2013-01-22_PWETC_AgendaPacket - Laserfiche WebLink

Home

Appendix B Statistical Modeling The data from the sorting events were entered into R. W. Beck's specially designed statistical model (Model). The Model has been developed in Microsoft Excel for easy accessibility and use. The Model statistically manipulates the data to calculate the mean, 90% confidence intervals, and standard deviation for individual material categories. In addition, the Model is structured to identify where specific samples could be considered statistical outliers. The mean represents the mathematical average or average percent of material composing the recyclable materials stream. The confidence interval is an expression of accuracy. It provides the upper and lower limits of the actual mean for all the recyclables in each sample, based on the sorting and sampling of the materials. For example, the 90% confidence interval represents that there is a 90% level of confidence that the true population mean falls within the upper and lower bounds of the confidence interval. The 90% confidence interval is the generally accepted standard by the industry for solid waste and recyclable materials composition studies. In general, the more samples that are sorted, the narrower the confidence interval. The narrower the intervals, the less variability in the data. We believe it is critical when comparing the recyclable materials composition results that the confidence intervals are considered along with the mean percentages. In addition, the total quantities of materials collected for each route were compiled during the sorting and sampling process. This data was analyzed separately to identify the mean quantities collected with the various collection methods and the applicable standard deviations. Results of Data Analysis The Project Team compared the results of the following recyclable materials collection methods: • Comparison of the Monday dual stream collection before the pilot started to the Monday single stream collection during the pilot program. • Comparison of the aggregated results of the single stream collection method to the aggregated results of the dual stream collection method. • Comparison of the current dual stream, bi- weekly method (Friday control route) to the dual stream, weekly method (Tuesday). • Comparison of the current dual stream, bi- weekly method (Friday control route) to the dual stream, bi- weekly method with additional education (Wednesday). • Comparison of the current dual stream, bi- weekly method (Friday control route) to the dual stream, bi- weekly method with larger, 22- gallon bins (Thursday). ■ Comparison of the aggregated results of the dual stream, bi- weekly collection method (Wednesday through Friday) before the pilot started to the results during the pilot program. B -4 B1605