East Massachusetts Geodemographic Classification<br />Stas Sushkov<br />Maria Sushkova<br />
Introduction<br />Geodemographics is the analysis of people by where they live (Sleight, 1996)<br />Geodemographic classification categories neighborhoods based on their socio-economic and lifestyle characteristics.<br />Geodemographic segmentation is a multivariate statistical classification technique for discovering whether the individuals of a population fall into different groups by making quantitative comparisons of multiple characteristics with the assumption that the differences within any group should be less than the differences between groups. (Wikipedia)<br />
Program evaluation </li></ul>Source: adapted from Harris et al. 2005 <br />
ResearchObjectives<br />To build the geodemographic classification of East Massachusetts based on results of 2000 US Census<br />To create the multipurpose basis for further research analysis<br />To develop and test methodology for more comprehensive geodemographic classifications to be elaborated <br />
Research Methodologies <br />The methodology applied for creation of East Massachusetts classification was developed by Dr. Dan Vickers (University of Sheffield) and described in his work “Multi-level Integrated Classification Based on the 2001 Census”. Dr. Vickers has created The National Classification of Census Output Areas for UK, this classification was officially endorsed by the Office for National Statistics (UK). <br />http://www.sasi.group.shef.ac.uk/area_classification/index.html <br />
Clustering Process<br />Selection of cluster objects (operational taxonomic units) <br />Variables selection<br />Variables standardisation<br />Clustering method selection<br />Identification of cluster number<br />Interpretation, <br />testing and mapping of clusters <br />
Classification Inputs<br />The 4,340 block groups of 9 Massachusetts counties were used: Essex, Middlesex, Suffolk, Norfolk, Bristol, Plymouth, Barnstable, Worchester <br />Total number of households: 2,117,000<br />Overall population: 5,510,000<br />
Project Results<br />Various approaches to the classification design were evaluated and more appropriate ones were selected<br />The geodemographic classification was visually represented and tested in ArcGis, Google Earth and Maptube<br />5 super clusters were identified, narratively described and mapped.<br />Geographic database of cluster locations has been created <br />
Mapping Clusters<br />Mapping clusters in ArcGis using full boundaries – each block group belongs to a certain cluster. Although the population clustering is obvious, uneven census sizes gives wrong perception of the population density. Thus low populated areas dominate on the map <br />
Mapping Clusters<br />Mapping clusters in Arcgis using block group centroids represents cluster distribution more accurately and corresponds with population density. <br />
Mapping Clusters<br />Mapping the classification visualizes the geodemographicclusterisation and shows that the population is clustered. For example brown dots are grouped representing one of the cluster areas. <br />The created clusters can also be exported to Google Earth… <br />
Mapping Clusters<br />or Maptube.org – the online resource which allows to create public maps for free. East Massachusetts Geodemographic Classification Map can be easily accessed there.<br />
Naming and Describing Clusters <br />5 clusters of the classification were named:<br />Common City Dwellers<br />City Strugglers<br />Wealthy Suburbs<br />University Students and Downtown Residents<br />Suburban Middle Class<br />
#1 Common City Dwellers<br />Represented by the “average” middle class population residing mainly in high populated metropolitan areas. <br />Close to average household income ($48,000*; mean - $55,600) and educational level (20% with bachelor or higher degree). House value is around $185,000 which is 13% lower than the mean.<br />High percentage (49%) of Common City Dwellers rent their primary residence, in comparison with 38% mean.<br />Within this cluster the share of households with 2+ cars is 30% while the mean is 42%. <br />*Note: Here and further medium household income, medium house value and medium rent are shown based on the values of 2000 year when the US 2000 Census took place.<br />See this cluster 3D on Google Maps<br />
#3 Wealthy Suburbs<br /><ul><li>Represented by people with high medium household income ($82,000; mean - $55,600) and high medium house value ($307,000; mean - $211,000) , who mostly reside in suburban areas.
Low share of black (1%; mean - 6%) and foreign born (8%; mean - 3%) population.
High proportion of households with 2+ cars (64%; mean 49%) and low percentage of households with no cars.
High share of population with bachelor or higher degree (35%; mean - 23%).
Low proportion of one parent households (6%; mean – 14%).
High average rent price - $800 (mean - $680), and higher than average share of population who pays high mortgage and housing costs (8%; mean – 3%).</li></ul>See this cluster area 3D on Google Maps<br />
#4 University Students and Downtown Residents <br />
#4 Downtown Residents and University Students<br /><ul><li>This cluster members mostly resides in vicinity to universities, or in downtown of metropolitan cities.
Extremely high proportion of people do not use a car as a means of transportation to get to work (64%; mean - 29%), instead they take public transit (29%; mean - 10%), use a bicycle or walk (23%; mean - 5%). Low share of households with 2+ cars (12%; mean – 42%).
High percentage of population with bachelor or higher degree (41%; mean 23%). Medium income ($49,000; mean – $55,600), medium house value ($276,000; mean – $211,000).
High average rent - $910 (mean - $680) and low proportion of residence owners (28%; mean 61%).</li></ul>See this cluster area 3D on Google Maps<br />
84% (mean - 72%) use car to go to work, 55% (mean- 42%) of households have 2+ cars. Only 5% of households have no car.
Household median income is almost the same as the mean - $56,000, but average house value ($175,000) is lower than the mean ($211,000).
Large proportion of owner occupied households (78%; mean – 62%).</li></ul>See this cluster area 3D on Google Maps<br />
References<br />Callingham, M. (2005), From areal classification to geodemographics, paper presented at the Demographic User Group Conference, Royal Society, London 10th November 2005.<br />Debenham, J. E. (2002), Understanding Geodemographic Classification: Creating The Building Blocks For An Extension, Working Paper 02/1 School of Geography, University of Leeds [online] http://www.geog.leeds.ac.uk/wpapers/02-1.pdf<br />Harris, R., Sleight, P. and Webber, R. (2005), Geodemographics, GIS and Neighbourhood Targeting, London, Wiley<br />Longley, P. A. (2005), Geographical Information Systems: a renaissance of geodemographics for public service delivery, Progress in Human Geography, 29(1)<br />Sleight, P. (2004) Targeting customers: How to Use Geodemographic and Lifestyle Data in Your Business, Henley-on –Thames, World Advertising Research Centre<br />Vickers, D. (2006) , Multi-level Integrated Classification Based on the 2001 Census, The University of Leeds<br />Webber, R. and Farr, M. (2001) , MOSAIC-From an area classification system to household classification, Journal of Targeting, Measurement and Analysis for Marketing,10(1).<br />