Results of Simulation Experiments with Data from Microbiological Environmental Monitoring. One hundred values were randomly drawn from the data population, and a control level (i.e., the 99th percentile) was calculated using the different approaches. Then how much of the remaining data population could be covered by these calculated control levels was assessed and compared to the percentile of interest (e.g., if a calculation approach covered 98% of the data population for a 99th percentile of interest, then the error would be 1%). Furthermore, a Chi-squared goodness-of-fit test was performed in order to identify the best-fitting parametric model for each data set, and the coverage of the control levels calculated always using the best-fitting model was assessed (indicated as “Best fit” in the table).

Data for Simulation Derived from:Normal DistributionPoisson DistributionNegative Binomial DistributionZINBGamma DistributionFormula by Hussong and MadsenNon-Parametric PercentileBest fit (Chi-squared test)
Figure 3A1.02%1.00%0.59%0.59%0.45%0.50%0.58%0.59%
Figure 3B2.51%6.38%1.28%1.29%0.95%2.89%0.70%1.11%
Figure 3C1.63%9.24%1.18%1.18%0.49%4.33%0.47%0.49%
Figure 4A1.20%18.20%1.00%0.81%0.92%2.18%1.34%0.92%
Figure 4B2.38%20.91%0.64%1.51%0.72%10.11%0.68%0.68%
Figure 4C1.72%24.02%0.40%0.65%0.55%12.81%0.97%0.43%
Average individual models1.74%13.29%0.85%1.01%0.68%5.47%0.79%0.70%