{"id":250552,"date":"2025-09-22T14:09:32","date_gmt":"2025-09-22T21:09:32","guid":{"rendered":"http:\/\/www.esri.com\/arcgis-blog\/?post_type=blog&p=250552"},"modified":"2025-09-22T17:10:36","modified_gmt":"2025-09-23T00:10:36","slug":"better-breaks-define-your-thematic-maps-purpose","status":"publish","type":"blog","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-online\/mapping\/better-breaks-define-your-thematic-maps-purpose","title":{"rendered":"Better Breaks Define Your Thematic Map\u2019s Purpose"},"author":4161,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"open","ping_status":"closed","template":"","format":"standard","meta":{"_acf_changed":false,"_searchwp_excluded":""},"categories":[22941],"tags":[26451,34821,24581,23391,32661],"industry":[],"product":[36581,36831,36551,37011],"class_list":["post-250552","blog","type-blog","status-publish","format-standard","hentry","category-mapping","tag-cartography","tag-policy-maps","tag-smart-mapping","tag-spatial-analytics","tag-thematic-mapping","product-arcgis-living-atlas","product-js-api-arcgis","product-arcgis-online","product-esri-demographics"],"acf":{"short_description":"The purpose of this blog is to discuss how a typical thematic map of a percentage comes into focus and how you give it purpose.","flexible_content":[{"acf_fc_layout":"content","content":"

Originally published in 2018, now updated to reflect the latest Map Viewer user interface.<\/em><\/p>\n

Because I am a geographer who makes a lot of thematic maps, over time I\u2019ve noticed the key moments in the decision making process that dramatically influence each map. The purpose of this blog is to discuss how a typical thematic map of a percentage comes into focus and how you give it purpose.<\/p>\n

To start, we need data, and an idea of what we want to map. We recently hosted up this layer of U.S. county health rankings data<\/a> from the Robert Wood Johnson Foundation<\/a> and University of Wisconsin Health Institute<\/a> and made a few maps for policy making<\/a> from it. The layer contains dozens of useful measures, each waiting to be turned into useful information in a thematic map.<\/p>\n

The software can map it, but it takes a human to make it meaningful. In this blog we\u2019ll cover how the software (in this case, ArcGIS Online) starts the thematic map, and how a human improves what the software suggests to give the map purpose.<\/p>\n"},{"acf_fc_layout":"content","content":"

Choose a topic<\/h2>\n"},{"acf_fc_layout":"content","content":"

Let\u2019s pick just one subject among the many attributes in this gold mine of data: Percent Low Birth Weight<\/a>. It represents the percentage of all births in a county that meet the standard of low birth weight. So we have data.<\/p>\n

We need an idea for the map. It is easy to imagine a choropleth map of the counties, each colored by its Low Birth Weight percent. Pretty straightforward.<\/p>\n

As always, let\u2019s explore the data on the map first, to compare what we know about the subject to what\u2019s on the map, and then make a thematic map of it.<\/p>\n

That first step (exploring the data) is key \u2013 unfortunately a lot of people simply want to get the thematic map done as quickly as possible without thinking critically about the data. They choose a default classification technique, verify that the map shows some variation in colors, and call it a day, when in reality that map is unfinished.<\/p>\n"},{"acf_fc_layout":"blockquote","content":"

How can you tell a thematic map has been rushed into use without a specific purpose?<\/p>\n

1)\tDefault colors, default outlines, default classification settings
\n2)\tThe breaks used to set the colors have no intrinsic meaning \u2013 they are the numbers generated by a classification algorithm.
\n3)\tThe colors have not been chosen to emphasize the interesting part of the data.
\n4)\tThe legend contains unnecessary levels of precision<\/p>\n"},{"acf_fc_layout":"content","content":"

