Cluster Analysis Overview

For purposes of the State of the Cities reports financial characteristics of Washington’s cities and town are analyzed based on grouping cities in “clusters.”

Cities were initially broken out into clusters for the 2005 State of the Cities report. Cities were first separated into two groups based on 2000 U.S. Census defined metro areas and non-metro areas. These Metropolitan Statistical Areas (MSAs) are a geographic entity defined by the federal Office of Management and Budget based on the concept of a core area with a large population nucleus, plus adjacent communities having a high degree of economic and social integration with that core.

A hierarchical cluster analysis was then performed to divide cities into smaller groupings. This analysis used the following variables:

• Population size (U.S. Census 2000)
• Median household income (U.S. Census 2000)
• Assessed property value per capita (Department of Revenue 2002)
• Retail sales tax revenue per capita (State Auditor’s Office, Local Government Financial Reporting System 2002)
• Population growth, 1990-2000 (U.S. Census 1990 and 2000)

Based on the results of the hierarchical cluster analysis, 14 city clusters were formed.

For the 2009 State of the Cities report, the original cluster analysis was validated through discriminant analyses using the original data. These analyses confirmed the original cluster formation and indicate the cluster structure holds up over time.

A second series of discriminant analyses were then performed using updated variables. This provided the ability to identify cities that moved from one cluster to another based on changing fiscal conditions between reports. The updated variables were:

• Population size (Office of Financial Management 2007)
• Property assessed value per capita (Department of Revenue 2007)
• Retail sales tax revenue per capita (Department of Revenue 2007)
• Population growth, 1997-2007 (Office of Financial Management 2007)

No two cities are the same and categorizing these diverse entities can be difficult. However, the clustering of cities based on financial characteristics offers a tool to view basic differences in resources and needs. Although variance exists between the cities within each cluster, it is clear that within the context of the larger city family there are more similarities than differences.

Alphabetical list of all cities and respective cluster membership

List of cities that changed cluster membership from 2005 to 2009 (pdf, 46 kb)