Bloom two-sigma effect

The Bloom two-sigma effect is the empirical finding — central to the education thread of Marc Andreessen’s It’s Time To Build — that one-to-one Tutoring yields a two-standard-deviation gain over conventional classroom instruction. Andreessen wields it as a “why haven’t we built this?” indictment: we know one-to-one tutoring works that well, and we have the internet, so the failure to build systems that match every learner with a tutor is a failure of will, not knowledge. In the same passage he notes the last major K-12 innovation was Montessori (1960s) and that fifty years of education research never reached deployment.

In Kyle’s notes the effect is the conceptual hook for Synthesis School, the alternative-education startup he tags directly against this argument as a worked example of “build new education systems” — the kind of system that could, in principle, deliver tutoring-scale gains at scale.

Context: The “two-sigma problem” comes from educational psychologist Benjamin Bloom’s 1984 paper, which reported that students tutored one-to-one performed about two standard deviations better than students in conventional classrooms — and framed the open challenge of finding group-instruction methods that approach the same effect.

Where this appears

• It’s Time To Build — cited as the known two-standard-deviation gain from one-to-one tutoring, used to argue we’ve failed to build tutoring-at-scale systems despite having the internet.
• Synthesis School — tagged as a worked example of the build-new-education-systems call the effect motivates.