Today’s #FutureHigherEd idea is a novel use of big data to match students with peers and mentors and improve the learning experience. In 2010, Tim Renick at Georgia State Univesity started to examine millions of student grade reports looking for markers that might be predictive of classroom success. Over time, Renick and his team developed analytical models that would alert students and advisors when an intervention is needed. The resulting data-driven interventions improved graduation rates and saved GSU students over $3 Million in tuition payments. It also saved the university money and those savings justified hiring more advisors. The GSU project helped launch a revolution in the use of data to improve outcomes and was the subject of an Education Advisory Board case study which can be found here.
Terry Hosler’s submission to the openIDEO Challenge takes this idea in another direction: using student data to match students with peers and mentors. It’s an important concept because this kind of support strongly influences achievement.
For a student to be successful attending a university or college away from home takes a ‘village of support’ including guiding communication both before and once arriving on campus with both the student and their network of support (family, mentors, and, when the student chooses, a broader community to increase their confidence and comfort level.
This need is there for all students out of their normal cultural environment.
This cultural match is particularly crucial for low-income, first-generation, rural students who are academically astute but unsure of higher education opportunities ‘away from home’ due to lack of experiences and feeling ‘like a fish out of water’.
How can Hosler’s idea be combined with the GSU predictive analytics model to identify not only culturally similar peers, but also a community of mentors and tutors who are likely to be effective at helping students improve classroom performance?