Teaching Safety
Without turning it into a rules list
Read this before you teach Chapter 24. It is the most important chapter in the book and the easiest one to teach badly.
The problem
Almost every safety unit a student has ever encountered has been a rulebook: wear this, guard that, here are the Focus Four, here is the fall protection trigger height, here is the quiz. Students can pass that assessment and leave with the belief that safety is a compliance function performed by a department, adjacent to the real work of building.
That belief is not just incomplete — it is the belief that produces incidents. A project manager who thinks safety is the safety manager's job will make schedule decisions without ever considering their safety consequences, and those decisions are where hazards actually come from.
Chapter 24's threshold concept is:
Safety is a property of the production system, not a rulebook. Hazards are produced by decisions — about schedule, sequence, staffing, procurement, and pressure — made weeks earlier by people who were not thinking about safety at all. Fix the system that produced the hazard, not just the hazard, or you have merely reset the clock on the next one.
Your job is to make that concept survive contact with an exam.
The regulatory content still matters. Teach it properly.
None of the above is an argument for skipping the technical material. A manager who cannot recognize an unsafe excavation, cannot compute a fall clearance, or does not know what a competent person is required to do is dangerous regardless of how systemically they think.
So teach it, and teach it accurately:
- The Focus Four — falls, struck-by, caught-in/between, electrocution — with real mechanisms, not just names.
- The hierarchy of controls — elimination, substitution, engineering controls, administrative controls, PPE — and the uncomfortable observation that PPE is the weakest control and the one the industry reaches for first.
- Fall clearance arithmetic. Work it. A six-foot lanyard tied off at a twelve-foot anchor does not work, and a student who has computed free-fall distance plus deceleration distance plus harness stretch plus safety factor plus worker height will never forget why.
- Competent person versus qualified person as defined terms with specific duties.
- The General Duty Clause as the thing that catches the hazard no specific standard addresses.
- The multi-employer worksite — creating, exposing, correcting, and controlling employer — because it is the general contractor's actual legal exposure.
Refer to 29 CFR 1926 by subpart topic. Do not have students memorize section numbers; have them learn to find the requirement. And note that state-plan states differ and other countries have entirely different regulators.
The centrepiece: run the scaffold near-miss as a role-play
This is the highest-value ninety minutes in the course. Do not lecture it.
The scenario
Week 34 of the Northgate job, a Tuesday, north elevation, level 3. A frame scaffold was partially modified overnight by a different trade to run conduit; a plank was lifted and not re-secured. A mason tender, Milo Serrano, stepped onto it at 7:20 a.m.; the plank shifted and he caught himself on the top rail. No injury. Safety Director Bea Salgado stopped work on the elevation for the day.
The investigation found three things:
- No competent-person inspection tag for that shift — the tag was two days stale.
- A scaffold modified by a trade that did not erect it, with no re-inspection.
- A crew running behind after the steel acceleration, under an unwritten pressure to make it up.
How to run it
Set it six days after the event, at the close-out meeting where the investigation report is finalized. Assign roles: the safety director, the project manager, the general superintendent, the VP of operations, the masonry subcontractor's foreman, and — if your class is large enough — the owner's representative and the scaffold subcontractor.
Give each role a private brief with their own incentives. The project manager and the VP made the acceleration decision. The masonry foreman's crew was behind. The safety director wants finding three in the report.
The whole exercise is the argument over finding three.
Findings one and two are easy. They have obvious corrective actions, they belong to identifiable parties, and nobody is threatened by writing them down. Finding three says that a decision made in a trailer, about money and about days, put a man on an unsafe plank three weeks later. Writing it down means the company's own management is a root cause.
What to watch for
Classes almost always try to resolve it by making finding three vaguer — "schedule pressure" as a general observation rather than a specific decision. Let them try. Then ask the question that breaks it open:
"If you write it that way, what changes next time?"
The answer is nothing. A finding that names no decision produces no corrective action. That is the entire lesson, and it lands far harder when the class discovers it than when you assert it.
The landing
Close by having them design the corrective actions, and insist on the distinction:
- Corrective actions for findings one and two are hazard fixes — a tag regime, a scaffold modification-control procedure requiring re-inspection by the erecting contractor.
- The corrective action for finding three is a system fix — a standing safety impact review as a required step in every acceleration decision, with named questions: what hazards does this decision create, what controls are required, what do those controls cost, and who is accountable.
Then ask how they would know, six months later, whether any of it worked. That is the leading-versus- lagging indicator conversation, and they will arrive at it themselves.
Reinforce it every time the class recommends acceleration
This is the part that actually changes behavior, and it costs you two minutes per occurrence.
Chapters 20, 27, 29, 31, and 41 all contain decisions where a class will reasonably recommend compressing a schedule. Every single time, ask the same question:
"Who is going to be on a scaffold in three weeks because of what you just decided?"
By the fourth or fifth repetition, students start raising it themselves before you do. That is the behavior you are trying to install — not a rule they can recite, but a question they ask automatically when they are about to make a schedule decision.
The canonical Northgate acceleration is the anchor: $168,000 bought back 17 of 23 days, and it also produced trade stacking, a deck-edge rework event, and a measurable spike in near-misses in weeks 34 through 36. Chapter 20's case study prices the unpriced portion of that decision. Use it.
Leading versus lagging indicators
Students default to the recordable incident rate because it is a number and it is available.
Teach them why it is nearly useless as a management instrument: it measures the past, it is statistically noisy on a single project, and — most importantly — it can be improved by suppressing reports rather than by preventing harm. Chapter 24's second case study is exactly this: a project with an excellent recordable rate and a serious injury, where a safety incentive program paying for zero recordables had suppressed near-miss reporting for eight months.
Then teach the leading indicators that actually predict: inspection frequency and quality, near-miss reporting rate (and the counterintuitive point that a rising near-miss count is usually good news), pre-task planning completion, training currency, corrective-action closure time, and — the book's own contribution — schedule pressure as a measurable leading indicator.
A good discussion question: "Your recordable rate improved 40% this year. Name three explanations, only one of which is good news."
Handling the emotional weight
Two things to prepare for.
Some students have personal experience. In any construction cohort, someone has had a family member hurt or killed on a job site, or has been hurt themselves. Do not force disclosure and do not be surprised by it. If someone shares, treat it as the most valuable content in the room and then move on without dwelling — dwelling makes people regret having spoken.
Do not use graphic imagery. It is common in safety training and it is a poor teaching tool: it produces a brief emotional response, no durable behavior change, and it is genuinely distressing to people who have lived through something similar. The book deliberately uses a near-miss with no injury as its anchor for exactly this reason — nothing about the analysis requires anyone to have been hurt, and the systemic lesson is identical.
What success looks like
At the end of the course, a student who has learned this can:
- Name the Focus Four and describe the mechanism of each, not just the label.
- Apply the hierarchy of controls to a specific hazard and explain why PPE is the last resort.
- Write a job hazard analysis with the crew that will do the work, and explain why one written in an office and handed down is paperwork rather than planning.
- Read an incident report and distinguish proximate causes from systemic ones.
- Look at a schedule decision and ask, unprompted, what hazards it creates.
The last one is the only one that will still be operating in ten years. Assess for it.
If you only have one session
Cut the regulatory survey to reading and spend the whole session on the near-miss role-play. A student who leaves understanding finding three, and who has argued about it, has the thing that matters. A student who can recite fall-protection trigger heights and thinks safety is somebody else's department has not.