Evidence-analysis defense
This is a live exam at the bench. The student analyzes a piece of physical evidence — a fingerprint against a reference, a chromatogram, a fiber, a blood type — running the method start to finish and reaching an identification. Then the guide starts asking: why that method, what the result actually reports, how strong the match really is, and how far it could be wrong. There is no worksheet to copy and no figure to look up: the student stands over the evidence and defends the analysis — and its certainty — out loud. The point is not to claim proof, but to state honestly how sure the evidence lets them be.
| Criterion | Not yet | Approaching | Mastered |
|---|---|---|---|
| Method & technique | Runs the analysis incorrectly — contaminates the sample, skips a control, or applies the wrong method for the evidence. | Follows the method but handles a control or a step inconsistently, weakening the result. | Chooses and runs the right method for the evidence, uses proper controls, and avoids contamination throughout. |
| Reading the result | Misreads the result — calls a non-match a match, or cannot say what the test actually shows. | Reads a clear result but hesitates on a faint or ambiguous one, or explains it only in vague terms. | Reads the result correctly — a ridge comparison, a chromatogram band, a blood type — and explains what the method is and is not reporting. |
| Certainty & what a match means | Claims a certain match, or calls the result proof that the suspect is guilty. | States a conclusion but overstates its strength, or cannot explain why it is a likelihood rather than certainty. | States the conclusion as a likelihood, explains what a statistical match does and does not mean, and never claims 100% or proof. |
| Limits & sources of error | Treats the result as exact; names no limits or sources of error. | Lists a limit or two but cannot say how it weakens or strengthens the conclusion. | Identifies real limits — sample quality, a partial print, database size — and reasons about how each affects the strength of the conclusion. |
| Oral defense under questioning | Folds at the first follow-up or recites a memorized line that does not fit the evidence. | Answers some follow-ups, falters when asked to defend how sure they can honestly be. | Handles unrehearsed follow-ups with sound, on-the-spot reasoning, and holds the line that the analyst reports evidence while the court decides guilt. |
“The ridge detail matches the reference at twelve points with no unexplained differences, so I can say it’s a strong likelihood the two prints came from the same finger — not a certainty, and not proof this person is guilty. That call belongs to the court; my job is to report how strong the evidence is.”
“It matches, so it’s definitely the same person and they did it. I’m sure — it’s a 100% match.”
This assessment is AI-proof by design: it happens at the bench, with real evidence, in real time. No chatbot can run the comparison, read a result it did not produce, or hold up under a follow-up question about how sure it can honestly be. Mastery is shown by analyzing and defending — and above all by being honest about certainty: the analyst reports the strength of the evidence, and the court decides guilt.