In: Statistics and Probability
I need a step by step answer for the following question:
Four states were randomly selected, and their members in the U.S. House of Representatives (114th Congress) are noted. At a = 0.10, can it be concluded that there is a dependent relationship between the state and the political party affiliation of its representatives?
California | Florida | Illinois | Texas | |
Democrat | 39 | 10 | 10 | 11 |
Republican | 14 | 17 | 8 | 25 |
Ans:
Chi square test for goodness of fit:
H0:There is a Independent relationship between the state and the political party affiliation.
Ha:There is a dependent relationship between the state and the political party affiliation.
Expected count(fe)=row sum*column sum/overall sum
Observed(fo) | |||||
California | Florida | Illinois | Texas | Total | |
Democrat | 39 | 10 | 10 | 11 | 70 |
Republican | 14 | 17 | 8 | 25 | 64 |
Total | 53 | 27 | 18 | 36 | 134 |
Expected(fe) | |||||
California | Florida | Illinois | Texas | Total | |
Democrat | 27.69 | 14.10 | 9.40 | 18.81 | 70 |
Republican | 25.31 | 12.90 | 8.60 | 17.19 | 64 |
Total | 53 | 27 | 18 | 36 | 134 |
Chi square=(fo-fe)^2/fe | |||||
California | Florida | Illinois | Texas | Total | |
Democrat | 4.623 | 1.194 | 0.038 | 3.240 | 9.095 |
Republican | 5.056 | 1.306 | 0.041 | 3.544 | 9.948 |
Total | 9.679 | 2.501 | 0.079 | 6.784 | 19.043 |
Test statistic:
Chi square=19.043
number of rows,r=2
number of columns,c=4
df=(2-1)*(4-1)=3
p-value=CHIDIST(19.043,3)=0.0003
As,p-value<0.10,we reject the null hypothesis.
There is sufficient evidence to conclude that there is a dependent relationship between the state and the political party affiliation.