Data-Driven Road Trip Stop 5: KML files carry me Home (to New Hampshire)Reading time: 3 minutes
As they say in Texas, Howdy Y’all! I know that because I spent one whole night there on my final push back to New Hampshire. But in case you didn’t recognize my Southern drawl – it’s me again, Lily, except this time I’m not on the road anymore!
Now that I have returned to the place I call home, I have done plenty of reminiscing, sharing photos and stories, and…..analyzing! I spent 35 days driving, hiking, paddleboarding, rodeo-watching, camping, and creating data around the country. Now that all that is in the rearview, I was able to use Rapid Insight’s Construct to explore the data I created.
Exploring the Data
To start my exploration, I used an excel spreadsheet I created while planning the trip. This file came in handy since I already had all the planned destinations and their addresses in one place. Having those addresses was important because once I connected to the data in Construct, I was able to use the Geocoding node. If you haven’t seen it before, the Geocoding node is an Add-in node (download it here) that formats and converts addresses into latitude and longitude coordinates by comparing and taking advantage of Census data as well as Google and Geocodio APIs.
Now that I had corresponding latitude and longitude information, I was then able to push my data into a KML output node. This created a KML file that I could then use to plot points on a map for each of my destinations. I did this by importing the KML file created in Construct into Google Maps.
Seeing those points on a map gave me a really cool view of just how far and wide I had travelled. But I was also interested in finding out exactly how far I had driven. Once the points were plotted on the Google map, I was able to add a layer containing driving directions retracing my steps to show the routes I drove. This allowed me to view the mileage and duration for each trip – the data that I was interested in, but had not kept track of.
Since Google calculated this for me, I went back to Construct and used an Embed Data node to easily copy all the values over. The Embed Data node is perfect for situations like this, where you don’t necessarily have data in a file, but you have a handful of values that you need to work with. It is really easy to paste the data directly into Construct through the Embed Data node, and then begin working with it.
The Trip in Numbers
I used the functions in the Aggregate node to calculate some summary statistics on the trip data.
- In total I drove:
- ? 8,182 miles around the country
- ⏱️ 130 hours and 42 minutes
- ⛽ Plus a bit more on top of each for day trips and rest stops
- Farthest trip:
- Telluride, Colorado to Amarillo, Texas
- ? 769 miles
- ⏱️ 11 hours and 10 minutes
- Shortest trip:
- Mount Rushmore, South Dakota to Jackson, Wyoming
- ? 127 miles
- ⏱️ 2 hours 25 minutes
- Average trip:
- ? 430 miles
- ⏱️ 6 hours 53 minutes
Driving it Home
This Data-Driven road trip is certainly something I will never forget. But the memories go far beyond what I could even begin to share through these summary statistics. Creating this series titled the “Data-Driven” roadtrip was a fun play on words, but like I mentioned before, the data itself here isn’t giving us the whole picture. It’s rather supporting my stories and experiences. Looking back, this was more of a data-informed roadtrip.
And as I look back over this blog series, finding data to support those experiences is what made telling the stories so much fun! At Rapid Insight we strive to keep a data-informed mindset in all the work we do. Data is a critical piece of the puzzle, but its most important function is as a supporting, informing resource for investigations in both personal and professional pursuits, and my time on the roadtrip was no different!