A STATISTICAL-ANALYSIS OF STATIONARY SOURCE COMPLIANCE TEST AUDIT DATA

The U.S. Environmental Protection Agency (EPA) provides audit material s to organizations conducting compliance tests using EPA Test Methods 6 (SO2), 7 (NOx), 18 (organics by GC), 25 (organics as ppm C), 106 (vi nyl chloride), and 26 (HCl), and to those organizations conducting tri al burn testing for toxic volatile organics under the Resource Conserv ation and Recovery Act (Method 0030). These audit samples must be anal yzed and the results reported to the regulatory agency along with the compliance test results. Each regulation specifies accuracy (percent b ias) limits that must be achieved on the audit samples. Failure to mee t these accuracy limits may invalidate the compliance test results. A statistical analysis was done on the results from 4,321 compliance aud its to determine if the limits will be achieved by most organizations. This analysis determined that they are likely to be achieved more tha n 90% of the time for Methods 6, 7, and 26, and also for some of the o rganics measured by Methods 18 and 0030. However, they are not likely to be achieved even 50% of the time for Method 25 and for many Method 18 measurements. No decision could be made for Method 106 because of i nsufficient data.