Comparison results

Discuss, Discover, Learn, and Share. Feel free to share information.

Moderators: orrb, saw22

16_Points
Posts: 67
Joined: Thu Mar 26, 2009 8:00 am

Re: Comparison results

Post by 16_Points »

Stephany Louk-Denney wrote:Thank you to everyone who has responded so far. It's helpful to receive feedback from others in the field.
16_Points wrote:In another case, I report that mark "A" has minutiae in agreement with John Doe insufficient to identify and that all other subjects were excluded as the source of mark "A".
16_Points, do you write report this way? I'm interested to see if you do and how this holds up in court.
Yes, this is how I report out the comparison results. It has not been a problem in court, yet.

L.J. Steele, in reference to the "other subjects", consider that in most cases the examiner may be asked to compare multiple subjects (ask the investigator why these people were selected). The probative nature of a "partial match" becomes more interesting when you can exclude the remaining persons of interest in the case. It becomes less interesting if the crime scene mark is inconclusive to the general population (i.e. of no value).

In essence, the "partial match" scenario becomes complicated because you are claiming that the subject/defendant is part of a smaller pool of the general population that cannot be excluded as the source. Unfortunately, as Gerald has pointed out, we have no meaningful way of describing the size of this "smaller pool" of people. There is no statistical structure for a random match probability nor a likelihood ratio (not yet anyway). The question, then, in the absence of a meaningful statistical structure, should we report out the findings of inconclusive comparisons?

I am, obviously, saying we should with the caveat that we explain carefully what we know and what we don't know.

Pat, I am not opposed to adding additional conclusions like probable match/probable exclusion. Before doing that, it would be nice to have some well defined criteria for using them and a clear definition of what each means. Perhaps the IAI/SWGFAST/etc. would like to take up the challenge.
Pat A. Wertheim
Posts: 872
Joined: Thu Jul 07, 2005 6:48 am
Location: Fort Worth, Texas

Re: Comparison results

Post by Pat A. Wertheim »

"For Entertainment Purposes Only" (I like that!)
16_Points wrote:Pat, I am not opposed to adding additional conclusions like probable match/probable exclusion. Before doing that, it would be nice to have some well defined criteria for using them and a clear definition of what each means. Perhaps the IAI/SWGFAST/etc. would like to take up the challenge.
The IAI's position (still, I believe) is that there is no basis for a qualified conclusion, i.e., "possible, probable, likely." At the same time, there is pressure to abandon the idea of "absolute, to the exclusion of every other person on earth." Research is being conducted into probability models, but to the best of my knowledge none has yet been published or validated.

I think we have all had comparisons that resulted in a strong inclination to call the ident, but we couldn't quite go there "absolutely, to the exclusion of every other person on earth." I each of us have had a comparison in which we have said(at least to ourselves), "If I just had one more point," and we have spent hours or days with photo enlargements, or in Adobe Photoshop, trying to find that "one more point" so we could say "absolutely."

But in the absence of that "one more point," all we are allowed to say now is "Inconclusive."

So, as the Devil's Advocate here, I ask, "Is that science? Or is it dogma."

If we have that strong feeling but cannot find that "one more point," is it doing the court a disservice by simply saying "inconclusive," or on the other hand, would we be prejudicing the court by saying "probably."

I imagine things will change in the next year or two or five. I expect the IAI and SWGFAST will both play a part in a review and reevaluation of what constitutes good science. I am eager to see how this question is going to settle out.
Pat A. Wertheim
P. O. Box 150492
Arlington, TX 76015
Gerald Clough
Posts: 557
Joined: Wed Jul 06, 2005 6:27 am
Location: Lockhart, Texas
Contact:

Re: Comparison results

Post by Gerald Clough »

Pat A. Wertheim wrote: I each of us have had a comparison in which we have said(at least to ourselves), "If I just had one more point," and we have spent hours or days with photo enlargements, or in Adobe Photoshop, trying to find that "one more point" so we could say "absolutely."
Sure. And very likely, someone else would Individualize without that one more point. And someone else wouldn't be satisfied with the one more. That all competent examiners will conclude similarly is bunk. It just turns out that in most cases it would be hard to find an expert to disagree. (Boy! Is that naive of what?) The truth is that it's always an expression in a very high confidence belief. We can quibble about our confidence in what the absolute truth is, but it's always one degree or another of "probably," just as are even frank scientifically certain conclusions. There's always some tolerance.