Open this web map<\/a> which uses layers found in ArcGIS Living Atlas, and click on the “Layers” button in the top left to see the list of layers available. Turn off the layer “Low Birth Weight % – Above and Below theme (recommended)” and turn on the layer at the bottom of the list titled \u201cCounty Health Rankings 2018.\u201d<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":2940612,"id":2940612,"title":"Screenshot 2025-09-22 at 10.14.27\u202fAM","filename":"Screenshot-2025-09-22-at-10.14.27-AM.png","filesize":581380,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.14.27-AM.png","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-online\/mapping\/better-breaks-define-your-thematic-maps-purpose\/screenshot-2025-09-22-at-10-14-27-am","alt":"Turn on the last layer in the Layers list, called \"County Health Rankings 2018\" and the map should show a purple layer of counties.","author":"4161","description":"Turn on the last layer in the Layers list, called \"County Health Rankings 2018\" and the map should show a purple layer of counties.","caption":"Turn on the last layer in the Layers list, called \"County Health Rankings 2018\" and the map should show a purple layer of counties.","name":"screenshot-2025-09-22-at-10-14-27-am","status":"inherit","uploaded_to":250552,"date":"2025-09-22 17:14:47","modified":"2025-09-22 17:16:35","menu_order":0,"mime_type":"image\/png","type":"image","subtype":"png","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":1927,"height":839,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.14.27-AM-213x200.png","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.14.27-AM.png","medium-width":464,"medium-height":202,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.14.27-AM.png","medium_large-width":768,"medium_large-height":334,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.14.27-AM.png","large-width":1920,"large-height":836,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.14.27-AM-1536x669.png","1536x1536-width":1536,"1536x1536-height":669,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.14.27-AM.png","2048x2048-width":1927,"2048x2048-height":839,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.14.27-AM-826x360.png","card_image-width":826,"card_image-height":360,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.14.27-AM-1920x836.png","wide_image-width":1920,"wide_image-height":836}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

Click the three dots to the right of the layer name “County Health Rankings 2018” and choose the “Rename” option. Rename the layer to \u201cLow Birth Weight.\u201d Next, click on the layer to select it, and then click on “Styles” in the top right of the user interface\u00a0 and choose the \u201cChange Style\u201d button on the layer to explore the Percent Low Birth Weight attribute.<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":2940616,"id":2940616,"title":"Screenshot 2025-09-22 at 10.20.10\u202fAM","filename":"Screenshot-2025-09-22-at-10.20.10-AM.png","filesize":544771,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.20.10-AM.png","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-online\/mapping\/better-breaks-define-your-thematic-maps-purpose\/screenshot-2025-09-22-at-10-20-10-am","alt":"Click on the name of the layer to select the layer, and then click \"Styles\" at the top right to change the layer's style.","author":"4161","description":"Click on the name of the layer to select the layer, and then click \"Styles\" at the top right to change the layer's style.","caption":"Click on the name of the layer to select the layer, and then click \"Styles\" at the top right to change the layer's style.","name":"screenshot-2025-09-22-at-10-20-10-am","status":"inherit","uploaded_to":250552,"date":"2025-09-22 17:23:11","modified":"2025-09-22 17:24:05","menu_order":0,"mime_type":"image\/png","type":"image","subtype":"png","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":1927,"height":838,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.20.10-AM-213x200.png","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.20.10-AM.png","medium-width":464,"medium-height":202,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.20.10-AM.png","medium_large-width":768,"medium_large-height":334,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.20.10-AM.png","large-width":1920,"large-height":835,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.20.10-AM-1536x668.png","1536x1536-width":1536,"1536x1536-height":668,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.20.10-AM.png","2048x2048-width":1927,"2048x2048-height":838,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.20.10-AM-826x359.png","card_image-width":826,"card_image-height":359,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.20.10-AM-1920x835.png","wide_image-width":1920,"wide_image-height":835}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

