The Khan Academy is a set of freely-accessible online video-centric learning resources, principally focussing on declarative and procedural knowledge, covering a wide range of subjects at levels suitable for school-age and adult learners. Learning analytics figures in three ways: as the engine for services offered by the Khan Academy through the web pages; as access to data for analytics processes undertaken by third parties; and as a means of continuous design enhancement (not considered further here).
Khan Academy describes itself as being a “personalized learning resource… using state-of-the-art adaptive technology”, but the emphasis on personalisation is on recommendation by teachers/coaches, rather than by the system, although they have deployed recommendation engines and analytics-drive estimation of skill mastery, which drives adaptation of the exercises.
Khan Academy provides information to teachers/coaches on individual and class-level performance. This provides summary estimates of effort, engagement, and difficulty with the material. The learning materials are mapped to a set of skills, with various mastery levels for each; the teacher/coach can drill-down to this level and use the information on progress or difficulty to recommend materials for follow-on or under-pinning skills, or to instigate an alternative learning activity (e.g. outside Khan Academy).
Khan Academy provides a dashboard for learners which shows progress against skills (as for the teacher/coach) and activity pattern in time and against different skills.
Data access by third parties is via a web-standards based API and gives differentiated access according to the data type. Video, playlist, topic/skill maps, and exercise data is open access. User-level activity and progress logs are secured, requiring login and authorisation.
[This synopsis originally coded by the LAEP project - Learning Analytics for European educational Policy https://laepanalytics.wordpress.com/]