Is the difference between the two values due to an unexplainable random error or a systematic error that can be rationalized by a hypothetical model? The significance test is a statistical analysis used to validate whether the difference between two values can be explained by indeterminate errors or not. The null hypothesis assumes that the values compared are the same and any difference stems from indeterminate errors. The alternate hypothesis states that the difference must be real and cannot be explained by indeterminate errors. A significance level denoted by α sets a confidence level condition for the validity of the null hypothesis. The null hypothesis is rejected when values are present beyond the confidence level. The significance test is called one-tailed if rejection occurs for values at only one end of the normal distribution curve. In two-tailed significance tests, the rejection can occur for values falling at either end of the normal distribution curve.