erroneous ident question

[Ashley]

I was posed a question this morning that I found interesting and thought I would ask your opinion about. If you have a latent print that is said to be and ident and it infact is not an ident it is erroneous. But, what if you have a latent print that is said to be an ident, but it should have been ruled as inconclusive what is it called then? Since they could techincally be correct.

[sharon cook]

[Carl Speckels]

A difference in opinion. Was the 'ident' verified?

[16_Points]

I don't think you can say that the comparison "should have been inconclusive" unless the implication is that the latent was clearly insufficient or of no value. If the latent is clearly insufficient or no value, then you can argue that an incorrect identification was made (Erroneous Identifications I take to mean, specifically, identifying the incorrect person). However, latent print conclusions are highly subjective (i.e. based on an examiner's experience). So there can be a large "gray area" between one examiners "No Value" and another examiner's "Sufficient for Identification". The real question becomes, does your laboratory have a conflict resolution protocol and are these differences of opinion reviewed and recorded?

[sharon cook]

[David Fairhurst]

I'm going to address the opposite case from Sharon.

What if the latent really is insufficient?
What if you pass it around your lab and everyone there says it just doesn't have enough clear detail to be individualised?
What if you send it to another agency and everyone there says it's insufficient?
What if you post it here and no-one can identify it?

What if one person then contacts you and tells you who made the print?

They could be right. After all, even if we cannot tell who it is, a print can only have one donor. What if this one examiner has got the right guy?

I'd say it's a lucky guess.
I'd say it's not repeatable.
I'd say it's not scientific.
I'd say it's not admissible.

[Charles Parker]

Ashley I am going to throw my 2 cents into this thread. Keep in mind that what I say might not be main stream, and may not fit your agencies SOP’s or work environment (a new phrase I learned today---I like it).

Example: Let us say you have completed an examination and you have a latent print with eight ridge events in sequence and agreement and your conclusion is that both the latent print and inked print are from the same source. Now you take it for verification and the other examiner states that see some level of agreement (6 ridge events) but they have a problem with two that you have indicated. Saying that two are too close to the edge for them to consider or it is in a light area and they cannot say where the ridge actually ends. For whatever reason they only see 6 ridge events and say there examination is inconclusive----cannot exclude but do not have sufficient clarity and quantity to individualize (identify). Are you wrong----is that other person wrong----I do not think so as there is a sufficiency issue.

Now some are going to say----well you don’t really have and ID until you get it verified you dumb old man. OK we will play that tune. So now you take it to a third examiner and after careful study they agree with you that it is identified with exemplar. Is the 2nd examiner wrong----no you have a sufficiency issue! If you take it to the third and they agree with the second are you wrong---no you have a sufficiency issue!

So you see if you really have scenario where the issue is sufficiency then neither is wrong. However if the scenario is different and it is obvious that the LPE is just digging their heels in to be stubborn like a mule----then the issue is training. Of course you do know how to get the attention of a Missouri Mule?

Also in this country there are currently two philosophies dealing with sufficiency. One is that sufficiency is based upon individualization and the other is that sufficiency is based upon comparison. They may sound similar but they are not. Suitable for comparison treats exclusion at the same level as identification. This means you might have a latent print suitable to exclude but does not contain the level of information to identify.

The other thing is that some agencies (examiners) only use inconclusive when dealing with sufficiency issues of exemplars but do not and will not apply a sufficiency issue to a latent print. In other words they will only state inconclusive if the exemplar is of poor quality but never on a poor quality latent print. It either has value to identify or it does not----no middle ground.

16 Points said

“However, latent print conclusions are highly subjective (i.e. based on an examiner’s experience).

16 Points I am going to have to say I disagree. Latent print conclusions are objective as they are based upon “external—verifiable—phenomenon.”

16 Points said

“The real question becomes, does your laboratory have a conflict resolution protocol and are these differences of opinion reviewed and recorded?”

Some type of conflict resolution is good but my question would be if these two came to conflict resolution and the conflict was resolved would there be a need to record it. Both parties are now in agreement and issue is moot—so why record a resolved issue? Now if the conflict is not resolved then by all means record the different opinions.

The reason I ask is that I know of one agency that if a conflict is resolved it is not recorded only those that are not resolved. I guess it goes back to that subjective transparency thing.

