I have three columns, one for percentage White, percentage Hispanic, and percentage African American per census tract. I'm trying to use Python to extract the population subtype with the highest percentage into a new column called "highest_percentage". The highest percentage category will have 3 classes: 1,2, and 3. I can not figure out how to set a conditional statement so that I can automatically assign a value of 1,2, or 3 to the census tract to use in a classified map.

I was able to write code to find the highest percentage and assign the new field the same value as that percentage, but i'm trying to find which field has the highest value and assign it 1-3 based on a classification. 3 white, 2 Hispanic, 1 African American

I've attached a screenshot of the table and the code window below.

screenshot of arcmap attribute table

Arcmap python field calculator

second arcmpa attribute table sceenshot

  • Welcome to GIS SE! We're a little different from other sites; this isn't a discussion forum but a Q&A site. Please check out our short tour to learn about our focussed Q&A format. Please always present any code using formatted text (highlight it use the {} button) rather than pictures.
    – PolyGeo
    Commented Feb 14, 2019 at 5:29

2 Answers 2


I prefer to use an arcpy UpdateCursor for tasks such as these, mainly because I think the syntax is easier to work with and can be incorporated into larger workflows.

In your case, find the index of the largest percentage in each row. For example:

import arcpy

fc = r'C:\path\to\your\file_geodatabase.gdb\featureclass'

with arcpy.da.UpdateCursor(fc, ("african_american", "hispanic", "white", "index")) as cursor:
    for row in cursor:
        max_val = max(row) # Find the max value in each row
        max_index = row.index(max_val) # Find the index of the max value
        row[3] = max_index + 1 # Apply a correction on the index

enter image description here


Try this

Pre-Logic Script Code:

def maxnum(fields):
    if (fields[1]>max_value):
    if (fields[2]>max_value):
    return return_value



Not the answer you're looking for? Browse other questions tagged or ask your own question.