Next, click the “+ Field” button and search for the attribute \u201cPercent Low Birthweight.” Click the checkmark next to it and hit “Add” to see it on the map. Map Viewer evaluates the field you chose, and makes a suggestion for what map style to use. In this case, Map Viewer sees that the attribute is a rate or percent so it suggests the Counts and Amounts (Color) style of map<\/a>. This style shades applies a color to each county, based on the value found in the \u201cPercent Low Birthweight\u201d attribute for that county.<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":2940619,"id":2940619,"title":"Screenshot 2025-09-22 at 10.30.23\u202fAM","filename":"Screenshot-2025-09-22-at-10.30.23-AM.png","filesize":519792,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.30.23-AM.png","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-online\/mapping\/better-breaks-define-your-thematic-maps-purpose\/screenshot-2025-09-22-at-10-30-23-am","alt":"The \"Counts and Amounts (color)\" style is meant for use with a numeric attribute that represents a percent, rate, index or other normalized value, and shades each feature based its attribute value.","author":"4161","description":"The \"Counts and Amounts (color)\" style is meant for use with a numeric attribute that represents a percent, rate, index or other normalized value, and shades each feature based its attribute value.","caption":"The \"Counts and Amounts (color)\" style is meant for use with a numeric attribute that represents a percent, rate, index or other normalized value, and shades each feature based its attribute value.","name":"screenshot-2025-09-22-at-10-30-23-am","status":"inherit","uploaded_to":250552,"date":"2025-09-22 17:30:38","modified":"2025-09-22 17:32:28","menu_order":0,"mime_type":"image\/png","type":"image","subtype":"png","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":1384,"height":778,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.30.23-AM-213x200.png","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.30.23-AM.png","medium-width":464,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.30.23-AM.png","medium_large-width":768,"medium_large-height":432,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.30.23-AM.png","large-width":1384,"large-height":778,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.30.23-AM.png","1536x1536-width":1384,"1536x1536-height":778,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.30.23-AM.png","2048x2048-width":1384,"2048x2048-height":778,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.30.23-AM-826x465.png","card_image-width":826,"card_image-height":465,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.30.23-AM.png","wide_image-width":1384,"wide_image-height":778}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

Click \u201cTheme\u201d to explore this data a bit using some settings that decide how the data will be shaded.<\/p>\n

High to Low theme<\/h2>\n

This is where ArcGIS Online saves you time and makes you a better mapmaker by encapsulating good mapping techniques as “themes” within each map style. So far, all you did was touch an attribute, and the map lights up with a suggested \u201cHigh to Low\u201d theme using a very light blue to dark blue color ramp.<\/p>\n

Click “Style options” to see the histogram of the data. Understanding the histogram in relation to the colors applied to the data is how you take control of which story this map is going to tell.<\/strong><\/p>\n

 <\/p>\n"},{"acf_fc_layout":"image","image":{"ID":2940620,"id":2940620,"title":"Screenshot 2025-09-22 at 10.38.47\u202fAM","filename":"Screenshot-2025-09-22-at-10.38.47-AM.png","filesize":475717,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.38.47-AM.png","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-online\/mapping\/better-breaks-define-your-thematic-maps-purpose\/screenshot-2025-09-22-at-10-38-47-am","alt":"When you make a thematic map like this, you are using color to indicate variation in the data you are mapping. The histogram shows how color is being applied across the range of data found.","author":"4161","description":"When you make a thematic map like this, you are using color to indicate variation in the data you are mapping. The histogram shows how color is being applied across the range of data found.","caption":"When you make a thematic map like this, you are using color to indicate variation in the data you are mapping. The histogram shows how color is being applied across the range of data found.","name":"screenshot-2025-09-22-at-10-38-47-am","status":"inherit","uploaded_to":250552,"date":"2025-09-22 17:39:03","modified":"2025-09-22 17:40:34","menu_order":0,"mime_type":"image\/png","type":"image","subtype":"png","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":1382,"height":784,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.38.47-AM-213x200.png","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.38.47-AM.png","medium-width":460,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.38.47-AM.png","medium_large-width":768,"medium_large-height":436,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.38.47-AM.png","large-width":1382,"large-height":784,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.38.47-AM.png","1536x1536-width":1382,"1536x1536-height":784,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.38.47-AM.png","2048x2048-width":1382,"2048x2048-height":784,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.38.47-AM-820x465.png","card_image-width":820,"card_image-height":465,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.38.47-AM.png","wide_image-width":1382,"wide_image-height":784}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

ArcGIS Online shows you the color ramp next to a histogram of the data. For the \u201cHigh to Low\u201d theme, the little handles on the left side of the histogram indicate at what values dark blue or light blue are applied. In this case, counties with 10% low birth weight or higher are given a full dark blue color. Counties with 6.1% or lower are given a very light blue color. These extreme values are not the main story in this map style.<\/p>\n

Values between 10.0 and 6.1 are shaded a color somewhere between dark blue and very light blue, depending on where the attribute values fall. Sometimes referred to as \u201cunclassed\u201d or \u201ccontinuous\u201d color, it\u2019s value is that you get an overall pattern on the map, and you can see how neighboring counties vary relative to one another. You can see the variation in the data because this style emphasizes variation between the two handles (10 and 6.1 in this example).<\/p>\n

