GIS Cookbook: Recipe - Classifying Data  
Keywords: Census tract, census data, classification, reclassification, nominal data, interval data, null value
Category: Census Data
Software: ArcView 3.2

Problem: I have a shapefile of census tracts with attributes, how do I classify the data?

Description: When a shapefile is being displayed on your view, it is usually shown in a single color. This is because the information has not been classified into categories. When information has been classified by a given variable, the shapefile will display different colors representing different values for that variable in its attribute table. For example, if a census tract were to be displayed according to age, it may result in an image where red would stand for a tract with an average age above 60, orange for the 50 to 60 range, yellow for 40 to 49, and so on. When classifying data, you can choose the ranges in which your data will be divided. You are also able to select the colors that will be associated with each range, although it is important to keep your color scheme intuitive and easy to follow.

Scenario:
In this recipe, you will learn how to classify and display census tracts according to average age within each tract.


Methodology:
1) Create a new view and add your census tract shapefile to your view by using the Add Data button

.



2) Make the census tract theme active by clicking on it in your view window. Now select the Edit Legend

button. A window similar to the following should pop-up on your screen.



3) In the space next to Theme should be the name of the census tract layer. For Legend Type we will choose "Graduated Color." If you were to choose "Unique Color" a different and unrelated color would be assigned to each category. "Unique color" should be used if you are classifying nominal data (i.e. data that does not have an inherent order to it that would require an ordered color scheme to accompany it). "Graduated Color" is being used here because we want to display the average age within each tract. Age is interval data, so we want a color scheme that shows this order.

4) For Classification Field, select the category you would like to display. In this example we are using Avg_age.

5) We do not need to normalize the data by any other category because we are only displaying one attribute, therefore leave it as None. Your Legend Editor should look similar to the following picture.



6) Notice in the above picture, instead of beginning with 0, the first interval begins with –99. -99 is the null value used in the database to stand for a missing value. These values should be excluded to avoid skewing the image’s color scale. To exclude the null value, press the null value button

within the Legend Editor window.

7) Under Field choose the name of the category from which you would like to exclude the null value, then type in the value in the following space below.


Note: if you choose the "Include No Data Class in Legend" box, it will display the null value on your legend and label it as “no data.”

Select the OK button. Your legend editor should now exclude the null value from its first interval.



8) Select Apply at the bottom of the Legend Editor window. Notice if you do not like the color scheme or the way it displays in the view window, select Undo and make the necessary changes. To change the color scheme scroll down the Color Ramps option. Your View should now be classified.

Pitfalls:
Links:
ESRI Discussion Forums - Chat with others about your ESRI problems
ESRI homepage -
ESRI Canada (K-12) Tutorial - This is an ESRI tutorial page that talks about reclassifying census tracts within a geocoding tutorial. This recipe references this page.
Authored by: Sam Ying Modified: 9/10/03


Copyright © 2002-2014 by Regents of University of California, Santa Barbara
Cookbook: Ben Sprague, Ethan Sundilson, Carlin Wong, Sam Ying