Using proportional overlap when analyzing your data
| || |
Hello again MapInfoers.
Today I am here to explain how I use the MapInfo Pro Proportional Overlap to do a population analysis in Spain.
Every year, the Spanish Statistics Institute publishes a population register update.
This information is published at a census track level.
Census tracks are defined as a bracket of certain number of voters that live in an area. But people move, cities grow, and there are new development areas, etc . Town halls usually modify this information each year. New areas are created, other areas are split and others just reshaped. The changing areas mean that simple alphanumerical relations (joins) cannot be used to compare population changes.
Image 1: New administrative definition in different years
My objective is to understand where the population growth has happened in these past years. In order to do that I would like to have two different population figures in the same layer. Keeping in mind that the polygons don't match from 2007 and 2013, I am going to have to use the Proportion Overlap capability to achieve this. (Once again, the census tracts are the smallest unit of geography available to me. I don't have the option to aggregate up from smaller boundaries.)
I will compute 2007 population into the 2013 census tracks. Afterwards, I will be able to create in a thematic map the % of variation from 2007 to 2013. In this way I can plan resources if I am public administration or look for expansion sites with new business opportunities if I am a retailer.
The way to pass that information to the 2013 polygons is as follows. This example is using the 64 bit version of MapInfo Professional but the same thing can be done with the 32 bit version.
In the new ribbon interface go to the Table tab and use Update Column to show the well-known dialog box to update data in your tables.
In the dialog box we will make sure we add a new temporary column to our 2013 table. The value of that column will be the proportional sum of the population in 2007. We are adding that proportion when the areas are intersecting.
If your datasets are very large this process might take some time.
Once the operation is complete, I can see where the population is growing and where it is decreasing in this 5 year time period.
Next, I am going to create a thematic with the expression as shown below. The purpose is to have a % ratio growth over the 2013 census :
Now I just need to play a little bit with the ranges in order to have a clear picture of the areas where the population growth might impact my analysis.
As might be expected, between 2007 and 2013 people have left the center of the cities. This has identified potential new development areas, and with the register information for 2013, I have accurate population counts in these new areas for 2013.
Now this data can be used to provide the population with resources and services (schools, stores, etc) to make their lives easier and better.
I hope you find this useful!.
Article by Juan Carlos Cuesta Pérez, EMEA Presales Systems Engineer
When not helping Pitney Bowes customers to be Location Intelligent, Juan Carlos enjoys taking photos, hiking, going to the beach and other outdoor activities with his family.