Where did these values come from? They are 1 standard deviation above the mean (10.0) and below the mean (6.1). Looking at the legend or the histogram, you can see that the mean is 8.1 for this data set. (Note: this is the average of the data, not necessarily the true national average, because counties vary widely in population from hundreds to millions.)<\/p>\n

At this point, I always go search the documentation or online for what the literature has to say about this subject. In this case, the source data did not provide the national average for percent low birth weight, but a broader search found several indications that 8.1% happens to be the national average for this topic. This is useful information to have as you think about how to style this map.<\/p>\n

This default is just a starting point, it is NOT the one-size-fits-all solution for making maps.<\/strong><\/span> It is a useful map style for initial exploration of the data, so that you can ask yourself: \u201cWhat part of this data is interesting?\u201d<\/p>\n

From the histogram of the data we see a pretty normal bell curve, with a little skew toward higher values.<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":2940621,"id":2940621,"title":"Screenshot 2025-09-22 at 10.46.57\u202fAM","filename":"Screenshot-2025-09-22-at-10.46.57-AM.png","filesize":40228,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.46.57-AM.png","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-online\/mapping\/better-breaks-define-your-thematic-maps-purpose\/screenshot-2025-09-22-at-10-46-57-am","alt":"The histogram shows how color varies in between two handles.","author":"4161","description":"The histogram shows how color varies in between two handles.","caption":"The histogram shows how color varies in between two handles.","name":"screenshot-2025-09-22-at-10-46-57-am","status":"inherit","uploaded_to":250552,"date":"2025-09-22 17:47:10","modified":"2025-09-22 17:48:05","menu_order":0,"mime_type":"image\/png","type":"image","subtype":"png","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":373,"height":626,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.46.57-AM-213x200.png","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.46.57-AM.png","medium-width":156,"medium-height":261,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.46.57-AM.png","medium_large-width":373,"medium_large-height":626,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.46.57-AM.png","large-width":373,"large-height":626,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.46.57-AM.png","1536x1536-width":373,"1536x1536-height":626,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.46.57-AM.png","2048x2048-width":373,"2048x2048-height":626,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.46.57-AM-277x465.png","card_image-width":277,"card_image-height":465,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.46.57-AM.png","wide_image-width":373,"wide_image-height":626}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

A color ramp that has only one color biases the eye toward only the high values in the data. The darker colors are applied to the higher values, but even the middle of the color ramp (near the 8.1% national average) is already leaning blue\u2026 so if the story needs to focus mainly on areas where Low Birth Weights are happening at unusually high rates, the High to Low theme is a good option. Those counties who are doing well on this topic are deemphasized in a High to Low theme when a color ramp that uses only one color is used (e.g. all shades of blue).<\/p>\n

High to Low theme does not really care about a national average or mean, unless you adjust a break to use such a figure.<\/p>\n

Let\u2019s explore the same data using classification, to see where it starts the map.<\/p>\n

Natural Breaks<\/h2>\n

Here\u2019s the same layer, but with \u201cClassify Data\u201d turned on, which defaults to a Natural Breaks method<\/a>. The darkest color is assigned to values at or above 11.3 so compared to the Unclassed approach above, in this style it is \u201charder\u201d for a county to earn that darkest color. The values between 8.79 and 11.3 all get the same color, as do all values between 6.93 to 8.79 and all values below 6.93.<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":2940623,"id":2940623,"title":"Screenshot 2025-09-22 at 10.52.30\u202fAM","filename":"Screenshot-2025-09-22-at-10.52.30-AM.png","filesize":569827,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.52.30-AM.png","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-online\/mapping\/better-breaks-define-your-thematic-maps-purpose\/screenshot-2025-09-22-at-10-52-30-am","alt":"Classification eliminates of the detailed variation in the data, forcing all counties into one of four classes.","author":"4161","description":"Classification eliminates of the detailed variation in the data, forcing all counties into one of four classes. ","caption":"Classification eliminates of the detailed variation in the data, forcing all counties into one of four classes. ","name":"screenshot-2025-09-22-at-10-52-30-am","status":"inherit","uploaded_to":250552,"date":"2025-09-22 17:52:46","modified":"2025-09-22 17:55:54","menu_order":0,"mime_type":"image\/png","type":"image","subtype":"png","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":1757,"height":843,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.52.30-AM-213x200.png","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.52.30-AM.png","medium-width":464,"medium-height":223,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.52.30-AM.png","medium_large-width":768,"medium_large-height":368,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.52.30-AM.png","large-width":1757,"large-height":843,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.52.30-AM-1536x737.png","1536x1536-width":1536,"1536x1536-height":737,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.52.30-AM.png","2048x2048-width":1757,"2048x2048-height":843,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.52.30-AM-826x396.png","card_image-width":826,"card_image-height":396,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-10.52.30-AM.png","wide_image-width":1757,"wide_image-height":843}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

