The principles
Reviewed · 1 min read
Fourteen principles make up the library. Seven are proven in real classrooms by randomized trials. The rest are strong but rest more on lab work, and the site says so.
Each principle follows the same order, so you learn the layout once: a 60-second card with the finding, what to do, the grade and the best evidence; then a 5-minute brief with why it works, what it looks like in each grade band, common mistakes, what a leader looks for, what we don't know, and who to read next. The seven proven principles also have a playbook and a full evidence file.
14 of 14 principles
Principle 01
Working memory is small
Evidence grade: Strong3 to 5
chunks working memory holds at once, not seven
Students can hold only about four new pieces of information in mind at once. Overload it and learning stops.
Principle 02
Explicit teaching for new material
Evidence grade: Strong164 studies
explicit teaching beat unassisted discovery
When material is new, students learn more when the teacher explains and models it than when they are asked to figure it out on their own.
Principle 03
Worked examples
Evidence grade: Strongg = 0.48
gain in math from worked examples, across random-assignment studies
Beginners learn math faster by studying a fully solved problem and then trying a near-twin than by solving problems cold.
Principle 04
Check every student
Evidence grade: Strong+2 months
of GCSE progress in a 140-school randomized trial
Teachers who see every student's answer several times a lesson catch mistakes before they set, and their students learn more.
Principle 05
Feedback on the task, not the student
Evidence grade: StrongOver 1 in 3
feedback interventions lowered performance (607 effect sizes)
Feedback usually helps, but about one time in three it makes performance worse. The difference is where it points the student's attention.
Principle 06
Quizzing from memory
Evidence grade: Proven in classroomsAbout 0.5 SD
achievement gain from quizzing across 222 classroom studies
Pulling an answer out of memory strengthens it more than rereading or reviewing notes does.
Principle 07
Spaced practice
Evidence grade: Strong317 experiments
the longer you need to remember, the longer the gaps
The same amount of practice, spread over days and weeks, is remembered far longer than practice crammed into one session.
Principle 08
Mixed math practice
Evidence grade: Proven in classrooms61% vs 38%
mixed vs. blocked practice on a surprise test a month later
Practice sets that mix different problem types beat sets of one type, because students have to decide which method to use.
Principle 09
Systematic phonics
Evidence grade: Proven in classrooms0.24 to 0.41
range of phonics effects across analyses, all positive
Children learn to read words better when letter-sound patterns are taught directly, in a planned order, than when phonics is taught as it comes up.
Principle 10
Knowledge-building curriculum
Evidence grade: Proven in classrooms+16 percentile points
in reading, grades 3–6, in a lottery study
What students already know is one of the biggest factors in what they can read and learn next. A planned sequence of science, history, and the arts from kindergarten raises reading scores years later.
Principle 11
Fractions on the number line
Evidence grade: Proven in classrooms5 randomized trials
of a 12-week program for 4th graders at risk in math
Students who struggle with math learn fractions better when they first learn fractions as sizes, placed on a number line, rather than only as parts of a pizza.
Principle 12
Small-group tutoring
Evidence grade: Proven in classrooms0.29 SD
average gain across 89 randomized trials of tutoring
Frequent tutoring in very small groups is one of the most reliable ways to help students who are behind.
Principle 13
The curriculum you choose
Evidence grade: Proven in classrooms110 schools
in a randomized trial of four math curricula
Curriculum is not neutral. In a large randomized trial, the math program a school was assigned changed how much its students learned.
Principle 14
Direct Instruction programs
Evidence grade: Strong328 studies
positive effects in reading, math, language and spelling
Fully designed, sequenced programs in the Engelmann tradition ("big DI") have produced consistently strong results over fifty years of studies.