[Graham F]

Sharon, I think you may have to go back to bed and sleep on it once more (meowwww!!!) 8)

If someone comes along later and says it is inconclusive, then that person is wrong--either not qualified to give an opinion, or paid by the defense to muddy the waters.

Just as others are paid by the prosecution side to overstate their evidence. The first examiner and the verifier could overstate the comparison as an identification, when it should have been inconclusive and the independent third party may well be correct. If all prosecution identification evidence were correct, there would have been no Mayfield, or McKie debacle, albeit these were erroneous cases, not inconclusive.
Let’s face it, what makes an examiner an examiner, or verifier – depends if they have had any training, experience and the area where they are employed. Many examiners know of others in the game that have had limited, or in some cases no training. They could have been pressured by their work environment, investigating officer, DA’s office, or just plain old bias. Just ask Mr. Dror.

[David Johnson]

Ashely originally asked

But, what if you have a latent print that is said to be an ident, but it should have been ruled as inconclusive what is it called then?

If you are just asking what do you call it then it really does not fall into the typical Type one or Type two categories. This is mainly because we dont really know ground truth.

So you would just have to call it was it is... an unverified identification by the first analyst. It would then go through your agencies conflict resolution process.

There was a Hall and Player, Will the introduction of an emotional context affect fingerprint analysis and decision-making?, where they gave a large number of analysts the same print and it shook out that some people ID'ed it, some were inconclusive, and some said no value. In this case they know ground truth so they can say that the people who did not ID the print made a type 2 error.

In case work we have no idea so you can not assign error in that sence. As a group your agency will have to determine if the first or second person is in error and what significance you attach to that.

[Neville]

Many years ago we had an expert who would result jobs as unsuitable because we had such a massive backlog, he could plough through great stacks of work in a day, the boss thought he was great???? However we later searched some of those same jobs and got hits. Does that mean they are not idents then; these idents were verified by two other experts.

Some times at the end of a very long day I have resulted something as not identifiable and the very next morning identified it and had it checked by two other experts, does that mean it was not an ident.

I doubt it; No one is perfect no one can say that's not me.

After 8hrs at a scene with poor light have you never looked at a lift and said that is great I will take that and when you get back to the office realise it is a pile of old ... of cause you have, this is just the other side of the same coin.

It has perhaps something to do with bias but more likely to do with being human. Isn't that why our work is checked.

I do not have a problem with something being resulted as unsuitable and then being identified, it is only a problem if you think it is. That may be because it is your decision to make it unsuitable and some one else thinks it is identifiable. Once it has been identified and every one agrees with the ident, then it can't be unsuitable, surely.

[16_Points]

Example: Let us say you have completed an examination and you have a latent print with eight ridge events in sequence and agreement and your conclusion is that both the latent print and inked print are from the same source. Now you take it for verification and the other examiner states that see some level of agreement (6 ridge events) but they have a problem with two that you have indicated. Saying that two are too close to the edge for them to consider or it is in a light area and they cannot say where the ridge actually ends. For whatever reason they only see 6 ridge events and say there examination is inconclusive----cannot exclude but do not have sufficient clarity and quantity to individualize (identify).

. . .

So now you take it to a third examiner and after careful study they agree with you that it is identified with exemplar.

16 Points said

“However, latent print conclusions are highly subjective (i.e. based on an examiner’s experience).

16 Points I am going to have to say I disagree. Latent print conclusions are objective as they are based upon “external—verifiable—phenomenon.”

You can't have it both ways. Are conclusions objective? Are they based on external-verifiable-phenomenon? Examiner A says there are eight "ridge events" in agreement. Examiner says there are six "ridge events" in agreement. So which is it? Just saying it's a sufficiency issue is a bait and switch tactic. Objective, external, verifiable phenomenon are either there or they are not. They are real, or they are not real. Otherwise they aren't objective, external, and verifiable. So I don't buy your argument. Then Examiner C comes along and says, "Yeah, sure, I see eight ridge events in common. It's an ID." Yep, I can see the court brief now, "Two out of three latent print experts agree it's an identification . . ." Sounds like those toothbrush commercials I used to see as a kid. "Two out of three dentists recommend Oral-B to their patients."

Objective implies measurable criteria for decision making. I see a minutiae in a poorly developed area of the latent. Do I count it? Well, what is the signal to noise ratio? When I was in the university, we used a simple formula to determine if a spectrographic peak would "count". The peak had to be at least three times larger than the background "noise". If yes, you count it. If no, then you don't. DNA analysts, if I understand correctly, do something similar.

