An article at Mobilize.org reported that the median credit-card balance for undergraduate students was \(\$ 1770\) for those who carried a balance from month to month. A professor at a community college believes that the median credit-card balance of students at his college is different than \(\$ 1770\). To test this hypothesis, he obtains a random sample of 20 students enrolled at the college who carry a creditcard balance from month to month and asks them to disclose their credit-card debt. The results of the survey are presented in Table 3 in dollars. Do the data indicate that the median credit-card debt of students at the professor's college differs from \(\$ 1770\) at the \(\alpha=0.05\) level of significance?

Approach A normal probability plot indicates that the sample data do not come from a population that is normal, and the sample size is small, so parametric procedures for testing the mean cannot be used. However, the data were randomly selected, so we can use the nonparametric sign test for the median.