📄 Direct examination of Dr. John Gerdes (morning, part 3) — Thursday, August 3, 1995
📅 Aug 3 — Day 128
🛡️ Barry Scheck🔬 Dr. John Gerdes
chain_of_custodydnalab
Address:
C:\DEPT103\CRIMINAL\1995\AUG\3\DIRECT-EXAMINATION-OF-DR-JOHN-.DOC
TRIAL
▲ Day 128 of 167

Direct examination of Dr. John Gerdes (morning, part 3)

Witness: Dr. John Gerdes
Examiner: Barry Scheck
Called by: Defense • Date: Thursday, August 3, 1995 • Utterances: 26
Barry Scheck concludes his direct examination of Dr. John Gerdes by establishing that the proficiency tests taken by the crime labs in this case — LAPD, DOJ, and Cellmark — were not representative of actual case conditions. Scheck reads from the NRC report on laboratory error and uses Gerdes to confirm that none of the labs were tested on degraded, aged, or mixed bloodstain samples of the kind found in this case.
1 (The following proceedings were held in open court:)
2 MR. SCHECK:

Doctor, I'm going to start reading a paragraph that is at page 89 of the NRC report and ask is that your view of it and ask to you explain it.

3 MR. CLARKE:

Objection, hearsay as phrased.

4 THE COURT:

Overruled.

5 MR. SCHECK:

"Laboratory errors happen even in the best laboratories and even when the analyst is certain that every precaution against error has been taken. It is important to recognize that laboratory errors on proficiency tests do not necessarily reflect permanent probabilities of false positives or false negatives results. One purpose of regular proficiency testing under standard case conditions is to evaluate whether and how laboratories have taken corrective action to reduce errors. Reported errors should be based on proficiency tests that are truly representative of case materials with respect to sample quality, accompanying description, et cetera. Tests based on pure blood samples would probably underestimate error rate and tests based primarily on rare and extremely difficult samples, which might be useful for improving practice, would probably overestimate. Although the California Association of Crime Lab Director proficiency tests was less than ideal, being opened rather than blind and not requiring reporting of size measurements, the materials appeared to have been representative of standard case work."

6 MR. SCHECK:

Now, do you agree with that judgment of the national research council?

7 DR. GERDES:

I do.

8 MR. SCHECK:

Do you agree--why is it important to have proficiency tests on samples that are representative of case work?

9 DR. GERDES:

Well, I think as I tried to explain in that paragraph, you want to make the tests at the same difficulty as what you are going to try and analyze, so if you make the test too easy, you don't really get an accurate reflection of what your error rate is, how often you might make a mistake. So you want to try and mimic exactly the type of thing you are going to analyze as possible.

10 MR. SCHECK:

And in terms of having a blind test administered by an outside agency, why is that important?

11 DR. GERDES:

Well, if you know you are being tested, you just naturally try a little harder, and if you--so both in the clinical field--in the clinical field on our proficiency tests we have to state--I even have to sign that I'm going to handle the sample the same way I would handle any other sample that came in from a patient and you just put it in with that batch that day and run it and the same technician runs it so you don't have the hot shot in the lab that is better at this do it. So that way you have an accurate reflection of exactly how many errors might occur under your standard operating procedures.

KEY QUOTE
12 MR. SCHECK:

Now, Dr. Gerdes, you mentioned that the national marrow donor program--that is an external blind?

13 DR. GERDES:

Yes.

14 MR. SCHECK:

Are all the proficiency tests that your lab undergoes from the various proficiency agencies, are all of those external blinds?

15 DR. GERDES:

No, they are not.

16 MR. SCHECK:

Some of them are open tests?

17 DR. GERDES:

Yes.

18 MR. SCHECK:

Now, in terms of the proficiency tests just in terms of the samples that LAPD ran, were those, in your judgment, representative of the kind of samples that were being analyzed in this case?

19 DR. GERDES:

You are referring to the validation specimens?

20 MR. SCHECK:

Well, I'm actually referring to--well, both the validation specimens and the CTS and CAP proficiency tests?

21 DR. GERDES:

All of those specimens represented specimens with adequate DNA. They weren't degraded, they weren't aged or they didn't mimic the crime scene specimens in any way.

KEY QUOTE
22 MR. SCHECK:

And this case involves mixtures of bloodstains, some of which are degraded?

23 DR. GERDES:

Yes.

24 MR. SCHECK:

And were they, in any of the labs that you have looked at DOJ, the tests they took, Cellmark, LAPD, were they doing proficiency tests on those kind of sample?

25 DR. GERDES:

No.

26 MR. SCHECK:

Your Honor, I have no further questions.

Temperature

procedural

Key Quotes (3)

Dr. John Gerdes
All of those specimens represented specimens with adequate DNA. They weren't degraded, they weren't aged or they didn't mimic the crime scene specimens in any way.
The definitive attack on proficiency testing validity — the labs were tested on clean samples, not the degraded mixtures central to this case.
Dr. John Gerdes
No.
Gerdes's one-word answer confirming that none of the labs — DOJ, Cellmark, or LAPD — were doing proficiency tests on representative crime-scene-type samples. The bluntness lands hard as the final answer before resting.
Dr. John Gerdes
If you know you are being tested, you just naturally try a little harder... so you don't have the hot shot in the lab that is better at this do it.
Explains why open (non-blind) proficiency tests systematically understate real-world error rates — labs put their best people on known tests.

Evidence (2)

Informal
NRC (National Research Council) report, page 89 — passage on laboratory error rates and proficiency testing standards
read into record and affirmed by witness
Informal
LAPD validation specimens and CTS/CAP proficiency tests
discussed and criticized as unrepresentative of case-condition samples

Notable Exchanges (1)

Barry ScheckDr. John Gerdes
Scheck systematically walks Gerdes through the NRC's framework for evaluating proficiency tests, then applies it to all three labs in the case, eliciting confirmation that none ran tests on degraded or mixed samples. The examination ends on that damaging note.
strategic

Credibility Attacks (1)

⚔ LAPD Crime Lab / Cellmark / DOJ
methodological critique
Gerdes confirms that proficiency tests run by all three labs used clean, high-quality DNA samples — not degraded, aged, or mixed bloodstains representative of the crime scene evidence — meaning their error rates on case-condition samples remain unknown and likely understated.

Objections

1 objections (0 sustained, 1 overruled)
Proceeding 7941 • 26 utterances • Defense witness
Criminal Trial
Department 103
⚖️ Start
📂 AUG 3, 1995 📄 Direct examination of Dr. John
AUG 3, 1995 KRT DvH TD