So what are the criteria for measuring "ridge events" (I call them points)?

[Charles Parker]

16 Points, for years I struggled with the concepts of Subjective and Objective in Friction Ridge Analysis. Some of the current authors would state this part was subjective and this part was objective. I would lean one way and then read something and go the other way. It always seemed strange to me and contradictory to the science courses I took in college. I always thought if one part of the issue was subjective then the whole thing was. But then I read an article by William Leo published in the Journal of Forensic Identification [58 (1), 2008], titled “Subjective”—The Misused Word. A light went on with that article and I do not struggle with it anymore. You see William Leo’s position is not based upon the lay definition of subjective or objective but the legal definition. Since court testimony is an important part of my job duties, I use the legal definition. Here is the legal definition as stated in Black’s Law Dictionary, Eighth Edition.

Subjective: Based on an individual’s perceptions, feelings, or intentions, as opposed to
externally verifiable phenomena.

Objective: 1. Of, relating to, or based on externally verifiable phenomena, as opposed to an individual’s perceptions, feelings, or intentions <the objective facts>. 2. Without bias or prejudice; disinterested.

Mr. Leo also had a little test at the end his article which goes:

The Fingerprint Examiner’s Subjective Test

1. Is your conclusion based on A or B?
A. An examination and comparison of the features of the prints
B. Your personal perception, feeling, or intentions.

2. Can your conclusion be verified?
A. Yes
B. No

3. Can you show or demonstrate to another examiner what you examined that provided you with the information that allowed you to form a conclusion?
A. Yes
B. No

If you answered “A” to the above three questions, you have proven that the conclusion of a friction ridge comparison is an objective conclusion, not a subjective one. If you answered “B” to any or all of the above three questions, perhaps you should re-evaluate how you are doing friction ridge examinations, or you could become a palm reader.

I would suggest reading his article. If you do not have access to the JFI, I have his article in PDF, just e-mail me and I will send you a copy.

Now let me see if I can answer some of your posts.

As far as the quotations you made of my post. You forgot an important part of the Triad (External Verifiable Phenomena) and that is verification. The examples had not gone through verification.

16 Points Said

You can't have it both ways. Are conclusions objective? Are they based on external-verifiable-phenomenon? Examiner A says there are eight "ridge events" in agreement. Examiner says there are six "ridge events" in agreement. So which is it?

Conclusions that have been verified are objective based upon the legal definition. If you want to use the lay definition, by all means do so.

16 Points Said

Just saying it's a sufficiency issue is a bait and switch tactic. Objective, external, verifiable phenomenon are either there or they are not. They are real, or they are not real. Otherwise they aren't objective, external, and verifiable. So I don't buy your argument.

You will have to explain more why you think a sufficiency issue is a bait and switch tactic. It is not that they may or may not be there, they very possibly could, and the observer just cannot see them (cannot see an atom can we—yet). Have you ever done an examination where you see 16 points and the other examiner sees 17 but you have the same conclusion? Sorry you do not buy the argument---others agree with it and I am sure you can find others that do not.

16 Points Said

Then Examiner C comes along and says, "Yeah, sure, I see eight ridge events in common. It's an ID." Yep, I can see the court brief now, "Two out of three latent print experts agree it's an identification . . ." Sounds like those toothbrush commercials I used to see as a kid. "Two out of three dentists recommend Oral-B to their patients."

Do you really feel that is a good analogy? But to stay on the issue of the point you are making, have you heard of the problems with an agency in California because they did not report differing conclusions. I guess your commercial is coming to be true---in California.

16 Points Said

Objective implies measurable criteria for decision making. I see a minutiae in a poorly developed area of the latent. Do I count it? Well, what is the signal to noise ratio? When I was in the university, we used a simple formula to determine if a spectrographic peak would "count". The peak had to be at least three times larger than the background "noise". If yes, you count it. If no, then you don't. DNA analysts, if I understand correctly, do something similar.

So what are the criteria for measuring "ridge events" (I call them points)?

You stated “However, latent print conclusions are highly subjective”, and I thought that is what we were talking about conclusions and not “criteria of measuring points”. Some types of measurement can be subjective but I am speaking of objective conclusions from a legal definition.

