Confidence Interval
Parametric Confidence Interval MCL Test
For MCL/ACL Comparisons
See Also:
Land's Formula Confidence Interval
Two-Sample Test of Proportions
Description:
The parametric Confidence Interval Test provides a method to compare the concentrations of parameters in a well to a comparison level such as a Maximum Contaminant Level (MCL) or Alternate Contaminant Level (ACL).
This method follows the procedures described in the USEPA 1989 document "Statistical Analysis of Ground Water Monitoring Data at RCRA Facilities, Interim Final Guidance," page 6-3.
Use:
The Confidence interval is useful for intra-well comparisons. No background monitoring location is required. A comparison level such as an ACL/MCL must be provided for the parameter. For the parametric confidence interval, data should be normally distributed. If transformed data are used, the transformed data should follow a normal distribution, and the interval compared to the transformed comparison level. ChemStat will automatically transform the comparison level when transforming values.
Remarks:
The parametric confidence interval is presented in detail in the 1989 guidance, along with two example data sets, that are included with ChemStat.
The 1992 guidance indicates that confidence intervals should not be used to compare to an MCL because it is possible for samples to exceed the MCL, and yet the mean of confidence limit would still fall below the MCL, indicating no contamination, when in fact, regulatory limits had been exceeded. The 1992 guidance also points out that unlike tolerance intervals, confidence intervals do not indicate how often individual samples will exceed the MCL.