Jump to main content.


Statistical Significance

Note: EPA no longer updates this information, but it may be useful as a reference or resource.


The performance data obtained during ERP programs provide the opportunity for regulators to consider whether to adjust compliance assistance or other strategies directed at the sector. ERP involves selecting a random sample of shops, prior to compliance assistance and self-certification, in order to assess baseline performance. Afterwards, agencies select and assess the performance of a second, independent random sample of shops. Regulators can then compare the percentage of shops using certain best practices during baseline inspections to the percentage of shops using these same practices after the self-certification phase is complete.

One of the advantages of applying this statistically based approach is that regulators can use data from these samples of facilities to make inferences about changes in behavior of shops in the entire population of facilities subject to ERP. For those indicators that show a statistically significant change in performance between baseline and follow-up inspections, regulators can be confident (to a certain degree) that the performance of the overall population of shops changed.

On the other hand, for those indicators that do not show a statistically significant change, regulators can only be certain that the shops in the sample made those changes. If a change in performance is not statistically significant, it could still be true that the population's performance changed between baseline and post-certification, but regulators cannot be as certain that this occurred.


Local Navigation


Jump to main content.