« « The limitations of SQL and Access

My new perspective on math – it’s a journalistic tool! » »

Changes in how we travel across U.S. borders

Posted by on Dec 30, 2009 in Blog, data visualizations, projects | No Comments

Traveling is something we often take for granted nowadays — at least, I know I do.  And as situations in my life have changed, I’ve been thinking about all the different types of transportation I use.  I’m a bit more reliant on my car in the Chicago suburbs, and while I griped about the Metro while in D.C., I miss its reliability.  I took buses more in DC than I ever did in Chicago, and my acquaintance with commuter trains grew while I went to undergrad in the Boston suburbs.

It’s a large part of people’s lives, and deserves further study.  That’s the motivation behind this interactive graph I cooked up using data from the Bureau of Transportation Statistics.  It’s interesting to see just how much more people are using cars to cross over from the United States to Canada or Mexico, and how that compares to traffic from buses, trains and pedestrians.  I was actually quite surprised to see just how prevalent walking is in some states, second most in overall traffic to passenger vehicles.

Here, we have an visual exploration of how total traffic across the U.S. border has changed since 1996, in terms of how much each state and each mode of transportation contributes to the overall amount of those coming in and out of the U.S. on land.  Air travel was not included in the dataset, and rightfully so, do you count the planes that are passing over the border, or just ones that land in Canada or Mexico?  It seems as if it adds a lot of unnecessary issues.

Also, the data only goes up through July 2009, so when it looks like there is a drop from the previous year, it may be because a full 12 months aren’t recorded.  This seems to be a common issue in data analysis that I’ve come across in other projects too.  I don’t think you should leave it out, because that’s denying the user information, but it does merit an explanation.  I welcome feedback on this, one of my more ambitious projects to date.  (More analysis and personal reflections after the jump.)

I thought the best way to evaluate this data set was by enabling the user to both see the big picture and make comparisons across states or modes of transportation.  (Full disclosure: This data set was from my final project for my digital frameworks class, but I didn’t get to complete it as thoroughly as I wanted.)

I still haven’t done as much analysis as I would like, but a new quarter calls.  (And if you’re wondering what the distraction was, it was designing and implementing the web site for a quarter-long team reporting project on religion and the environment, as well as data crunching on a piece Medill did with the Tribune looking at former staffers from the Hill now lobbying for organizations with health care interests.)

On the technical side, this was an adventure deeper into ActionScript than I had ever gone before, and much of it was based on this tutorial from Flowing Data.  I was exploring Adobe’s Flex program for the first time, and it was both challenging and rewarding.  Using Flare, a library created by the UC Berkeley Visualization Lab, helped me to take a lot of shortcuts, but at the same time, I found you really have to get inside someone else’s head to understand where different code is stored.  This emphasized the importance to me of keeping my own code as clean as possible — something I’ve heard often, but rarely experienced from the other side.  Flare was actually very well-documented in the comments, it was just an issue of searching through many different classes for the right line.  I suppose it’ll get easier with experience.

I wonder about this lesson as it applies to journalism. So often, I think of reporting as an individual job, but if we share work we have done in the past, we could be more efficient and provide even more information to the public.  I guess it’s an argument to stop bragging about how my chicken-scratch handwriting gives me the skills to write in investigative code no one can read, and think more about how to better facilitate the exchange of information with colleagues.  It’s the same rationale as to why interactive projects come out better when done as part of a team.  Sure, I can report, analyze, design and produce packages myself, but it would come out so much better if I were working on a specialized part of it with others who had expertise in different components.

My current dilemma is discovering exactly what my specialty is.  That’s why I’m embarking on a data visualization independent study on Monday, a major component of my final quarter at Northwestern.  The journalism industry and I share this commonality: We must work to determine where is it we fit in this changing world.  It’s not a question of if there’s a place for us, but a question of how best to discover and focus on our greatest strengths.

« « The limitations of SQL and Access

My new perspective on math – it’s a journalistic tool! » »