Chapter 4 Self-Check Quiz: The Forensic Laboratory

Twenty-five questions — multiple choice and short answer — to test your grasp of the chapter. Answer before opening the key at the bottom. The short-answer items reward the honest verbs (conformance vs. correctness, necessary vs. sufficient) over slogans.

Multiple choice

1. A "full-service" crime laboratory is best described as: - A) a single room where one analyst performs all forensic tests - B) a facility organized into specialized sections, each with its own analysts, instruments, and standards - C) a database that automatically matches evidence to suspects - D) the office of the medical examiner

2. Accreditation to ISO/IEC 17025 primarily certifies that a laboratory: - A) produces correct results in every case - B) uses only methods that are scientifically valid - C) operates a documented quality-management system and is competent in its declared scope - D) is independent of law enforcement

3. Which statement about an accredited lab is TRUE? - A) Accreditation guarantees the underlying method is foundationally valid - B) An accredited lab can faithfully run a method (e.g., bite-mark comparison) that lacks scientific validity - C) Accreditation makes contamination impossible - D) Every analyst in an accredited lab is incapable of fraud

4. Quality assurance (QA) differs from quality control (QC) in that: - A) QA is the sample-by-sample check; QC is the overarching system - B) QA is the overarching system of policies and training; QC is the sample-by-sample check on a run - C) they are two names for the same activity - D) QA applies only to DNA; QC applies only to drug chemistry

5. A negative control in an analytical run is a sample that: - A) is known to contain the target and must test positive - B) is known to be free of the target and must test negative - C) is the actual evidence under examination - D) is run only after the report is issued

6. The most revealing form of proficiency testing is one that is: - A) declared, so the analyst can prepare - B) blind — introduced into the casework stream disguised as a real case - C) graded by the analyst themselves - D) built only from clean, single-source, easy samples

7. Method validation is best described as: - A) the second analyst's review of a finished report - B) testing a method against samples of known truth, before casework, to characterize its performance and error rate - C) the accreditation assessment performed by an outside body - D) the chain-of-custody log

8. Developmental validation differs from internal validation in that: - A) developmental validation proves a method can work in principle; internal validation proves a given lab can run it correctly - B) internal validation is done once by the method's developer - C) they are the same thing - D) developmental validation is performed on every case

9. Why is high sensitivity a contamination liability? - A) sensitive methods are slower - B) a method that can detect a few cells of a real perpetrator can equally detect a few cells from an analyst or a previous sample - C) sensitive methods cannot use controls - D) it is not a liability

10. A "staff elimination database" exists so that: - A) underperforming staff can be removed - B) a staff member's DNA, if it appears in a result, can be recognized as contamination rather than mistaken for a perpetrator - C) the lab can bill each analyst's time - D) proficiency tests can be graded

11. "Dry-labbing," as in the Annie Dookhan case, means: - A) running tests in a low-humidity room - B) reporting results for tests that were never actually performed - C) drying samples before analysis - D) a validated rapid-screening method

12. The Sonja Farak case primarily involved: - A) reporting tests never run - B) the analyst consuming the drug standards and case samples while continuing casework - C) a contaminated batch of swabs - D) an invalid comparison method

13. Fred Zain's misconduct is best characterized as: - A) an honest analyst applying a method that turned out to be invalid - B) systematic fabrication and misrepresentation of serological results across two states - C) a single isolated clerical error - D) a contamination problem caused by faulty reagents

14. When an analyst is found to have committed systematic fraud, the doubt extends to: - A) only the specific cases where fabrication is individually proven - B) every case the analyst touched, because their overall reliability is destroyed - C) no past cases, only future ones - D) only cases that went to trial

15. The "independence problem" refers to: - A) analysts working alone without a second reviewer - B) most crime labs being housed within, and answerable to, law-enforcement/prosecution agencies - C) labs refusing to share data - D) the lack of accreditation

16. Which is the BEST example of the principle that a lab safeguard is "necessary and insufficient"? - A) a lab with no quality system at all - B) accreditation, which raises the floor on reliability but does not guarantee any specific result is correct - C) a method that has no error rate - D) a contaminated reagent blank

17. A small county lab that lacks a DNA section will most likely: - A) refuse the case - B) refer the complex evidence to a larger state crime laboratory - C) perform DNA analysis anyway, without validation - D) destroy the evidence

