In Pursuit of the Holy Grail
Dr. David Wiley, Co-Founder, Chief Academic Officer
This chapter reviews Bloom’s classic 1984 article on the “2 sigma problem.” Personally, I find this article to be one of the most inspiring pieces of writing on education of all time. The article isn’t inspirational in a way that pithily rejects the education establishment while holding up a supposedly superior model, like the misquoted and misattributed “education isn’t the filling of a pail, but the lighting of a fire.” Rather, this article demonstrates, through a rigorous, randomized experiment, that the “average” student has incredible academic potential.
In the study, students were randomly assigned to one of three conditions. The Conventional condition had students in a typical classroom setting, with one teacher and 25 – 30 students. The Mastery Learning condition was similar, except that students in this condition received detailed feedback on their tests and were given opportunities to retake equivalent tests in order to achieve mastery. In the Tutoring condition, students either worked one-on-one with a teacher or in a group of two to three students and one teacher. The result? The average student in the Tutoring condition performed two standard deviations better than the average student in the Conventional condition. And because standard deviations are represented with the Greek letter sigma, this result became known as the 2 sigma problem.
“But the students in the Tutoring condition did better… what’s the ‘problem’?” you might ask.
If sigmas aren’t your thing, let me put the results in different terms. The average student in the Tutoring condition performed better than 98% of students in the Conventional condition. The “2 sigma problem” is that we now know that the average student is capable of truly amazing academic performance, but that as a society we can’t afford to help them realize that potential. (We can barely afford one teacher for a classroom of thirty students. Imagine the cost of hiring a full-time tutor for every single student!) When you see that kind of potential going unrealized on a massive scale, it’s a problem.
Bloom gives this call to action in the article:
“If the research on the 2 sigma problem yields practical methods – which the average teacher or school faculty can learn in a brief period of time and use with little more cost or time than conventional instruction – it would be an educational contribution of the greatest magnitude.” (p. 6; emphasis in original)
The educational technology establishment’s response to this call has largely been to develop “intelligent tutors” in order to scale tutoring to all students. Since teachers are the expensive part of the equation, the logic goes, the obvious answer is to replace them with software.
At Lumen, we reject this dehumanizing interpretation of Bloom’s call to action. Our approach is to recognize that, as Bloom wrote elsewhere, different students need different amounts of help. Some students will earn As with no tutoring at all. And those students that do need help sometimes don’t need it all the time. Based on this realization, Lumen’s courseware makes it easy for faculty to (1) see which students need help, (2) see which specific topics they need help on, and (3) invite them to come to office hours (or another individual or small group setting) to get the help they need. We hypothesize that this approach can both scale to meet the needs of all learners and honor and value faculty-student relationships.
About This Series
At Lumen, everything we do is focused on improving student learning. You already know that we create awesome and affordable interactive courseware, engage in both data-driven and community-driven continuous improvement, and support faculty professional development. You might not know that we also do things like our “Not a Book Club,” in which the whole company is invited to engage with the research on learning.
We recently finished How Learning Happens: Seminal Works in Educational Psychology and What They Mean in Practice by Kirschner and Hendrick (2020). This is a terrific and really accessible book that summarizes and explains some of the most important research about how we learn. We had a great time discussing the book’s chapters (there’s one chapter for each research article) and talking about how that research shows up in our courseware and professional development designs. I thought it would be fun to share some of those insights with you in a brief series discussing several of the book’s chapters, so here we go.