Visualizing NC Fatal Crash Data in 3 Steps

Bella Wei
Stratifyd
Published in
3 min readApr 8, 2016

We found the data for fatal car crashes that happened in North Carolina from 2010 to 2014 on OpenData. The data set takes out personal identifiable information and adds other important dimensions such the crash date, the type of crash, alcohol involvement, road surface, lighting condition, teen driving, and etc. What does the crash data tell us?

3 steps

In 3 simple steps, Signals analyzes and visualizes the data without using SQL or R. Let’s find out how!

Step 1: Upload without SQL

Open Data

We exported the data in CSV from OpenData, and pasted it into the Signals Upload Window. The data size is 1.9MB; it has over 5900 records and about 20 columns including latitude and longitude formation formation {34.8648322965, -79.0950421482}.

CVS
Submit

And, hit Submit.

Step 2: Visualize without using R programming

Where are the fatal car crashes happening in NC? A Signals heat map gives you a quick glance of the spots where crashes happen at a high rate.

Screen Shot 2016-04-06 at 6.18.49 PM

Zooming in on Signals Geo Map, we take a closer look at the major cities in NC.

  • Charlotte: US 16 and I-85
Charlotte
  • Raleigh: US 401 and I-440, I-40
Raleigh
  • Fayetteville: NC Hwy 87 and I-95
Fay

Step 3: Analyze with ease

Screen Shot 2016-04-08 at 10.26.13 AM

Signals automatically displays all the categorical information from your data set. Now you can dive in with ease.

  • Crash Types: beware when you are making a lane departure! It is the most common crash type of all.
crash type
  • Alcohol Involvement
alchohol
  • Frequent Crash Route: Most fatal crashes happen on NC state highways, followed by local and interstate.
Screen Shot 2016-04-08 at 3.17.43 PM

For some visualization platforms, one would have to import and prepare the data using Microsoft SQL server, and use R programming to visualize the data. Signals can visualize and analyze your data, structured or unstructured, in just 3 steps.

Signals is fast, easy, and you do not need to be a programmer or data scientist to use it. Would you like to try Signals for yourself? Sign up for a free trial of our platform. If you have a project in mind, email us and we are happy to connect!

--

--