In: Statistics and Probability
We are interested in studying whether education level affects people's credit default risk (failing to pay their credit debt). A random study was conducted to see the percent of those who default on their credit card was conducted.
Highest level of education attained % defaulted Number of people studied
High School 36% 100
Some College 20% 160
Bachelor's degree 15% 180
Advanced degree 6% 50
Test the null hypothesis of whether the population proportions of credit default rate is the same for all levels of education.
Provide the test statistic value from your analysis and P-value
Conduct chi-squre test using Excel:
H0: p1 = p2 = p3 = p4, Population proportions of credit default rate is the same for all levels of education
H1: At least one pi is different, Population proportions of credit default rate is not the same for all levels of education
Test statistic = 2.73
Degrees of freedom: df = k-1 = 4-1 = 3
Level of significance = 0.05
p-value (Using Excel function CHISQ.DIST.RT(test statistic,df)) = CHISQ.DIST.RT(2.73,3) = 0.435
Since p-value is greater than 0.05, we do not reject he null hypothesis and conclude that p1 = p2 = p3 = p4. Population proportions of credit default rate is the same for all levels of education.