Comparing is recommended (e.g., comparing input data to data from a different source). It is likely that data from two different sources will not be exactly the same. How should the determination of close enough be made? If the p-value that the null hypothesis is true is less than .05, a determination of close enough can be made. If the means differ by more than 2 standard deviations of the sample mean, a determination of close enough can be made. The model user, possibly with help from the analyst, should set the criteria. If the p-value that the null hypothesis is false is less than 5%, a determination of close enough can be made