These breaks are where the Natural Breaks algorithm found a mathematical reason to divide the data up into the four breaks it was told to use. There are now eight different numbers to interpret in this map\u2019s legend (see above), and no explanation of their significance. We see a medium blue color begins at 11.3% – is this to be considered a \u201chigh\u201d rate? It’s not the same dark blue as is found in the Unclassed color ramp.<\/p>\n

In which shade of blue does the national average 8.1% fall into? Unless we adjust a break to use 8.1%, we can\u2019t really speak to that figure effectively on the map. All this map says is that some places have it worse than others, but we have not provided a standard of comparison by which we would leverage the use of color.<\/p>\n

The challenge with this style of map lies in the reality that the breaks have no real-world significance nor any direction relationship with the histogram. It is essentially a cookie-cutter applied over the top of the histogram.<\/p>\n

Here is the same layer, but with 10 natural breaks. It\u2019s essentially the same map, but now the legend is a little more challenging to read and interpret.<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":2940627,"id":2940627,"title":"Screenshot 2025-09-22 at 11.01.41\u202fAM","filename":"Screenshot-2025-09-22-at-11.01.41-AM.png","filesize":575226,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.01.41-AM.png","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-online\/mapping\/better-breaks-define-your-thematic-maps-purpose\/screenshot-2025-09-22-at-11-01-41-am","alt":"Adding more classes seems to add more detail because the range of values in each class gets smaller.","author":"4161","description":"Adding more classes seems to add more detail because the range of values in each class gets smaller.","caption":"Adding more classes seems to add more detail because the range of values in each class gets smaller.","name":"screenshot-2025-09-22-at-11-01-41-am","status":"inherit","uploaded_to":250552,"date":"2025-09-22 18:02:31","modified":"2025-09-22 18:03:14","menu_order":0,"mime_type":"image\/png","type":"image","subtype":"png","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":1758,"height":838,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.01.41-AM-213x200.png","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.01.41-AM.png","medium-width":464,"medium-height":221,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.01.41-AM.png","medium_large-width":768,"medium_large-height":366,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.01.41-AM.png","large-width":1758,"large-height":838,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.01.41-AM-1536x732.png","1536x1536-width":1536,"1536x1536-height":732,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.01.41-AM.png","2048x2048-width":1758,"2048x2048-height":838,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.01.41-AM-826x394.png","card_image-width":826,"card_image-height":394,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.01.41-AM.png","wide_image-width":1758,"wide_image-height":838}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

With 10 classes, the full color ramp is used again, and we can see more detail around those darkest blue counties. But if a legend with 8 numbers for an author to explain and a reader interpret is difficult, a legend with 20 numbers is far more difficult. In the legend above, can you find which class would contain the national average 8.1%, and then find a sample county at or near that average? There are nine shades of blue to choose from, and this legend infers that you should be able to distinguish among them.<\/p>\n

Whether your map has 4 classes or 10 classes or is not classified, the classified legend on a web map is often a poor way for someone to understand the actual value in any single county. Can a county be eligible for funding if they are at 14.39% or above? At this point, because we have not assigned any specific meaning to the classes such as \u201c>14.39 (Eligible for funding)\u201d, the legend is really there to simply orient the user about what color means, generally. Remember, we are making the maps to help connect a specific audience to a specific topic, so the pattern shown needs to relate to real world numbers found in the literature about a topic, current or proposed policies related to the topic, or other relatable numbers derived from comparisons to previous years or other measures.<\/p>\n

A web map lets the user simply click on a feature to learn its value. A pop-up<\/a> can provide the specific value as needed.<\/p>\n

