What is hexagonal binning?

Hexagonal binning (hex bins for short) is a data aggregation technique. It uses an overlay of hexagon shapes rather than geopolitical boundaries to display different levels of data. In the world of GIS and cartography, it's a powerful tool for visualizing a large amount of location-based information in a meaningful way. 

At PiinPoint, we use hex bins to help you interpret demographic data like population, income, ethnicity, age, and households at a high-level.

Display Techniques

Standard Data Aggregation Methods

In the US, demographic data is generally aggregated at the block group level, and in Canada, at the dissemination area (or DA). You can learn more about levels of geography here

While blocks and DAs display the most granular and accessible form of population statistics, they can be difficult to interpret across a large geography because they are all different sizes. This is because DAs/Blocks are normalized; and as such should be relatively equal in the number of people, however not relatively equal in size. This can cause you to misinterpret the data.

What makes hex bins different?

Cartographers use hex bins to represent data equally. All areas are compared using the same sized shapes, so that when measuring one against another, you're comparing apples to apples.

Mapping Examples

Toronto and Total Population

For example, let's say you're looking to find out where there are the most people living in Toronto. Stats Canada has normalized all the DAs in the city, making sure that each one has approximately 400 households:

The problem is, with each DA being a different size, you get a smattering of results when trying to rank them all by Population. There are dark red areas right beside green ones, simply because the size of the shapes are drastically different.

When we switch the aggregation from randomly sized polygons to evenly sized hexagons, we can pull much more meaningful trends across the geography:

We know which areas have more or less of a given demographic because of an area constraint, not a population constraint.  

Las Vegas and Family Households

In Vegas, block groups are very large around the perimeter of the city. Take this block southwest of Summerlin here, which is in the 100th percentile for all blocks in Vegas, at 2,180 Family Households:

Because it covers such a large area, more Family Households are being ranked in that block, leaving you with a heatmap that looks like the one above.

In actual fact, areas like this one here are more interesting, where dozens of smaller blocks rank high, but more importantly cover a small area so we can actually know where the population lies:

Hexagons make an immense difference in this case. This second visualization shows much more accurately where there are Family Households:

Using Hex Bins in PiinPoint

There are two options in PiinPoint for producing a chloropleth map, either by blocks/DAs, or via hex bins.

To create a chloropleth map:

  1. Set your screen resolution to the geography you want to analyze (Note: you must be closer than search level 12. More on search levels here).
  2. Pick some demographics through the Target Market panel's library
  3. Then choose your heatmap style using the Block/Hexagon toggle:

------

🖥 Technical Notes

For any variable that deals with an absolute count (e.g. Total Population, Total Households, etc.), the data is redistributed from the block groups into hex shapes based on its percent overlap with the underlying block groups. 

For example, if a hexagon covers 10% of one block with 100 people in it, and 60% of another block with 100 people in it, then the population for the hexagon would be 70 people. 

On the other hand, if the hexagon is completely filled by a single, larger block with 100 people (say it covers 80% of the block's shape) then the hexagon's population will be 80 people.  

------

Still have questions? For more information, connect with your Customer Success Manager in the Chat tool or email support@piinpoint.com

Did this answer your question?