Abstract: The statistical difference between the amount of recycled resources and the amount of collected waste in statistical caliber had caused difficulties in assessing the performance of waste separation. The mega-city Shenzhen was taken as an example， conducted a stratified sampling survey on the renewable resources recovery points. Through field survey and account checking of 170 recycling points， it was estimated that the recycling amount of renewable resources in Shenzhen was 2.707×106 t/a. The category of recycled renewable resources presented the market selection phenomenon of “grab the big profit and not collect the small profit”. The recycling points in Shenzhen were mainly located outside the original boundary of specific area. The area of recycling points and the resident population were the key factors that limited the amount of renewable resources. By comparing with the national and domestic first tier cities， it was confirmed that the indistinguishable sources of renewable resources recovery were the important reasons for the difference in the statistical caliber of the so-called “two networks”. The per capita recycling amount of renewable resources，the recycling amount per unit area of recycling points and other supporting indicators were proposed for improving the evaluation index system of waste classification and integration of “two networks”.

Key words: waste separation, market selection, integration of “two networks”, statistical caliber, evaluating index