The Poetics of Big Data 

SEPTEMBER 2013

Poetry and algorithms rarely go together--and perhaps for good reason. But what happens when the tools of data science are used to analyze long-term engagement in a digital poetry community for youth? And what implications does this have for teaching and assessing literacy?

A Movie Project That Became a Poetry Platform 

“If you don’t learn to write your own life story, someone else will write it for you,” is Power Poetry’s motto. Launched in April 2012, it is the world’s first online/mobile poetry community for youth. Power Poetry’s mission is to move its users to action through empowered literacy. It is not just a digital community for poets – it is a community of socially engaged poets who use innovative multimedia art as a tool for educational and social impact.  

The award-winning documentary To Be Heard is the inspiration for Power Poetry. Hailed by the New York Times as “…one of the best documentaries of the year”, the film follows three young poets whose struggle to change their lives begins when they start to write poetry. To Be Heard is the culmination of 12 years work Power Poetry's founder spent with youth in NYC public schools as co-creator of the Power Writers program. The Power Writers’ mission is to empower youth through the mastery of language and cultural literacy. The program provides youth an in-school, student-centered, safe space that supports them as they build literacy skills, connect with poet mentors, and sharpen their critical thinking skills. Power Poetry uses digital technology to bring this mission to scale by empowering youth from across the U.S.—and the world—to build literacy skills that lead to personal transformation and social change.

Using Data to Help Accelerate Language Mastery

Big Data and technology are used frequently in business settings.  But with limited resources—and limited spending dollars for tech—non-profits have been historically behind the times in using technology and data to improve their programs and services.  This is precisely why analytics start-up SumAll.com created SumAll.org: to provide data that non-profits can use to deepen their impact.  

Power Poetry was fortunate enough to secure a partnership with SumAll.org, and together, they embarked on an unparalleled journey to analyze 50,000 poems submitted to PowerPoetry.org by youth ages 13-24. Our goal was to assess the (1) quality of a poem and, (2) linguistic progress of users who submitted multiple poems over time. The most striking insight that emerged from this exploratory study is this: submitting more poems over time leads to higher quality poems. Writing more poetry, then, may to heightened language mastery. But there is still much work to be done to deepen this preliminary study.

An Algorithm That Predicts the Quality of a Poem

SumAll.org analyzed 50,000 poems and quantified characteristics that may influence or indicate the quality of a poem and its mastery of language. The algorithms we created looked at word counts, abstractions, phonetics, rhymes, tonality, punctuation, n-grams, and many other features in order to assess quality.  

This preliminary research has broad implications for the field of education and student literacy assessment.  It also suggests that poetry is an incredibly effective tool through which to learn writing and measure literacy