Funnel plots for count data in Tableau – brute force approach

Many years ago, my former employer (and precursor to Public Health England or PHE) – the Association of Public Health Observatories, released a series of Excel spreadsheets in which the basic statistical calculations for various funnel plots were described and demonstrated.  You can still find these resources online (proportions, rates and DSRsISRs and SMRs, counts).  If you don’t know what the difference is between a rate, proportion or a count (nor how to identify them), read the accompanying guide here.

Put simply, a funnel plot is a two-axis control chart that takes into account sample size when performing significance testing.  It places a measure of variation on the Y-axis and a measure of population (or sample size) on the X-axis.  The control limits become tighter as the associated population becomes larger – hence the funnel.  This principle can be applied to a variety of measures, albeit with different statistical underpinnings. Read more

Hexmaps – the old fashioned (stupid) way

Some time ago, a colleague had been impressed by the Guardian’s hexmaps of the 2015 general election. He issued a challenge to try and recreate the basic premise using whatever software and techniques we had available and some uncontraversial data (population and deprivation data).

The basic idea of this is that:

  • constituencies are given a block of hexagons, the amount of which is proportionate to their population;
  • those hexablocks are supposed to look like the constituencies themselves
  • the hexablocks are then stuck together and ‘tessalate’ with eachother
  • the shared borders of the constiuency should be reflected by it’s hexablock
  • the overall aggregation of these hexablocks should look recognisably like the shape of it’s parent (in the Guardian’s case, the UK; in my case, Wakefield district)

Read more