Integration guide.

Learning how to do science is not a sealed subject. Every skill worth teaching — a careful observation, an honest measurement, a fair test — has a history, a fight over the data, and a set of consequences that reach into ethics and public life. When we teach a skill as if it were a step to memorize, we strip away exactly the parts that make it stick — the story, the argument, the stakes. This guide is the playbook for putting those parts back.

Integration is not decoration. It is not a “fun fact” tacked onto the end of a lesson. It is a deliberate method for making each skill reach outward — into history, reading, and writing first, and then into ethics and statistics — so that learning how to do science becomes something a student can think with rather than just recall.

Why integration matters for retention

Memory is associative. A step stored on its own, connected to nothing, is a fact with one fragile thread holding it in place. The same step connected to a story, a controversy, and a consequence is held by a dozen threads — and when one fails, the others keep it from falling out of the mind. This is not a teaching opinion; it is how human memory is built.

A historical comparison gives these skills a concrete purpose. Semmelweis's 1847 intervention and the dated mortality record supported preventive action, but the clinics were not randomized identical groups. Compare particular periods and denominators, examine other differences, and distinguish the evidence available then from later explanations. A short retrospective source does not justify a single motive for every critic or an undated permanent ward rate.

The goal of integration isn’t to make lab skills “more interesting.” It’s to make them harder to forget — because the student understands not just what the step is but how we learned to trust it and why it mattered.

The integration spine — what radiates, and how to choose

Integration is not freeform. Every unit radiates the same structured set of connections off the science-skill spine, organized in three tiers plus a quantitative lane. This is what keeps the cross-domain work rigorous instead of random.

• Core spokes — always required. History, Reading, and Writing. Use a cited historical or evidence-focused source section, including an assigned educational or textbook passage where appropriate. Distinguish original observations from retrospective explanation. Produce a student-authored response or approved accessible equivalent with evidence and a limitation. Optional trade books and elective topics do not replace this work.
• Standard spokes — required where they fit. Statistics (rates, comparison groups, honest reading of numbers) and everyday examples (which real object — a bean seedling, a paper airplane, a fizz tablet dissolving — makes the skill concrete). Most units carry these naturally; where a unit genuinely doesn’t, we don’t force it — we move it to the elective pool below rather than fake a connection.
• Elective spokes — pick a few. A menu the guide assigns from, or the student chooses from — say two of five: Statistics extension, Ethics debate, History deep-dive, Communication & persuasion, Design & data visualization. Electives are additive depth, never a substitute for the core. Letting students choose feeds wonder and lets faster students go deeper as extension work.

The statistics lane. Numbers are not a spoke — we use them, we are not a math program. But learning to do science leans on measurement and simple statistics more than almost anything else, so every unit names the specific quantitative skill it actually requires, mapped straight back to the practice: reading a ruler or graduated cylinder to the right precision, turning a raw count into a rate, averaging repeated trials, plotting points and reading a trend, comparing two groups fairly. Students do the numbers inside the investigation, where they mean something, not as a parallel curriculum. The unit-by-unit lane is tabled below.

The core three — History & Ethics · Reading · Writing — run in every unit. Statistics and everyday examples run wherever they fit. Electives are chosen, not assigned by default. And the numbers are always present — but always in service of the skill.

How it’s assessed. Integration is graded as its own strand on the unit rubric, separate from the skill-mastery criteria. A student can be Mastery on the skill and only Proficient on integration, or the reverse — which keeps the skill bar pure while still rewarding the cross-domain depth that makes the learning stick.

The repeatable method

Integration sounds like an art, but it runs on a method — one you can apply to any unit, in this course or beyond it. There are four steps, and they always go in the same order.

1. Pick the unit’s big idea. Name the skill, not just a procedure. For design, define the comparison, assignment, observations and possible confounders before interpreting a difference.
2. Find a real historical, data, or ethics source. Cite its section and distinguish original records from later analysis. Clearly labeled synthetic practice can rehearse the same reasoning, but cannot be presented as the historical evidence.
3. Build a question students investigate. Turn the anchor into something to do, not just read — a rate to calculate, a position to argue in writing, a dataset to interpret. A good question forces students to use the skill to reach a conclusion of their own.
4. Connect back to the skill. Close the loop. After the investigation, name explicitly which skill the student just used, so the integration deepens the unit instead of distracting from it.

Skip step four and you get a history lesson wearing a lab coat. Do all four and the outside world becomes a lens that makes the skill sharper. The worked example below shows every step in action.

Worked example: Semmelweis and the handwashing data

The clearest demonstration of the method is the one we use to anchor the whole year: Ignaz Semmelweis and the handwashing data from Vienna General Hospital in 1847 — a clear case of testing competing explanations against evidence. The situation is simple to state: two maternity clinics in the same hospital had very different death rates from childbed fever. Its story reaches into history, statistics, ethics, and writing all at once.

1. The big idea. Comparisons can provide strong evidence without being randomized controlled experiments. The wards differed in staffing and practices, and the intervention also provides a before-and-after record. Identify possible confounders rather than assuming that everything except handwashing was identical.
2. The anchor. Semmelweis introduced handwashing in 1847 and recorded a substantial fall in mortality. Rates must be tied to dates and denominators: the often-cited roughly 18% and 2% values refer to particular months, not permanent rates for two identical clinics. Use the original records or an approved analysis, not an undated anecdote.
3. The question students investigate. Read dated admissions and deaths, calculate comparable rates, graph the record, and consider trend, seasonality, and differences between groups. Explain which observations support the intervention and which conclusions require additional evidence. Use the practice activity below before interpreting the historical record.
4. The connection back. Distinguish observational comparisons, nonrandomized interventions, and controlled classroom investigations. Communicate an evidence-based conclusion with limits; do not replace historical analysis with an unsupported judgment about a person's character.

The task is to explain how evidence supported a preventive intervention while retaining uncertainty about the design, interpretation and historical context. Communication should expose evidence to scrutiny; persuasion alone neither makes a claim true nor makes a finding valid.

Integration anchors for all eight units

The childbed-fever account is examined through each unit's different skill. Units 5 and 6 also use the preserved Significance historical admissions/deaths and model checks. Distinguish these numerical records and retrospective analyses from each task's separately labeled synthetic practice. Required source-linked writing remains. The historical clinics were not randomized identical groups; retain dates, denominators and alternative explanations.

The statistics & measurement lane, unit by unit

Numbers never drive a unit, but learning to do science uses them constantly — always anchored to the object or measurement in front of the student. Here is the quantitative skill each unit actually uses.

Run the course this way and the eight units stop being eight separate piles of procedure. They become eight windows onto the same truth — that doing science well is how humans learned to trust what they know, and that every skill on the page was once a hard-won lesson someone had to fight to be believed. That is the version of the craft a student keeps.