If you're pinned to the wall by a knowledgeable expert, you have to agree that there's just no absolute truth. There's only a belief in what that truth is very likely to be or an absolute belief that it's within some range of possibilities. And if another expert disputes your belief, they also have to admit that it's possible you're right. Someone charged with deciding an issue just has to either pick one belief, ignore both, or let some additional body of evidence settle which belief they decide to join. In the realm where our products are used, other participants will be arguing for belief in one or the other. In other words, if you turn it inside out, what makes you so sure that you shouldn't individualize without that one more point?
But in the absence of that "one more point," all we are allowed to say now is "Inconclusive."

So, as the Devil's Advocate here, I ask, "Is that science? Or is it dogma."
And as the Devil's Third Assistant Advocate, I ask, "If you're such an expert, why do others decide for you how to express your findings?"
"Nothing has any value, unless you know you can give it up."
16_Points
Posts: 67
Joined: Thu Mar 26, 2009 8:00 am

Re: Comparison results

Post by 16_Points »

Gerald Clough wrote:
But in the absence of that "one more point," all we are allowed to say now is "Inconclusive."

So, as the Devil's Advocate here, I ask, "Is that science? Or is it dogma."
And as the Devil's Third Assistant Advocate, I ask, "If you're such an expert, why do others decide for you how to express your findings?"
Easy, because they'll fire and/or de-certify me if I don't play by "their" rules . . .
Kathy Saviers
Posts: 48
Joined: Thu Jul 28, 2005 11:41 am
Location: Salem, Oregon

Re: Comparison results

Post by Kathy Saviers »

I use "cannot identify nor exclude" in the instance of "no find."

If I have a tiny section of palm that has plenty of Level Two Details but I have searched and searched and cannot find it, but I am not absolutely certain it is not there, I use "cannot identify nor exclude."

If the latent print is a tip or an edge and the only inked card in the file is really poor quality, I use "cannot identify nor exclude."

I try to qualify my statements by asking for a better rolled fingerprint card or palm print card.

Kathy
Gerald Clough
Posts: 557
Joined: Wed Jul 06, 2005 6:27 am
Location: Lockhart, Texas
Contact:

Re: Comparison results

Post by Gerald Clough »

16_Points wrote:Easy, because they'll fire and/or de-certify me if I don't play by "their" rules . . .
Of course that's pretty much the situation in most, if not all, official forensics. But not the case in much of the wide community of forensic expertise where a great many practice what I might call the "non-IAI" disciplines where I think it's fair to say that most have professional organizations that pretty well stay out of how they opine, other than publishing various articles commenting on methods and terminology. There are limits, of course. An expert can get so far outside the pale that they're deemed to have brought the field into disrepute and committed malpractice. Even if they aren't defrocked by their state licensing authority, being ousted from the national professional organization ends their expert witness work. A psychologist who gets the boot from the APA is not going to find many sponsors. Some of their constraints come from inside the law. For instance, the legal definition of "mental retardation" and it's morphing now into "intellectual disability." A psychologist would have to work within the definition.

My question, "If you're such an expert, why do others decide for you how to express your findings?", was intended to prompt some thought and discussion on the limits of expert discretion, because we're talking here about something that skirts the shoals of the rejection of opinions of confidence in less than absolute Individualization by our quasi-official advisory bodies. I think there's a line that must be walked but a line that must be drawn with care in standardizing and limiting the nature of conclusions. I think there are a couple of primary goals in crafting these things.

