A one-sided confidence interval quantifies our knowledge about the true population mean by bounding the range of likely values on one side of the sample mean. For example, after taking 25 random soil samples from a survey unit and measuring the lead concentrations, we might calculate a one-sided upper 90% confidence interval as 0.5 ppm on a sample mean of 11.9 ppm. The one-sided upper 90% confidence bound for this example is then 12.4 ppm. In general, use a one-sided confidence interval instead of a two-sided confidence interval to obtain the tightest upper (lower) bound on a sample mean.