expected | Ei=?row*?column/?total | nonsmoker | former | mderate | heavy | Total |
primary | 59.481 | 50.926 | 42.370 | 34.222 | 187 | |
secondary | 44.214 | 37.854 | 31.495 | 25.438 | 139 | |
university | 42.305 | 36.220 | 30.135 | 24.340 | 133 | |
Total | 146 | 125 | 104 | 84 | 459 | |
chi square | =(Oi-Ei)2/Ei | nonsmoker | former | mderate | heavy | Total |
primary | 0.20 | 0.19 | 0.04 | 0.09 | 0.53 | |
secondary | 1.18 | 0.700 | 0.641 | 1.693 | 4.21 | |
university | 2.70 | 1.866 | 1.141 | 2.858 | 8.57 | |
Total | 4.08 | 2.75 | 1.83 | 4.64 | 13.305 |
here as test statsitic 13.305 lies in rejectioin region we reject null hypothesis
we have sufficient evidence at 0.05 level to conclude that tehre is a significant relationship between smoking status and educational level.