One is uniformity and an attempt to create an conclusion space that helps define who is real. That's reasonable, given that, at least in most parts of the U.S., a latent print examiner is anyone who says they are one. (Not just latent print examiners. I encounter various forensic "experts" who put themselves forward with the flimsiest "qualifications.") Very few state laws address the issue and leave it to courts to decide who is an expert. It's rather different in some other expert fields. Rendering an opinion in testimony as a psychiatrist or any flavor of physician, for instance, generally requires a medical license in that state, because the law on medical licensing says you must have one to "diagnose." But remember that the more general perception type experts are usually excluded from testifying on fingerprint matters, not because they aren't IAI certified, but because they haven't conducted research in functional fingerprint issues.

The other reason is, I think, an attempt to limit the scope of conclusions to what can be sustained by consensus and supported by something that can be reasonably articulated. All such efforts are always ponderous. It takes a long time to get consensus. Really, in our field, most of this is articulated in very general ways. There's nothing in any SWGFAST document that speaks of just when an examiner should Individualize in any way that you could reliably use it to say with certainty if another examiner would so conclude from a given latent. In the absence of evidence of clear erroneous fact or frank sloppy examination, a body like the IAI is not going to consider action against one expert, just because of a conclusion that conflicts in a mutually exclusive way with the conclusion of another or many other examiners. The dogma is limited to broad principles. It's nothing really to do with "science." The hard sciences are arenas where competent investigators routinely not only disagree in fundamental ways but regularly engage in open intellectual warfare, as they always have.

But mostly we conform to the standards because by doing so, we're not expressing anything erroneous and because if we get outside those bounds, we risk a fight that our sponsors don't want to have to engage in. I'm not at all saying that's wrong. For every expert who goes outside the standard bounds and potentially contributing some additional legitimate value, there's someone else prepared to breech the same bounds to contribute something bogus.

Dogma has a legitimate place. We just don't want to be blindly autodogmatic. Beware of automaticdogs, too.

Image
"Nothing has any value, unless you know you can give it up."
Patrick Warrick
Posts: 37
Joined: Mon Jul 11, 2005 7:46 am
Location: Minnesota BCA-Northern Minnesota

Re: Comparison results

Post by Patrick Warrick »

I just don't see how reporting probable or possible conclusions are helpful. Noting similarites or discrepancies between impressions are recorded in the bench notes of the examiner. That documentation is available for disclosure to the attorneys and courts. Discussions with investigators and attorneys about the similarities can also be done (and properly documented in the case notes). This may or may not lead to testimoy regarding the observations. But reporting out that there were 4 or 5 features or similarity, without anything really to back that up to what that really means, could be misleading and prejudical. Attorneys and investigators already have their preconceived scenarios about the investigation and may unintentionally (or intentionally) give that statement more weight than it merits. A layperson on the jury doesn't have the training or the previous experience to properly understand the implications of such a statement.
"Rather leave the crime of the guilty unpunished than condemn the innocent."-Marcus Tullius Cicero, Roman statesman (106–43 B.C.)
16_Points
Posts: 67
Joined: Thu Mar 26, 2009 8:00 am

Re: Comparison results

Post by 16_Points »

Patrick Warrick wrote:I just don't see how reporting probable or possible conclusions are helpful. Noting similarites or discrepancies between impressions are recorded in the bench notes of the examiner. That documentation is available for disclosure to the attorneys and courts. Discussions with investigators and attorneys about the similarities can also be done (and properly documented in the case notes). This may or may not lead to testimoy regarding the observations. But reporting out that there were 4 or 5 features or similarity, without anything really to back that up to what that really means, could be misleading and prejudical. Attorneys and investigators already have their preconceived scenarios about the investigation and may unintentionally (or intentionally) give that statement more weight than it merits. A layperson on the jury doesn't have the training or the previous experience to properly understand the implications of such a statement.