If you want to use the lay definition or the one you were taught in college then by all means do so, as I have no objection to that. Just do not force me to state your position because you believe it. You have to testify according to your principals and I have to testify according to mine and I have no problem with stating that the conclusions derived from friction ridge analysis are objective according to the legal definition. In fact I have done so on several occasions. Everyone in my office also agrees with Mr. Leo’s article and testifies that it is objective as well.

Really, I encourage reading William Leo’s article as he is far better at stating it than I can give it justice here. You cannot tell it might give you a reasonable alternative to your current belief.

Thanks 16 Points for the posting.

[16_Points]

I feel like I have hijacked this thread, that is drawn it away from it's original intent (Which, coincidentally, is why I try to stay off the boards). So this will be my last post. Rather than go point by point, let me just say this. Members of the latent print community need to make a decision. Are we scientists that "do" science or not? If so, then we should be concerned first with scientific methods such as observations based on quantifiable measures and evidence based conclusions.

You can bandy about "lay" and "legal" definitions of "objective" all you like. Focus on the science. Conclusions must be based on evidence. Good evidence is measurable and documented for posterity. Do not confuse the "signal" (ridge event) with the "detector" (LPE). The latent lift doesn't change. If Examiner A sees 6 points and Examiner B sees 8 points, it is a detection issue. Either Examiner A is not sensitive enough to detect the additional 2 points, or Examiner B is reporting a false signal as real (two points that aren't really there). They cannot both be right because the signal (latent print) is the same. The detectors need to be calibrated, meaning some type of measurement criteria (such as signal to noise ratio). There is nothing "lay" about this; real scientists from chemists to nuclear physicists do this everyday. This is what scientists do. I get enough rhetoric from Simon Cole and Michael Saks. It's time to put it away and get back to the nuts and bolts of the scientific process such as signal detection, measurement criteria, and calibrated decision thresholds.

[ER]

16,

I can see that you have the best of intentions for the latent print discipline. But for all the discussion in this and other threads about what direction our discipline should take, I think that you're drawing a line in the sand that doesn't have to be there. The world of latent prints isn't as black and white as you're making it.

"Members of the latent print community need to make a decision."

Just because you disagree with Charles doesn't make his methods unscientific. Just because he doesn't document things the way you would like to see them done or to the extent you would see them done, doesn't make his conclusions invalid.

"Are we scientists that "do" science or not? If so, then we should be concerned first with scientific methods such as observations based on quantifiable measures and evidence based conclusions."

The question should come down to this: Can another qualified latent print examiner look at the evidence and the notes, understand what was done, and come to the same conclusion?

If yes, then we have good science. If no, then take better notes. It shouldn't be any harder than this.

The scientific process as we were taught in grade school is a bunch of hooey. Nobody 'does' science like that. You look at the evidence and make a conclusion. If other qualified people can look at the same evidence and reach the same conclusion, then publish it. There is no requirement that every scientist has to be measuring "signal detection" or "calibrated decision thresholds" to reach valid conclusions. That's ridiculous.

To strap us down and burden us with the straitjacket of over-documentation in the hopes that we'll be considered 'more scientific' is a pipe dream (even if some agencies are doing it). If we change things as you suggest, the egg-heads and defense lawyers that are shouting the loudest will just find something else to shout about in a couple years. If we keep doing good work without changing, they'll still move on to something else when this issue is no longer on the headlines.

[Kasey Wertheim]

Wow. OK.
So there may be more follow up on that, but here's an attempt to get back on track...

The original question provided the assumption that the comparison "should have been ruled as inconclusive". So staying with your premise - that this truly is an inconclusive, and nobody should identify it - I have seen a couple of suggestions.

Call it:

Carl: "Differing opinions"
16 points: "Incorrect identification"
Charles Parker: "Erroneous sufficiency determination"
David Johnson: "Unverified ID"

The original scenario said it should be ruled as inconclusive. Therefore, I'm going to expound on that - that in fact either that latent print or the known print it was compared to was insufficient for identification purposes, or there was not enough common area in order to make a conclusive determination of identity. Regardless of which of those 3 possibilities it was, in the scenario an examiner called it an ID when they didn't have enough information to make the call. There isn't additional information about how this was known - this is an assumption of the scenario that it "should have been ruled as inconclusive", so I'll accept your assumption and propose "invalid", "incorrect" or "inappropriate" as potential synonyms.

-Kasey