6.4.3 Data-Driven Design Decisions

• Performance Metrics: Gathering data on prototype performance, such as time spent on tasks, completion rates, and comprehension levels. This data provides insights into how well the solution supports learning and informs adjustments.

• Feedback Loops: Using feedback loops to gather insights continuously, ensuring that each iteration of the design aligns with both learner needs and instructional goals. Feedback loops enhance responsiveness to user needs, leading to more effective solutions​(20211220 LE Toolkit.

  ---

From the Learning Engineering Toolkit

The learning engineering process always draws on data to guide decisions, and inside the creation phase a team keeps asking how data bearing on the targeted learning might be gathered and examined — both to aid learning and as feedback for improving the solution. [LET-01] Letting data drive design choices matters because people's instincts about what actually helps others learn are so often off the mark; looking at evidence from lightweight trials keeps the effort trained on the design features that count and away from expensive wrong turns. [LET-06]

A telling example comes from Kaplan's LSAT preparation. The obvious answer — over an hour of video with practice folded in — was pitted against a worked-examples design meant to ease the demands placed on working memory. Kaplan tested four conditions: video plus workbook running roughly ninety minutes, fifteen worked examples, eight worked examples, and a group that did no preparation. [LET-06] Both worked-example groups did significantly better, the video-and-workbook group barely outperformed those who prepared not at all, and the strongest results came from the group that studied just eight worked examples for about nine minutes. [LET-06] The team formed a hypothesis, built an alternative, tried it out in context, and let the evidence pick the winner. [LET-06]

Findings like these are deeply tied to their circumstances: learning engineering studies focus on particular products and groups of learners rather than broadly generalizable conclusions, so a design that wins in one instance may not carry over to another. [LET-06] A team can build on a simple result by constructing predictive models and A/B testing individual features — say, altering the properties of worked examples or how long learners spend on each — and then refining iteratively as new learner data arrive. [LET-06]

Data can also steer a design effort in a new direction. At Carnegie Learning, a plan to author fresh MATHia content pivoted toward studying how the existing content was already being used; the team switched to authoring a small core set of problems and then watching how problem presentations were distributed to spot gaps in the bank that still needed filling. [LET-06] Allowing existing data to challenge assumptions reshaped both the product and the authoring process considerably. [LET-06]

Further Reading

Source: wrgr/lecommons — curated by the learning engineering community. Confidence: medium — lecommons-curated; not yet independently expert-validated in this context. To validate or challenge any item: use the Edit button on this page. Upgrading confidence from medium → high requires expert sign-off.

Landmark Academic Papers

• Cognitive load during problem solving: Effects on learning — John Sweller (1988). Cognitive Science · doi:10.1207/s15516709cog1202_4 · ~4,000 citations · tier: foundational

  The founding paper of Cognitive Load Theory. Established that working memory limitations impose hard constraints on instruction design. CLT-derived principles (worked examples, split-attention, redundancy) are among the most widely replicated findings in educational psychology and are standard design rules in learning engineering. Source: lecommons/landscape/data/papers.json · ID: LE-LS-AP-008 · confidence: medium · expert-validated: false

• Design Experiments: Theoretical and Methodological Challenges in Creating Complex Interventions in Classroom Settings — Ann L. Brown (1992). Journal of the Learning Sciences · doi:10.1207/s15327809jls0202_2 · ~2,800 citations · tier: foundational

  The most-cited paper in JLS history. Established design-based research as the canonical methodology for testing engineered learning interventions in authentic classroom settings. Defined the epistemological standards that learning engineering uses when validating systems in the wild. Source: lecommons/landscape/data/papers.json · ID: LE-LS-AP-003 · confidence: medium · expert-validated: false

• Abstract planning and perceptual chunks: Elements of expertise in geometry — Kenneth R. Koedinger, John R. Anderson (1990). Cognitive Science · doi:10.1207/s15516709cog1404_2 · ~650 citations · tier: foundational

  Foundational paper mapping human cognitive processes into computational production rules. Established the methodology for encoding domain expertise into ITS knowledge components — the direct basis for Cognitive Tutor and all subsequent production-rule ITS. Source: lecommons/landscape/data/papers.json · ID: LE-LS-AP-001 · confidence: medium · expert-validated: false

