demographic stats for choosing where to live


New Member
Given the vast amount of demographic information on sites like:

Could I could take a subset of categories like Education, Homeownership, and land per person, for example, and optimize for the best place to live?

What I'm wondering is - what branch of statistics is this?

I'd love to make an internet tool for myself and others which allows a user to choose a bunch of different categories and effectively say optimize %of population with college education, minimize high rate of unemployment, maximize person/square mile, etc... There could be a 100 different parameters or so.

But I don't know where to start on the statistics side. Any advice much appreciated. Thanks!


TS Contributor

I don't exactly understand what your on line application would do:p, but regarding statistics, you should start researching about Principal Component Analysis and Factorial Analysis. Both tools could help and will allow you to build indexes, so you can summarize the information of the variables involved.

Hope this helps
Terzi is correct and PCA, cluster analysis etc are the most common methods used to do this. This field is commonly known as geodemographics, or demographic profiling. It is often used in market research, by companies such as Mosaic and Experian. I use this technique in exploratory research, but also map the results with GIS (Geographic Information Systems) software, such as ArcGis. There are heaps of websites with this sort of information, if you Google market segmentation, geodemographics etc etc, you should get some useful information. Try the Australian Bureau of Statistics website ( too, search for SEIFA, this has a useful methodology section. However, keep in mind something called the "ecological fallacy" (Census data is aggregate data, and you cannot infer the nature of individuals from aggregate statistics) - simply, just because a Census District or Suburb has a high proportion of highly educated individuals, doesn't mean your new neighbours will be the same!