Grading Rubrics
Rubrics for the project notebook, its individual deliverables, the case-study discussions, and the simulations. All are designed to be handed to students at the start of the term — a rubric students have not seen is a grading tool, not a teaching tool.
Rubric 1 — The Project Notebook (summative)
The capstone assessment. Use it at the end of a one-semester course, or at the end of Semester II in the two-semester sequence.
Weighting note. The notebook is worth 30–35% of the course grade in the syllabi provided. If you weight it lower than 25%, students will treat it as optional and it will not get done.
| Criterion | Weight | Exemplary (90–100) | Proficient (75–89) | Developing (60–74) | Insufficient (<60) |
|---|---|---|---|---|---|
| Completeness | 15% | All assigned deliverables present, in order, with a table of contents | One or two minor deliverables missing or thin | Several missing; sequence unclear | Large gaps; not a coherent record |
| Internal consistency | 25% | Quantities flow into the estimate, the estimate into the schedule and SOV, and the SOV into the pay application and EVM. Numbers reconcile across documents | Mostly consistent; one or two reconciliation breaks that are acknowledged | Documents contradict each other; student appears unaware | Each document invented independently |
| Technical correctness | 25% | Arithmetic correct throughout; CPM passes verifiable; forecasts defensible | Minor errors that do not change conclusions | Errors that change conclusions | Method fundamentally misapplied |
| Documented assumptions | 15% | Every deliverable states its basis: sources, inclusions, exclusions, and what was assumed | Assumptions present but incomplete | Assumptions implied, not stated | No basis given; numbers appear from nowhere |
| Professional judgment | 10% | Recommendations are defensible, risks are named and owned, trade-offs stated honestly | Sound but unexamined | Recommendations asserted without reasoning | Judgment absent or indefensible |
| Presentation and usability | 10% | A stranger could pick it up and understand the project in twenty minutes | Readable with effort | Disorganized | Not usable as a record |
The two criteria that matter most
Internal consistency (25%) is the one that distinguishes a student who did forty-two assignments from a student who managed a project. If the takeoff says 1,240 CY and the estimate prices 1,180 CY and nobody noticed, that is the failure this rubric exists to catch — and it is exactly the failure that costs money in practice.
Documented assumptions (15%) is the professional habit the whole book is trying to install. Chapter 12's threshold concept is that an estimate is a priced bundle of assumptions; a notebook whose numbers have no stated basis has missed the point regardless of its arithmetic.
The honest note about arithmetic
Do not fail a student for an arithmetic error that they caught, documented, and corrected. Do mark down an uncorrected error that propagated through four subsequent documents, because that is precisely what happens on a real project and precisely what the discipline exists to prevent.
Rubric 2 — Individual weekly deliverable (formative)
Use this weekly. It is deliberately fast — four checkmarks and two lines of comment, under three minutes per student.
| Check | |
|---|---|
| ☐ | Present and on time |
| ☐ | Complete — contains everything the checkpoint asked for |
| ☐ | Consistent — reconciles with the previous deliverables |
| ☐ | Basis stated — assumptions, sources, and exclusions written down |
Mark: ✔ Complete · ↻ Revise and resubmit · ✗ Incomplete
Then two lines of comment: one thing that is right, and the single most important thing to fix.
Do not write more than that. Weekly formative feedback that takes fifteen minutes per student will not survive week five, and feedback that arrives late is worth almost nothing.
Rubric 3 — Case-study discussion
For graded participation in the case studies, which are the book's primary discussion vehicle.
| Level | Description |
|---|---|
| 4 — Advances the discussion | Brings specific evidence from the case; identifies a consideration others missed; builds on or respectfully challenges a classmate's reasoning; changes their own position when given a good argument |
| 3 — Contributes substantively | Engages with the actual facts and numbers; answers questions with reasoning rather than assertion |
| 2 — Participates | Present and responsive; contributions general rather than specific |
| 1 — Passive | Attends; does not engage |
| 0 — Absent or disengaged | — |
The behavior to reward most is changing position under a good argument. It is rare, it is the actual skill, and students will not do it unless you visibly value it.
The behavior to mark down is confident assertion without evidence, however plausible. This is the industry's most common professional failure and the classroom is a cheap place to correct it.
Rubric 4 — Simulations
For the five simulations in field-simulation-guide.md. Assess the debrief, not the performance —
a student who negotiated badly and understands exactly why has learned more than one who happened to
win.
| Criterion | Weight |
|---|---|
| Preparation — used the role brief; knew their own numbers and constraints | 20% |
| Argument quality — reasoning grounded in the documents and the arithmetic, not volume | 30% |
| Listening — responded to what the other party actually said and to their constraints | 20% |
| Debrief insight — can articulate what happened, why it turned where it did, and what they would do differently | 30% |
Do not grade on the settlement number. In the change-order simulation, a contractor who settles at $121,000 with a clear account of why their documentation limited them has demonstrated more than one who extracted $180,000 from an inexperienced partner.
Rubric 5 — Threshold concept mastery
The book carries fifteen marked 🚪 threshold concepts (listed in the front matter's Learning Paths file). They are the most durable thing in the course and deserve their own assessment.
Use this on exams and orally. Ask a student to explain one in their own words, with an example that is not the book's.
| Level | Description |
|---|---|
| Mastery | Explains it in their own words, generates a novel example, and can state what someone who hasn't grasped it believes instead |
| Understanding | Explains it correctly and applies it to a supplied scenario |
| Recognition | Can define it; cannot apply it to an unfamiliar case |
| Recall | Reproduces the book's wording without evident understanding |
The Mastery band's third element is the useful one. "What does someone who hasn't got this think instead?" cannot be answered from memory. A student who can say "they think a GMP caps what the owner will pay" has genuinely crossed the threshold.
Rubric 6 — Written communication
Apply to the memos, notices, RFIs, and narratives throughout the notebook. This is a professional skill the course is uniquely positioned to assess, because the book teaches it explicitly.
| Criterion | Exemplary | Insufficient |
|---|---|---|
| Purpose | The reader knows within one sentence what is being asked or asserted | Buried or absent |
| Specificity | Dates, quantities, references, and dollar impacts stated | Vague: "significant delay," "additional cost" |
| Evidentiary quality | Would still make sense to a stranger in a year | Requires context only the author has |
| Tone | Factual; no blame, speculation, or legal conclusions | Editorializing, sarcasm, or argument |
| Actionability | States what is needed and by when | Leaves the reader to infer the ask |
The diagnostic: "framing crew, Area B, 6 workers idle 0700–1130, 27 labor hours, awaiting response to RFI 112 issued 4/2" versus "waiting on RFI." Both describe the same day. Only one is worth money.
A note on grading a course built on a free textbook
Every answer in this book is public. Quiz answers are behind <details> blocks, worked solutions are
in Appendix J, and any student can read ahead.
This is not a problem to solve; it is a reason to assess differently. Assess production, not
recall: the notebook, the simulations, the oral threshold-concept checks, and open-book exams that ask
students to do something rather than remember it. The exams in additional-assessments/ are written
on that assumption.
The same logic applies to generative AI. Decide your policy, state it in week one, and choose assessments that are robust to it. The most robust of all is a student standing in front of you for fifteen minutes walking through a notebook they built and answering "why is this number what it is?"