• Improving students' help-seeking skills using metacognitive feedback in an intelligent tutoring system — Ido Roll, Vincent Aleven, Bruce M. McLaren et al. (2011). Learning and Instruction · doi:10.1016/j.learninstruc.2010.07.004 · ~600 citations · tier: highly_cited

  Demonstrated that LE methods can be applied beyond domain knowledge to model and improve learner metacognition and self-regulation. Proved that intelligent tutors can engineer help-seeking behavior, not just subject-matter proficiency — broadening the scope of what LE can target. Source: lecommons/landscape/data/papers.json · ID: LE-LS-AP-007 · confidence: medium · expert-validated: false

• The doer effect at scale: Investigating correlation and causation across seven courses — Rachel Van Campenhout, Bill Jerome, Benny G. Johnson (2023). Proceedings of LAK23: 13th International Learning Analytics and Knowledge Conference · doi:10.1145/3576050.3576088 · ~80 citations · tier: contemporary

  Leveraged telemetry from large-scale online platforms to isolate the causal effect of active practice (doing) vs. passive reading on learning outcomes across seven courses. Provides large-sample empirical validation of one of learning engineering's core design principles at scale. Source: lecommons/landscape/data/papers.json · ID: LE-LS-AP-012 · confidence: medium · expert-validated: false

• Generalizable Learning Engineering Adoption Maturity Model — Shawn Blake-Plock, Scotty D. Craig, Emily Czerwinski et al. (2025). Journal of Learning Engineering (preprint/proceedings) · tier: contemporary

  Defines a multi-level organizational capability model for assessing and advancing an institution's LE practice. Provides a structured framework for moving from ad-hoc instructional design toward systematic, data-driven learning engineering — the field's primary tool for organizational self-assessment. Source: lecommons/landscape/data/papers.json · ID: LE-LS-AP-013 · confidence: medium · expert-validated: false

Policy, Reports & Grey Literature

• The Job of a College President — Herbert A. Simon (1967). Educational Record (American Council on Education) · link

  The founding document of learning engineering. Simon coined the term 'learning engineer' and made the case that teaching effectiveness is a distinct, learnable expertise grounded in cognitive science — not a byproduct of subject-matter expertise. The philosophical foundation for everything that followed. Source: lecommons/landscape/data/grey_literature.json · ID: LE-LS-GL-001 · confidence: low · expert-validated: false

• Online Education: A Catalyst for Higher Education Reforms — Karen Willcox, Sanjay Sarma, Philip Lippel (2016). MIT Online Education Policy Initiative (MIT OEPI) · link

  Re-injected 'learning engineer' into higher education discourse at the moment MOOCs reached 58M+ global students. Recommended universities create dedicated learning engineering roles. Catalyzed institutional adoption of LE language and practices across US higher education. Widely cited as the document that triggered the modern LE movement. Source: lecommons/landscape/data/grey_literature.json · ID: LE-LS-GL-002 · confidence: low · expert-validated: false

• Learning Engineering Toolkit: Evidence-Based Practices from the Learning Sciences, Instructional Design, and Beyond — Jim Goodell, Janet Kolodner (2023). Taylor & Francis / Routledge · link

  The field's primary practitioner handbook. Frames LE as a verb — an iterative, evidence-based problem-solving process — not a technology set or a job title. Covers lean-agile methods, HCI design, data instrumentation, motivation modeling, and predictive analytics. Explicitly technology-agnostic: the principles apply equally to AI-driven software and low-tech interventions. Source: lecommons/landscape/data/grey_literature.json · ID: LE-LS-GL-007 · confidence: low · expert-validated: false

• Learning Engineering for Online Education: Theoretical Contexts and Design-Based Examples — Chris Dede, John Richards, Bror Saxberg (2019). Routledge · link

  First book-length treatment of learning engineering applied to digital and distributed contexts. Saxberg's formulation — LE as applying learning science at 'massive, affordable, data-rich scale' — became widely adopted. Documents how online architectures enable continuous telemetry that peer-reviewed journals document years later. Source: lecommons/landscape/data/grey_literature.json · ID: LE-LS-GL-008 · confidence: low · expert-validated: false

• Learning Engineering Adoption Maturity Model (LEAMM) — Shawn Blake-Plock, Scotty D. Craig, Emily Czerwinski et al. (2025). IEEE ICICLE · link

