Purdue Study On CTA Tweets Most Obvious Study Ever

This is the age of Twitter and, more than ever, people are taking to the messaging service to voice their pleasures and dislikes. In the case of the Chicago Transit Authority. Search the hashtag #cta and one can scroll for a while before finding a positive tweet about bus or rail service.

Three Purdue researchers theorized that transit systems across the country could learn a lot from negative tweets and set out to chart the levels of positive and negative tweets about CTA. They presented their findings at the annual Transportation Research Board conference in Washington yesterday.

Using publicly time-stamped tweets and geographic location data, Craig Collins, Samiul Hasan, and Satish Ukkusuri ran their findings across a "sentiment algorithm" to track the level of positive and negative tweets about transit system. Below is a graph from Independence Day, where sentiment about CTA can be affected by, for example, Taste of Chicago.

Shocking, yes?

Collins, Hasan and Ukkusuri say that public transportation agencies across the country can use Twitter to learn a lot about what needs to be improved about their infrastructure and service, similar to how marketing firms and movie studios use Twitter and other social media to gauge consumer reaction to films, events and products. Even without the use of Twitter, CTA could use a lot of work.

Contact the author of this article or email tips@chicagoist.com with further questions, comments or tips.

By Chuck Sudo in News on February 7, 2012 9:10 AM 0 Comments

Comments [rss]

INTPLibrarian

FWIW, I often use Twitter to search for why there seems to be a delay on a particular eL line. It's a better up-to-date resource than anything else I've found.

ChicagoD

When this was blogged on Atlantic Cities people commenting actually argued for the usefulness of the research. Sheesh.

sat3911

There is a usefulness. Just because everyone assumes it is obvious doesn't mean it is true. For example everyone knows that sentiment about the CTA is going to be poor because of the overloading of the system due to the fireworks, but looking at their data it turned out to match up to a fire.

Because they proved the obvious, they now have the correlation for past events showing a likely cause and effect. Now they can work on propagation rates and see if it is fast enough to be diagnostic for unknown events. They can also watch to see if their mitigating efforts are having an effect and change responses if they aren't.

This is science and in the scientific method you prove the obvious before basing your next hypothesis on it. Not all science is sexy.

/Soapbox