In: Statistics and Probability
A local town is considering a budget proposal that would allocate tax dollars toward the renovation of a new police station. A survey is conducted to measure public opinion concerning the proposal. A total of 150 individuals respond to the survey: 50 who report being republican and 100 who report being democrat. The frequency distribution is as follows:
Opinion
Favor Oppose
Republican |
35 |
15 |
Democrat |
55 |
45 |
- Based on these results, is there a significant difference in the distribution of opinions for republicans versus democrats? Test at the .05 level of significance.
- Compute the phi-coefficient to measure the strength of the relationship.
null hypothesis: Ho: there is no difference in distribution of opinions for republicans versus democrats
alternate hypothesis:Ha: there is a significant difference in the distribution of opinions for republicans versus democrats
degree of freedom(df) =(rows-1)*(columns-1)= | 1 |
for 1 df and 0.05 level of significance critical region χ2= | 3.841 |
applying chi square test:
Expected | Ei=row total*column total/grand total | Favour | Oppose | Total |
Republican | 30.00 | 20.00 | 50 | |
Democrats | 60.00 | 40.00 | 100 | |
total | 90 | 60 | 150 | |
chi square χ2 | =(Oi-Ei)2/Ei | Favour | Oppose | Total |
Republican | 0.8333 | 1.2500 | 2.083 | |
Democrats | 0.4167 | 0.6250 | 1.042 | |
total | 1.250 | 1.875 | 3.125 |
as test statistic 3.125 is not higher than critical value; we fail to reject null hypothesis
we do not have sufficient evidence to conclude that there is a difference in the distribution of opinions for republicans versus democrats
b)
phi coefficient =sqrt(X2/n)=sqrt(3.125/150)=0.14