11.2 Ethical Considerations in Learning Engineering

Ethical considerations in learning engineering arise from the field's defining characteristics: reliance on detailed learner data, algorithmic decision-making that affects learner trajectories, and scale that can amplify both benefits and harms. Ethical practice is not separable from technical practice — it is embedded in every design choice from instrumentation to deployment.

The Ethical Stakes of LE Systems

The U.S. Department of Education's AI-in-education report documented that algorithmic systems can cause direct harm to learners — facial recognition e-proctoring that systematically failed darker-skinned, transgender, and neurodivergent students is the most prominent example — and that these failures carry civil-rights implications. [LE-LS-GL-004] The Journal of Learning Analytics has developed frameworks for auditing algorithmic fairness and examining power dynamics in educational data systems. [LE-LS-JO-004] IEEE ICICLE's professional standards establish learner welfare and equity as primary values, above institutional efficiency. [LE-LS-CO-001]

Core Ethical Domains

Data privacy and learner rights (11.2.1) addresses data minimization, consent, FERPA compliance, and learners' rights to explanation and correction. xAPI and IEEE 9274 provide the technical infrastructure for consent-aware data collection. [LE-LS-SG-002]

Equity and fairness (11.2.2) requires proactive audit for disparate impact across race, gender, disability, and socioeconomic status — building equitable systems rather than correcting inequitable ones after deployment. [LE-LS-GL-004]

Responsible AI (11.2.3) encompasses transparency, explainability, human oversight, and accountability in AI-powered learning systems. The practitioner's obligation is to ensure that algorithmic recommendations can be understood, challenged, and overridden by instructors and learners. [LE-LS-GL-007]

The Generalizable LEAMM includes ethical governance as a maturity dimension — mature LE organizations have institutional review processes and stakeholder engagement embedded in their design cycles, not applied as post-hoc review. [LE-LS-AP-013]

From the Learning Engineering Toolkit

In the Toolkit, Jordan Richard Schoenherr argues that learning engineering is inherently ethical, with ethical decisions arising at every phase of the process. [LET-07] Working through the process model, the ethical questions specific to each phase come into view. The challenge phase means understanding the ethical features of the problem space, including the learner population and its sociocultural setting; the creation phase brings out vulnerabilities and the risk of disadvantaging some learner groups; implementation raises privacy, consent, and autonomy; and investigation demands keeping bias out of how data are read. [LET-07]

The Toolkit describes ethical affordances as context-appropriate design features that honor recognized ethical principles and stakeholder values, whereas anti-affordances fall short of doing so. [LET-07] It frames ethics as a competency built on ethical sense-making — interpreting principles, reconciling clashes between values that often cannot be measured against one another, and spotting the ethically salient aspects of a situation — rather than simply ticking off a list of principles. [LET-07]

Learning engineers draw on professional codes of conduct from several fields — education (AECT, for example), talent development, engineering (IEEE and ACM), and psychology (APA) — along with national and international guidance. [LET-07] The book takes up the APA's five principles — beneficence and nonmaleficence, fidelity and responsibility, integrity, justice, and respect for people's rights and dignity — and maps them across the phases of the process in a Sense-making Ethical Evaluation Matrix (SEEM). [LET-07] Value-oriented design approaches — value sensitive design, reflective design, design justice, and anti-discrimination design — offer further ways to weave values through the design process. [LET-07]

Chapter 17 turns this into an ethical decision-making tool, SEEM-ED, that recasts ethical principles as concrete questions posed throughout the process. [LET-17] It is intentionally more than a checklist: each question carries a rating scale and follow-up questions, and a representative set of stakeholders scores them on their own — anonymously where social pressure could distort results — to bring the design's ethical weak points to light. [LET-17]