Equal Interval<\/h2>\n

Here\u2019s the same layer, but with \u201cClassify Data\u201d turned on, and now using the Equal Interval method<\/a>. The darkest color is now assigned to values at or above 21 so the effect is that it is very hard for a county to earn that darkest color. The values between 14.9 and 21 all get the same color, as do all values between 8.8 to 14.9 and all values below 8.8. All classification methods are heavily influenced by the min\/max values in the data set, so it’s useful to recognize that when assessing how the software arrived at its class breaks.<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":2940630,"id":2940630,"title":"Screenshot 2025-09-22 at 11.11.31\u202fAM","filename":"Screenshot-2025-09-22-at-11.11.31-AM.png","filesize":549251,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.11.31-AM.png","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-online\/mapping\/better-breaks-define-your-thematic-maps-purpose\/screenshot-2025-09-22-at-11-11-31-am","alt":"Equal interval drains the pattern from most thematic topics because the min and max values have a huge impact on the interval sizes.","author":"4161","description":"Equal interval drains the pattern from most thematic topics because the min and max values have a huge impact on the interval sizes.","caption":"Equal interval drains the pattern from most thematic topics because the min and max values have a huge impact on the interval sizes.","name":"screenshot-2025-09-22-at-11-11-31-am","status":"inherit","uploaded_to":250552,"date":"2025-09-22 18:11:48","modified":"2025-09-22 18:13:49","menu_order":0,"mime_type":"image\/png","type":"image","subtype":"png","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":1765,"height":837,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.11.31-AM-213x200.png","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.11.31-AM.png","medium-width":464,"medium-height":220,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.11.31-AM.png","medium_large-width":768,"medium_large-height":364,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.11.31-AM.png","large-width":1765,"large-height":837,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.11.31-AM-1536x728.png","1536x1536-width":1536,"1536x1536-height":728,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.11.31-AM.png","2048x2048-width":1765,"2048x2048-height":837,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.11.31-AM-826x392.png","card_image-width":826,"card_image-height":392,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-11.11.31-AM.png","wide_image-width":1765,"wide_image-height":837}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

The map now looks very \u201csoft\u201d and the histogram\/color ramp tells why: most of the counties fall within the lowest category. To many people, this map would suggest low birth weights are not really much of a problem anywhere except that one northern Colorado county.<\/p>\n

That\u2019s because equal interval takes the maximum value minus the minimum value in the data, and divides that by the number of classes to set the interval. If the min value were 0, the breaks would shift. If the maximum were not 27 but 270, the breaks would shift, dramatically. Outlier values have a big effect on this option. Note that the national average 8.1% would fall into the lowest category.<\/p>\n

Quantile<\/h2>\n

Here\u2019s the same layer, but with \u201cClassify Data\u201d turned on, and now using the Quantile method<\/a>. Quantile ensures that each color has an equal number of features in it where possible. If you have 1000 features, Quantile stuffs 250 features into each of the four colors in your ramp. Each class has the same number of features no matter what is actually going on with the data. Your map can have two features in two different classes with exactly the same percent low birthweight. The darkest color is now assigned to values at or above 9.08, the values between 7.8 and 9.08 all get the same color, as do all values between 6.7 to 7.8 and all values below 6.7. The national average 8.1% now earns the second darkest blue. Quantile ensures you\u2019ll have lots of colors on the map, but they\u2019ll have no intrinsic meaning for this layer.<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":2940638,"id":2940638,"title":"Screenshot 2025-09-22 at 12.30.54\u202fPM","filename":"Screenshot-2025-09-22-at-12.30.54-PM-scaled.png","filesize":384013,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.30.54-PM-scaled.png","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-online\/mapping\/better-breaks-define-your-thematic-maps-purpose\/screenshot-2025-09-22-at-12-30-54-pm","alt":"The quantile classification method only tries to put an equal number of features in each class.","author":"4161","description":"The quantile classification method only tries to put an equal number of features in each class. ","caption":"The quantile classification method only tries to put an equal number of features in each class. ","name":"screenshot-2025-09-22-at-12-30-54-pm","status":"inherit","uploaded_to":250552,"date":"2025-09-22 19:31:23","modified":"2025-09-22 19:32:28","menu_order":0,"mime_type":"image\/png","type":"image","subtype":"png","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":2560,"height":1344,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.30.54-PM-213x200.png","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.30.54-PM-scaled.png","medium-width":464,"medium-height":244,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.30.54-PM-scaled.png","medium_large-width":768,"medium_large-height":403,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.30.54-PM-scaled.png","large-width":1920,"large-height":1008,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.30.54-PM-1536x806.png","1536x1536-width":1536,"1536x1536-height":806,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.30.54-PM-2048x1075.png","2048x2048-width":2048,"2048x2048-height":1075,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.30.54-PM-826x434.png","card_image-width":826,"card_image-height":434,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.30.54-PM-1920x1008.png","wide_image-width":1920,"wide_image-height":1008}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