I think there is a misconception regarding the reporting of sub-threshold "matches" or "partial matches". This is not an invitation for experts to go off willy nilly and report any three or four minutiae they find in common between a mark and a known print. That's just ridiculous. This scenario presupposes that the mark has some discriminating power that lends it comparative value (e.g. of value for exclusion). Marks that are globally inconclusive to the world's population are not included. Rather it's meant for those situations when you find say seven or eight minutiae in common with good focal points (core and delta) etc. and you need just a bit more for an "individualization". It seems absurd that a mark so close to the margin of individualization is thrown away or marked inconclusive and NOT reported. You are, in effect, saying that there is, "Nothing to see here. Move along." When in fact, there is good evidence for an association between the source of the mark and the source of the known print. Not an identification, but still good strong evidence.

I am not a lawyer. And I leave the lawyering to those who are "experts" in it. If attorneys (prosecution or defense) wish to abuse or misuse legitimate data and findings, that's for the court and respective professional societies to sort out. I do not feel the responsibility to "police" the actions of lawyers.

I can only ask myself, "Is my testimony a fair and accurate representation of the work I actually did?"

I also think it is a mistake to treat the jury as incompetent laypeople. As I mentioned previously, juries are asked to apply complex legal standards and reach conclusions regarding a variety of issues. A jury is just as likely to be inexperienced and untrained in matters of law as they are in matters of science. Nonetheless, the jury is called upon to tackle these issues. It may seem valiant to alleviate a jury of this burden by offering sanitized expert testimony, but it may also be seen as presumptuous. The expert may be with holding information which may prove important or crucial to the finders of fact.

No one is suggesting that this is simple. As I mentioned previously, it would be nice to have an authoritative body appoint a committee of esteemed experts to establish and publish a proposed protocol for determining when and how these "Inconclusive" results are reported.
Gerald Clough
Posts: 557
Joined: Wed Jul 06, 2005 6:27 am
Location: Lockhart, Texas
Contact:

Re: Comparison results

Post by Gerald Clough »

True. It's not really the responsibility of experts to try to work around gaps in the legal system. Let's see if I can put into reasonable terms.
  • - There's factual information about the impressions that's a product of an examination.
    - The examiner believes it is information gained strictly through the examiner's task of comparing the impressions and that the observation is an accurately represents the two impressions and the observed agreements or disagreements between them.
    - The examiner does not observe anything that rules out a common source and therefore cannot declare Exclusion.
    - The examiner does not form a professional belief that they have a common source.
    - If the examiner were to check off one of the three conventional conclusions, it would be Inconclusive, or similarly, if asked if the examiner had formed a belief with regard to the source, the examiner would respond that no such belief was formed.
    - The examiner, then, reports that the examination could neither conclude a common source nor exclude a common source and reports the observations made during the examination in a way that summarizes the data obtained.
I included the broad reporting of observations of both agreement and disagreement, because it seems to me that if we are talking about reporting what could be had by examination, it cannot be limited to only cases where there is agreement without details not in agreement. We originally began talking about depending on bare conclusion terms for meaning and about the relative virtues of terms other than Individualization or Exclusion. If we believe it's useful to report agreement that falls just short of Individualization, we are also bound to report disagreement that falls just short of Exclusion. For instance, the examiner observes a portion of ridge flow consistent with the record impression. Perhaps is looks very much like it could, if the missing portion had been impressed, show that both impressions were whorls, and the Known individual has three whorls. Between the latent and each of the Known's whorls, there is a disagreement. A detail appears in one that does not appear in the other in the position it would have to occupy if they were one of the whorl fingers, and none of the other fingers could have produced the latent. It is exactly the same kind of near conclusion.

I'm just not sure where the reporting/not reporting line would be or if there should be one at all. There's almost inevitably a "nudge-nudge-wink-wink" feel of an unspoken conclusion when anything is pointedly reported, even when that's not what's intended at all. (Arguably, that's right back to the responsibility for use being up to others.) I suppose that's back to why the "official" position is the three conventional conclusions.
"Nothing has any value, unless you know you can give it up."
16_Points
Posts: 67
Joined: Thu Mar 26, 2009 8:00 am

Re: Comparison results

Post by 16_Points »

Gerald Clough wrote:
If we believe it's useful to report agreement that falls just short of Individualization, we are also bound to report disagreement that falls just short of Exclusion.

Indubitably.
Post Reply