Statisticians sometimes report 50 percent confidence intervals, with the margin for sampling error called the probable error. For example, an estimate X of the average useful life of a television is said to have a probable error of 3 years if there is a 0.50 probability that the interval X̅ ± 3 will include the value of the population mean.

Calculate the probable error if a random sample of 25 televisions has an average useful life of 8.2 years with a standard deviation of 2.5 years.