Behind the technologies, there is also a new learning science, based in the following educational measurement principles:

  • A system that learns the learner – knowledge, understanding, progress as represented in knowledge artifacts and interactions, everything the student does.
  • A system that learns all learners – data adds to the intelligence of the system, e.g. corpus comparison based on target level/assessment/tagging by topic; validated change suggestions.
  • Personalized learning + social learning – to measure each student in vibrant, horizontal knowledge ecologies, and not (just) alone.
  • The test is dead, long live assessment – every measurement can be formative (immediate feedback to improve work; summative assessment is no more than a retrospective view of learner progress).
  • Semantic legibility – “construct validity”; always pointing to causation (the details of how the learning happened), not just correlation (the overall outcome).
  • Just-in-time and just enough – aptness of feedback as evidence in actioned feedback and progress measures; feedback on feedback to improve feedback.
  • Progress syntheses – “How am I doing?” asks the student or the teacher, and when they do, they get clear visualizations (belying the bigness of the underlying data).


To find out more about our approach to evidence-of-learning, read our technical papers:

Sources of Evidence-of-Learning: Learning and Assessment in the Era of Big Bata

Interpreting Evidence-of-Learning: Educational Research in the Era of Big Data