The IRS is interested in knowing whether people who have an accountant prepare their tax returns have fewer errors than people who prepare their own returns. A random sample of 500 people who had their returns professionally prepared reveals that 125 had errors. A random sample of 450 returns of people who prepared their own returns reveals that 128 had errors. Test, at the 1 %

level of significance, the hypothesis that there is no difference between the number of errors for returns prepared by an accountant and the number of errors for returns prepared by the individual.