Indexes are widely used in geodemographics, because they provide an easy embedded reference point which is the national average. At AGS, we use indexes for –

  • CrimeRisk
  • WeatherRisk
  • QuakeRisk

Absolute values are not necessarily meaningful to the user. Indexes are also widely used in AGS’ Panorama segmentation analysis, where we compare the frequency of a particular segment in a client sample to the local population. This allows us to benchmark the relative attractiveness of a store or product to each lifestyle segment, which can then be used for location or direct marketing targeting.

For CrimeRisk, 100 is the national average, usually population weighted so that the value of the variable is equal to 100 for the United States as a whole. Distributions tend to be highly positively skewed, especially on crime indexes where values of 20 times the national average are not uncommon. Below is a sample distribution from CrimeRisk.

 

 

 

 

 

 

 

 

A value of 50 means that half of the national average value, and a value of 200 means twice the national average. Or expressed as a percentage, the value of 75 is 75% of the national average, 125 is 125% of the national average. This distribution provides an easy reference to the average because it is a known point.

If these are not desired, any value can be reclassified by using the CLASS() function, which allows the user to specify the number of classes and their boundaries  For example, a crime risk index could be converted to a 1 to 5 scale using the function CRIMECLASS=CLASS(5,100,120,200,500) which would set the upper bounds of the first class to 100, and so on.