In: Statistics and Probability
The research department at the home office of New Hampshire Insurance conducts ongoing research on the causes of automobile accidents. A random sample of 400 policies written on single persons revealed 120 had at least one accident in the previous three-year period. Similarly, a sample of 600 policies written on married persons revealed that 150 had been in at least one accident. At the 0.05 significance level, is there a significant difference in the proportions of single and married persons having an accident during a three-year period?
uestion: Please answer the following, only:
Is this mean or proportion problem? Why?
What test should you use? Why?
What is the data that you extract from the problem?
What are the hypotheses?
Note: your answer should be hand written and attached to the response template.
(a).
The given problem is of proportion as the researcher compares proportions of single and married persons that met with an accident during a three-year period instead of their means.
(b).
There are two population proportions are compared:
-One population is for single person that had at least one accident in the previous three-year period.
-Second population is for married couples that had at least one accident in the previous three-year period.
Therefore, the appropriate test for this study is two proportion z-test.
(c).
Sample 1:
The number of single persons had at least one accident in the previous three-year period (x1) is 120.
The total number of policies written on single persons (n1) is 400.
Sample 2:
The number of married persons had at least one accident in the previous three-year period (x2) is 150.
The total number of policies written on married persons (n2) is 600.
(d).
The statistical hypothesis is,
Ho: There is no significant difference in the proportions of single and married persons having an accident during a three-year period.
Ha: There is a significant difference in the proportions of single and married persons having an accident during a three-year period.
In mathematical terms, the statistical hypothesis is,
Ho: p1 = p2
Ha: p1 ≠ p2