Skip to content
English
Back to platform
  • There are no suggestions because the search field is empty.

Normalization: what it is, how and when to use it

Provide context and make data more comparable with normalization.

What is normalization?

Usually, when you normalize something you're turning it into a fraction. Normalization is the process of taking a count and dividing it by something else in order to make a number more comparable or to put it in context.

When you normalize data, you eliminate the units of measurement for data, enabling you to compare data from different places. 

TL;DR: Normalize data for easier comparisons across geographies or groups, or to provide context and show rates rather than raw numbers.

Example of Normalization

Let’s say you are comparing unemployment numbers in different cities.

To compare Kansas City’s unemployment rate to St. Louis’s, divide the Kansas City unemployment number by the potential working population in the city. This way, instead of showing unemployment as 38,000 people, you get a rate of 3.3%, which can be compared to St. Louis’s 52,000 people at 3.6%.

Raw numbers don’t provide the necessary context for an accurate comparison. While St. Louis may appear to have higher unemployment, the normalized rates reveal the differences aren’t as large as they first seem.

How to Normalize

With mySidewalk Data

Normalization in mySidewalk allows you to easily adjust data for context.

  1. Find Normalization Options:

    1. In Maps: Select data for a layer on your map, then go to Data> Edit Data Styling > More Options to find normalization options.

      In other data components: Select data for your chart, then under the Data tab, click the dropdown under "Normalize by."

  2. Select the normalizer you want to use.

    1. Global normalizers: These normalizers are almost always available: Total Population, Total Households, Total Area (acres), and Total Area (Sq Miles).

    2. Universe normalizers: All data belongs to one or more "universe" - the highest level the data can be rolled up to. In other words, if one piece of data is a slice of the pie, the "universe" is the whole pie.

      1. For example, people in poverty under 18 years old can be part of several universes, or pies: people in poverty, people with poverty status determined, and people under 18 with poverty status determined.

      2. So for some data, mySidewalk may provide additional options that better suit the data. For example, employment data may show "Working Population 16 and Over" as an option.

  3. Set the number format to Ratio as Percent.

    1. Selecting Ratio as Percent multiplies the number by 100. Most of the time, when you normalize you're looking to create a percent. Changing the number format this way takes the fraction you've created by normalizing, and turns it into a percent.

    2. Do not do this if you are looking to find the density of something, such as Total Population per Square Mile.

      1. density_vs_rate

Tip: When working with poverty data, use "population with poverty status determined" instead of "total population." This excludes groups for whom poverty status isn't measured (such as folks who live in college residence halls or military barracks).

​​For more information, visit the Census guidance on poverty measures.

With Uploaded Data

To normalize your uploaded data with mySidewalk data:

  1. Georeference your data during upload to ensure it aligns with a geographic region. Learn more about georeferencing.

  2. Once uploaded and georeferenced, you can apply mySidewalk’s normalization tools to your data.