Learning Design Wiki

The Learning Design Wiki splits the design of teaching into separable parts and binds each part to the studies behind it, carrying 170 principles, 1,057 elements, 54 patterns, 1,010 designs, 1,903 products and programs, and 6,700 strategies as of October 2026. Every element and pattern links to claims, and every claim to a source, so a reader working on instructional design can start from a principle, drop to the elements it implies, and arrive at the citation.

The evidence layer is the part worth knowing. The wiki's own report counts 6,849 distinct studies behind 26,495 claims, of which 31% are experiments or systematic reviews and 2% are pre-registered trials or well-powered meta-analyses. It also reports that 98% of claims rest on a single study, and that only 24% of the studies print an effect size for the finding being cited.

What you can do with it

Browse by knowledge type rather than keyword. Principles state a design commitment, elements are the instructional building blocks, patterns assemble elements into a lesson-level design, designs apply a pattern to a named setting, and strategies are the finer moves. Claim pages list the studies bearing on one question as rows in a table instead of averaging them, so disagreement stays visible.

Who it is for

Instructional designers and instructors who want a design decision with its evidence attached, and researchers who want a mapped corpus of learning theories in design terms. Teams building generative AI design assistants can use it as a structured reference, since the site publishes a machine-readable bundle rather than a set of web pages.

Notes and caveats

Read the trust panel on every page. Language models maintain the wiki under one human maintainer, no person has checked any page, and the site labels a page a well-sourced lead rather than a finding. Claims carry a design tier, an effect size code and a rigour code, and "no effect size printed" means the source reported none, not that none exists. Content is CC BY-SA 4.0, so reuse must carry the same license, and the code is MIT.

Connected Concepts

Learning Design, Pedagogical Patterns, Theories and Frameworks, AI-Assisted Educational Research, Limitations in AIEd Research, Interpreting and Applying AIEd Research