Laserfiche WebLink
to oiher cities with weekiy collection. Not all of the residents may have been aware of weekly pickup <br />because approximately 20% of the Weekly area residents put their material out every other week. <br />Parficipation <br />•"Before" periad. The strongest participation rates in the "Befare" period were in the two most <br />affluent axeas, the Single�Stream Contrast area at 91% and the Contral area at 85%. The other areas <br />were grouped between 78% and $2%. <br />■"During" t�eriod. In the "During" period the Con#rast area retained the highest participation rate at <br />94% followed by the Two Bin area at 91 %, with the other areas grouped between about $6% and <br />88%. <br />■ Chan�e. The largest change in participation rate was in the Two Bin area, where the participation <br />rate rose by about 13%, while the increase for the other areas ranged from about 2% to7%. The <br />participation rate for the Cont7rol area increased the least, by only 2%; no significant change would be <br />expected in a Control area, and this change is relatively small. <br />Material Composition Sorts and Analysis <br />To compare the types an.d qtzantities of recyclable and non-targeted tnaterials collected in each of the various <br />pilot study areas, the pilot progratx� xncluded detailed material compositian sorts and analysis. As shown in <br />Appendix A, Beck/Krivit developed the connposition analysis categories, the materials composition protocol, <br />data shee#s, and health and safeiy plan, and provided the sfiatistical analysis. The protocol incorporated <br />American Society of Testing and Material (ASTM) Standards. <br />One sort was conducted in the two-month "Before" periad (June-July 2004) and two were conduc#ed in the <br />four-manth "During" period (August-November 2004). The number of sorts was a balance between having <br />sufficient samples for statistical analysis a.nd avaalable staff rime and other resources. Using the Beck/Krivii <br />protocol, City staff coordinated ihe cotnposition sorts, which were conducted at the City public works garage <br />by City staff and waste az�d recycling professionals from the metropolitat� area. For each sample nine <br />categories of recyclable and four categories of non-targeted materials were sorted and weighed. <br />The data frozn the sorting events were entered into R. W. Beck's specially designed statistical model, which <br />uses 90% confidence intervals to measure statistically significant differences (see the analysis in Appendi� A <br />for a further ex�lanation of stanstical issues}. In some cases, data from the pilot a.reas were aggregated to <br />increase the number of samples for statistical analysis: thus, in parts of the analysis data for both the Single- <br />Stream and Single-Stream Contrast areas were aggregated aztd compared with the aggregated data for alI the <br />dual-stream areas. <br />Analysis <br />The statistical analysis had two broad componen.ts, measuring if there was a statistically significant change in <br />pilot areas in the percentages of clifferent recyclable materials, and measuring if there was a statistically <br />significant change in the actual weight of certain materials. The follawing is a summary of key findings <br />frozxx the Beck/Krivit analysis in Appendix D, which includes numcrous comparison tables. <br />5ingle-Streann <br />The following were statistically significant: <br />18 <br />