The influence of changing the method, the sample, or the analyst on the analysis results is studied by altering only one in a pair of experiments. The t-test compares the statistical significance of one mean with another's mean value or with a known value. In a t-test, unpaired data are two independent sets of data from the same source. Paired data are two sets of related data, for instance, two sets from two different methods on a series of samples. The test statistic is derived from the selected confidence interval for a known value. In comparison to the tabulated test statistic value for a given degree of freedom at a chosen significance level, if the former value is within the latter's range, the null hypothesis is accepted. A one-tailed test looks for ONLY an increase or ONLY a decrease, while a two-tailed test looks for a change in either direction.