A random sample of 80 Tempe city employees average 13.1 years of education. A random sample of 75 Mesa city employees average 13.3 years of education. Which test should we use to determine whether the difference between Mesa and Tempe is statistically significant? Why? Should it be one or two-tailed?