By Teresa Sinopoli

This study uses 38 variables that describe the 170 retail strips in Metro Toronto as inputs to a cluster analysis.  The cluster analysis  uses the information to group together the retail strips systematically.  Each strip is assigned to the cluster analysis whose average value that it most resembles.  The procedure is iterated to evaluate many different alternatives.  The cluster process was halted when the 170 strips were grouped into ten optimal clusters.

The first part of the paper discusses the concept of retail strips and the underlying argument of the study.  The data and methodology are then described in detail.  The second part describes the retail strips -- first in aggregate; then in terms of the variation among them.  The final section introduces the results of the cluster analysis, in general and in terms of the specific clusters.