  Organizational capability matrix allowing enterprises to measure fidelity of their LE adoption across human-centered design, learning science integration, and data-driven iteration dimensions. Provides a structured pathway from ad-hoc instructional design to mature learning engineering practice. The field's primary organizational self-assessment tool. Source: lecommons/landscape/data/grey_literature.json · ID: LE-LS-GL-009 · confidence: low · expert-validated: false

Key People

• John Sweller (profile), University of New South Wales (1946–present) — Developer of Cognitive Load Theory

  Formalized Cognitive Load Theory (CLT) distinguishing intrinsic, extraneous, and germane load Source: lecommons/landscape/data/people.json · ID: LE-LS-PP-004 · confidence: medium · expert-validated: false

• Ann L. Brown (profile), University of California, Berkeley (1943–1999) — Pioneer of design-based research and collaborative learning

  Established design-based research (DBR) as the canonical methodology for testing learning interventions in authentic contexts Source: lecommons/landscape/data/people.json · ID: LE-LS-PP-006 · confidence: medium · expert-validated: false

• Bror Saxberg (profile), Formerly Chan Zuckerberg Initiative / Kaplan; founder of LearningForge (active 2000s–present) — Learning science practitioner; LE thought leader; industry-academic bridge

  Led learning science at Chan Zuckerberg Initiative, framing LE as 'precision education' analogous to precision medicine Source: lecommons/landscape/data/people.json · ID: LE-LS-PP-009 · confidence: medium · expert-validated: false

• Jim Goodell (profile), Quality Information Partners (QIP); IEEE ICICLE (active 2010s–present) — LE standards architect; Learning Engineering Toolkit co-editor; ICICLE community builder

  Co-edited the Learning Engineering Toolkit (Taylor & Francis, 2023), the field's primary practitioner handbook Source: lecommons/landscape/data/people.json · ID: LE-LS-PP-012 · confidence: medium · expert-validated: false

• Karen Willcox (profile), University of Texas at Austin, Oden Institute (active 2000s–present) — MIT OEPI lead; LE in higher education advocate

  Led MIT's Online Education Policy Initiative (OEPI), producing the 2016 catalyst report that re-introduced 'learning engineer' to higher education discourse Source: lecommons/landscape/data/people.json · ID: LE-LS-PP-013 · confidence: medium · expert-validated: false

Organizations, Conferences & Journals

• IEEE ICICLE — Industry Consortium on Learning Engineering (consortium) · link

  Source: lecommons/landscape/data/organizations.json · ID: LE-LS-CO-001 · confidence: medium · expert-validated: false

• HCII — Human-Computer Interaction Institute, Carnegie Mellon (research_center) · link

  Source: lecommons/landscape/data/organizations.json · ID: LE-LS-CO-005 · confidence: medium · expert-validated: false

• International Society of the Learning Sciences Annual Meeting (conference) · link

  Source: lecommons/landscape/data/organizations.json · ID: LE-LS-CE-005 · confidence: medium · expert-validated: false

• Journal of the Learning Sciences (journal) · link

  Source: lecommons/landscape/data/organizations.json · ID: LE-LS-JO-001 · confidence: medium · expert-validated: false

• Journal of Learning Engineering (journal) · link

  Source: lecommons/landscape/data/organizations.json · ID: LE-LS-JO-006 · confidence: medium · expert-validated: false

Programs & Initiatives

• Stanford GSE — M.S. Learning Design and Technology (LDT) (PC) · link

  Residential master’s focused on designing and evaluating technology-enhanced learning environments using learning sciences, design methods, and empirical study—one of the clearest Stanford pathways aligned with learning engineering practice (even when the degree title says LDT, not LE). Source: lecommons/site/src/data/programs_people_registry.json · ID: LE-PP-085 · confidence: medium · expert-validated: false

• Naturalistic Decision Making Association (CO) · link

  Research community studying expert decision-making in real-world conditions. NDM methods (critical decision method, cognitive task analysis) are the primary toolkit for eliciting expert knowledge in high-stakes training domains: military, healthcare, aviation, firefighting. Source: lecommons/site/src/data/programs_people_registry.json · ID: LE-PP-139 · confidence: medium · expert-validated: false

Lecommons enrichment applied 2026-04-17. All items pending expert validation. See wrgr/lecommons for source data and curation methodology.