Data Science

Data Science and Research


Conducted in 2011 and 2012, “Twitalectology” was one of the first research studies to use data from Twitter to map variation in dialects — for example, finding out where people say “pop” instead of “soda” (or “coke”). Twitalectology was featured in the New York Times and other major media, and was published as a graduate thesis in 2013.

More info on Twitalectology can be found here.

Social Media and Social Meaning

“Social Media and Social Meaning”, another graduate research project, explored quantitatively how the use of abbreviations on Twitter and Facebook (such as “u” for “you”) changed whether their user was more likely to be perceived as intelligent/unintelligent, male/female, etc. Presentation slides detailing the study and its results are available online.

Data Journalism

How To Use R For Sports Stats“, a three-part series written for TechGraphs, provides an introduction to R coding and basic data analytics using baseball as the chief data set. No prior experience with R or data analysis is necessary.