18. Which pair of failures requires DIFFERENT fixes, per the chapter? - A) a failed positive control and a failed negative control - B) analyst fraud (fixed by oversight/integrity controls) and invalid method (fixed by validation or abandonment) - C) intake and storage - D) QA and QC

19. Context management in lab workflow means: - A) giving the analyst the full police narrative so they understand the case - B) withholding domain-irrelevant information (like the suspect's identity) that could bias interpretation - C) managing the lab's appointment calendar - D) storing evidence at the correct temperature

20. Why does the Melendez-Diaz right to confront the analyst matter especially in light of the scandals? - A) it speeds up the backlog - B) a report is only as trustworthy as the human who signed it, who should be available to be questioned - C) it makes accreditation unnecessary - D) it eliminates the need for proficiency testing

Short answer

21. In two sentences, explain the difference between what accreditation certifies (conformance) and what a juror might wrongly assume it certifies (correctness).

22. A drug chemist's negative control develops color during a batch. State what should happen to that batch and why, and name the temptation that leads toward misconduct.

23. Give one strength and two limitations of proficiency testing as a measure of an analyst's real-world competence.

24. Explain, using the cold case, how the initial "accidental fire" classification had a second-order effect on the evidence by way of the lab's triage system.

25. State the one-line "what you can honestly say on the stand" about a lab result, given everything in this chapter about accreditation, controls, and independence.

Answer key (open only after attempting) **Multiple choice:** 1. **B** — a federation of specialized sections under one roof. 2. **C** — competence and a documented quality system within the declared scope. 3. **B** — accreditation certifies conformance, not validity; a lab can faithfully run an invalid method. 4. **B** — QA is the system; QC is the sample-by-sample check. 5. **B** — a known-clean sample that must test negative. 6. **B** — blind proficiency testing, disguised as routine casework. 7. **B** — proving a method on samples of known truth before casework. 8. **A** — developmental proves the method can work; internal proves *this lab* can run it. 9. **B** — sensitivity that detects a true perpetrator's few cells also detects a contaminant's few cells. 10. **B** — so staff DNA is flagged as contamination, not mistaken for evidence. 11. **B** — reporting tests never performed. 12. **B** — consuming the standards/samples while continuing casework. 13. **B** — systematic fabrication and misrepresentation across two states. 14. **B** — every case the analyst touched, because overall reliability is destroyed. 15. **B** — labs housed within and answerable to law enforcement/prosecution. 16. **B** — accreditation raises the floor but guarantees no specific result. 17. **B** — refer complex evidence to a larger state lab. 18. **B** — fraud vs. invalid method need different remedies. 19. **B** — withholding domain-irrelevant, biasing information. 20. **B** — a report is only as trustworthy as the human who signed it. **Short answer (model points):** 21. Accreditation certifies that the lab *has and follows* a recognized quality system and is competent in its declared scope — a statement about *process/conformance*. It does *not* certify that any particular result is *correct* or that the underlying method is *valid*; assuming "accredited = correct" is the error the chapter warns against. 22. The entire batch is invalid and must not be reported, because a contaminated/failed negative control means something is wrong with the run (contaminated reagent, dirty surface, carryover) and the unknowns cannot be trusted. The temptation, under backlog pressure, is to dismiss the anomaly as "obvious" and report the unknown anyway — the first step toward the §4.6 scandals. The control exists to *override* the analyst's expectation. 23. *Strength:* it is a routine, ongoing measurement of competence on samples of known truth — the very kind of error-measuring study many disciplines historically skipped. *Limitations (any two):* analysts often know it is a test and behave differently than in real casework; samples may be unrepresentatively easy and miss the hard, marginal cases where errors cluster; a high pass rate is not equivalent to a rigorously measured real-world error rate. 24. Because the death was recorded as a *probable accidental fire*, the case carried no suspect-in-custody or speedy-trial urgency, so it was triaged *low* and sent to the back of the state lab's months-long queue. The first (untested) assumption therefore didn't just shape the investigation's mindset — it delayed the very evidence that could test it. The assumption had a logistical, not just psychological, cost. 25. Model: *"This result was produced under an accredited quality system with documented controls, which makes it more reliable — but accreditation certifies process, not correctness; the method's validity and the possibility of contamination, error, or bias must each be addressed on their own, and you should weigh the conditions under which the work was done, not the credential alone."*