Poetry, Data and Social Gaming



Obesity may be America’s biggest health problem, but when it comes to education, the literacy gap poses the greatest threat to the future of America. Nearly two-thirds of fourth graders don't meet reading proficiency standards. And although millions of dollars are invested each year to help improve literacy, America continues to lag behind compared to other markets like the Netherlands and China.




Infographic: "OMG! We SRSLY have a Literacy Crisis!"

Infographic: "OMG! We SRSLY have a Literacy Crisis!"

Some educators and policymakers like Diane Ravitch, former U.S. Secretary of Education, question the effectiveness of such programs as the No Child Left Behind Act, which assumes that schools bear the full responsibility for helping students achieve literacy. Studies have also shown that a number of non-school related factors such as having family support affects the way children develop language skills. This begs the question, if the literacy gap is a societal issue that extends beyond the scope of traditional schooling, what tools can we give students to help them improve their mastery of language outside the classroom?


Poetry and data are two words that rarely go together, but to quantify students’ language skills, SumAll looked at hundreds of data points extracted from poems and created a logical regression model to describe the mastery of language in the poem. This was then applied to the poems submitted on Power Poetry, so that each poets’ progress could be tracked.

For the training set used to build the regression model, a total of 200 poems were examined, half of which were poems written by well-known professional poets and half were written by amateur writers.  The algorithms created looked at word counts, abstractions, phonetics, rhymes, tonality, punctuation, n-grams, and many other features in order to assess quality. The output of the model predicts the likelihood of the poem being written by a professional writer, with 1 indicating absolute certainty and 0 indicating that the poem was most likely written by an amateur writer. 


Our research shows the positive impact of technology on the mastery of language. Those who publish at least 10 poems on Power Poetry saw a visible progression of their language scores: starting out slightly below average (51% percentile) aspiring poets were able to advance to above average during the course of their 10 poem submissions to the 47-percentile. 

Scoring Micropoems on Twitter

Scoring Micropoems on Twitter

4% is a significant improvement. Given that powerpoetry.org knows the approximate location of where poems were submitted, it was possible to map poems and their language scores to respective low-income and high-income zip codes. Among powerpoetry.org users the literacy gap between low-income and high-income neighborhoods appears to be about 9 percentiles. Active poets seemed to be able to advance their language score by about 4-percentiles which is the equivalent of half of the gap between more affluent and less affluent zipcodes

SumAll.org & Volunteer Team

Power Poetry Team

Roland LegiardiI-Laura

Roland LegiardiI-Laura