Notes from a lab coordinator: what I’d want a new dean to read first.
This essay is the synthesis. If a new department chair, dean, or curriculum committee member asked for one piece of writing that captured how I think about the questions a large undergraduate science program faces, this would be it — written as counsel rather than as argument.
When a new dean arrives in a college that includes a large undergraduate science program — the kind that funnels students into nursing, allied health, and pre-medical careers across an entire region — the volume of competing requests for attention can be overwhelming. Faculty want resources; students want flexibility; accreditors want documentation; legislators want efficiency. Each request is reasonable. None of them, by themselves, tell the dean which decisions warrant the most caution.
What follows is one lab coordinator’s attempt to offer that guidance. It is not a list of demands, and it is not specific to any particular institution. It is a small set of working principles I have arrived at over years of running multi-section undergraduate lab programs — principles I have found useful for distinguishing decisions that can safely be moved quickly from decisions that, if moved quickly, will be regretted in three to five years.
Each principle below corresponds to a longer essay elsewhere on this site, linked in case the dean would like the underlying argument and citations. The point of this synthesis is not to re-argue any of them. It is to put them in one place.
Five working principles
Lecture and laboratory teach different things. They are not interchangeable.
Decades of cognitive-science and clinical-skills research converge on a single point: bench-based instruction produces four kinds of learning that lecture cannot reliably produce in its place. Procedural memory, decision-making under genuine uncertainty, perceptual identification on unlabeled real specimens, and social calibration on a shared physical task. Each is bench-dependent in the technical sense; each has documented downstream consequences in clinical practice.
The implication for curriculum decisions is straightforward. Proposals that move time from lab to lecture should be evaluated not only on what they preserve but on what they give up — specifically, on whether the affected learning categories can be produced by any other instructional channel available to the program. Often they cannot. Where they cannot, the substitution is not a neutral efficiency gain; it is a category change.
A gateway course is a structural decision, not a contained one.
Most academic decisions are contained: small enrollment, elective, low coupling to anything downstream. If the decision turns out to be wrong, it is reversible without lasting harm. A gateway course — thousand-student enrollment, required prerequisite for a half-dozen professional programs, tightly coupled to admissions and licensure decisions in adjacent programs — is structurally different. Wrong decisions in this category are difficult to detect within a single review cycle and difficult to reverse without harming a cohort that has already moved through.
The implication is not that structural decisions deserve special ceremony. It is that they warrant a higher evidentiary bar before they are made — the same bar a clinical guideline committee would apply to a change in standard of care, for the same reason. The cost of being wrong is paid by people who were not in the room when the decision was made.
Choose assessments to match the claim the course needs to defend.
Multiple-choice instruments and lab practicals measure different constructs. Multiple-choice formats sample content domains efficiently and reliably; they reach their ceiling at recognition and reasoning-from-given-information. Lab practicals demonstrate procedural execution, perceptual identification on real specimens, and applied decision-making. Neither dominates. Both are legitimate. The choice between them is a measurement question, not a values question.
Courses adjacent to credentialing decisions — licensure, board certification, professional-program admission — generally need criterion-referenced instruments that match what the credential measures. Not because they are harder. Because they certify what the credential certifies. A grade in a course that prepares students for a performance-based credential should be defensible as a claim about performance.
Substituting paper exercises for bench tasks is a hypothesis. Treat it like one.
The transfer-of-learning literature is unambiguous on a specific point: paper analogs of bench tasks preserve some learning categories (declarative knowledge, diagram recognition) and not others (procedural memory, perceptual identification, decision-under-uncertainty, social calibration). The substitution is, in technical terms, a far-transfer claim. Far-transfer claims have a documented history of underperforming intuition.
Where the substitution is being considered for cost reasons — usually the honest reason, and a legitimate one — three questions decide whether it is wise. What construct does the bench task measure that the paper analog does not? How would we detect, in the next one to three years, whether the substitution has produced a downstream gap? And is the cost saving, expressed honestly, worth the measurement loss? Programs that can answer these questions clearly make sound decisions either way. Programs that cannot make decisions anyway, but the consequences are detected later and by someone else.
Innovation and content reduction are not the same thing. The distinction is worth defending.
Evidence-backed pedagogical innovation — active learning, peer instruction, POGIL, team-based learning, high-structure course design — has earned its mainstream status on the strength of replicated, peer-reviewed evidence. These techniques deserve the energy and resources institutions have put behind them. None of this is in dispute.
Some curricular changes that share the label do not share the evidence. The pattern that distinguishes them from genuine innovation is not their shape; it is the absence of an answer to the question “what does the evidence say this produces?” A curriculum committee can make the distinction cleanly with a five-question checklist and an outcomes-tracking requirement. The cost of failing to make it is borne by the students who took the course in good faith and by the downstream programs that have to backfill what was reduced.
What I am not asking the dean to do
A short, important section: what this synthesis is not asking for. It is not asking for protection from change. It is not asking for resources beyond what the program needs to do its work. It is not asking for any colleague's proposal to be overruled, or for any committee's authority to be redirected. The five principles above are not a position against reform; most of them are positions in favor of more disciplined reform.
The ask is smaller and more specific than that. It is that decisions in the categories above — bench-vs-lecture allocation in lab-dependent courses, curriculum changes in gateway courses, assessment instrument selection in credentialing-adjacent courses, paper-for-bench substitutions, and proposals labeled “innovative” — be required to clear a slightly higher evidentiary bar before they are made, and that the outcomes be tracked so the institution can learn from them.
The administrative work is hard, and most of it is invisible. The request from this corner of the college is small: ask one extra question before approving a curriculum change, and require one extra document attached to the proposal. Both will pay for themselves within a year.
One concrete suggestion
A single, specific, easy-to-implement step that operationalizes everything above. The dean, or the curriculum committee chair, institutes a one-page Curriculum Change Brief as standard accompaniment to any proposed change in a gateway-course or credentialing-adjacent course. The brief asks for four things: the measurement claim the change is meant to support; the relevant peer-reviewed evidence in comparable contexts; the outcomes-tracking plan, with a timeline and named measures; and a one-paragraph honest accounting of what is being given up.
This is not gatekeeping. It is the documentation any program would expect of itself, formalized into a single page so the requirement does not depend on individual reviewers remembering to ask. Proposals with strong evidence behind them clear the brief easily and gain credibility from doing so. Proposals without strong evidence get the most useful possible response: a concrete answer about what evidence would be required before the change is wise, and an opportunity to gather it.
Closing
The undergraduate science programs that do this work well are not the ones that resist change. They are the ones that change carefully — piloting before scaling, measuring before declaring, building reversibility into proposals from the beginning. Careful change is slower, and more durable, and the kind that earns the trust of the downstream programs the gateway course feeds.
That trust, once built, is the most valuable thing a science program has. It is also the easiest to lose, because it is built on a track record that the next change can either extend or interrupt. Asking one extra question before approving a change, and requiring one extra page attached to the proposal, is a small administrative cost in service of preserving something that takes much longer to rebuild than to maintain.
Drafted May 2026. Comments and corrections welcome.