Standard Deviation<\/h2>\n

Here\u2019s the same layer, but with \u201cClassify Data\u201d turned on, and now using the Standard Deviation method<\/a>. The colors are “anchored” around the mean of the data (0 Std. Dev.). The darkest colors are now assigned to values above the mean based on which standard deviation class or range they fall in.\u00a0 This is a useful method when trying to get more fine-grained in understanding how quickly your data deviates from the mean on the map. But, the legend is mostly unintelligible to most people, because it no longer shows the actual percentages. Consider your audience before showing them a thematic map with this legend. You can hand-edit the label of each class, to say things like “>11.1% (Very High)” or perhaps “Very High (>11.1%).”<\/p>\n"},{"acf_fc_layout":"image","image":{"ID":2940639,"id":2940639,"title":"Screenshot 2025-09-22 at 12.33.06\u202fPM","filename":"Screenshot-2025-09-22-at-12.33.06-PM-scaled.png","filesize":384889,"url":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.33.06-PM-scaled.png","link":"https:\/\/www.esri.com\/arcgis-blog\/products\/arcgis-online\/mapping\/better-breaks-define-your-thematic-maps-purpose\/screenshot-2025-09-22-at-12-33-06-pm","alt":"Standard deviation classification is the only classification method that explicitly factors in the mean and standard deviation of the data.","author":"4161","description":"Standard deviation classification is the only classification method that explicitly factors in the mean and standard deviation of the data. ","caption":"Standard deviation classification is the only classification method that explicitly factors in the mean and standard deviation of the data. ","name":"screenshot-2025-09-22-at-12-33-06-pm","status":"inherit","uploaded_to":250552,"date":"2025-09-22 19:36:05","modified":"2025-09-22 19:40:06","menu_order":0,"mime_type":"image\/png","type":"image","subtype":"png","icon":"https:\/\/www.esri.com\/arcgis-blog\/wp-includes\/images\/media\/default.png","width":2560,"height":1347,"sizes":{"thumbnail":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.33.06-PM-213x200.png","thumbnail-width":213,"thumbnail-height":200,"medium":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.33.06-PM-scaled.png","medium-width":464,"medium-height":244,"medium_large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.33.06-PM-scaled.png","medium_large-width":768,"medium_large-height":404,"large":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.33.06-PM-scaled.png","large-width":1920,"large-height":1010,"1536x1536":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.33.06-PM-1536x808.png","1536x1536-width":1536,"1536x1536-height":808,"2048x2048":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.33.06-PM-2048x1077.png","2048x2048-width":2048,"2048x2048-height":1077,"card_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.33.06-PM-826x434.png","card_image-width":826,"card_image-height":434,"wide_image":"https:\/\/www.esri.com\/arcgis-blog\/app\/uploads\/2018\/06\/Screenshot-2025-09-22-at-12.33.06-PM-1920x1010.png","wide_image-width":1920,"wide_image-height":1010}},"image_position":"center","orientation":"horizontal","hyperlink":""},{"acf_fc_layout":"content","content":"

You can see from the image above that this standard deviation method slices the histogram neatly and applies a color ramp to those slices consistently. The \u201cHigh to Low\u201d style of color ramp spreads the blue color progressively across the classes. The map is mainly blue, because the center of the color ramp is itself a medium blue.<\/p>\n

When Should You Classify?<\/h2>\n

It\u2019s pretty interesting to see how the various methods dramatically shift the color around the map. If it makes you feel a little uncomfortable that there are so many options with widely varying effects, that\u2019s good, because your next step is to take control of where and when color is applied to the map, based on your purpose<\/strong>.<\/p>\n

Every map needs a purpose, and you can\u2019t get to purpose without exploring the data first. No matter how many methods are available to slice and dice the data into various colors, at some point each map author (or, manager of people making maps) needs to put meaningful numbers into the map legend: numbers they can explain, talk to and justify.<\/p>\n

In a classified style, does it matter that a county with a value of 15 is colored the same as a county with a value of 20.9 \u2013 in effect saying there is no difference between those two counties? It may, or it may not \u2013 we classify things like body temperature into:<\/p>\n