CLPEX Chat Board

I'm happy to see this document now out for the community to digest. Thank you to the FRS for publishing the draft version, and I hope that it can become official in some capacity soon.

Here are a couple of my personal thoughts on it.

First, I know there will be an initial shock to having 5 levels. However, this is not a major departure from the last SWGFAST draft document where the two "support" conclusions were present but just included within the inconclusive category. I think that it's totally appropriate to separate them out (just like every other forensic discipline).

Second, nowhere in this document does it say that YOUR agency HAS to use or regularly report all 5 levels with the "support for different" conclusion being the one that I haven't yet used. I've known that this was coming for awhile, and I think I'm ready to start using it in situations where there are no similarities, where the exemplars are complete enough, but where the latent doesn't have a delta or core. It'll definitely require new training and new protocols, but I think it's better than lumping this situation in with regular INC and definitely better than excluding without a core or delta.

I hope that the community can look for the positives and benefits from this standard. I know it's more difficult, more nuanced, and more complicated. But that's what latent print comparisons are. Our decisions used to be black and white (just like the ridges and furrows). While some comparisons are definitely that easy, many are more complicated than that. (And hey, the field decided way back in the 80s that black and white images in AFIS systems wouldn't work. We needed all those shades of grey... or at least 256.)

ER, thank you for drawing attention to this draft document.

The NIS OSAC FRS draft "Standard for Friction Ridge Examination Conclusions" codifies two new conclusions that may result from a comparison. In addition to Exclusion, Identification, and Inconclusive, we now have "Support for Different Sources" and "Support for Same Source." To me, those two "new" conclusions are not entirely objectionable.

However, the first bullet point in the "Qualifications & Limitations" section mandates that "An examiner shall not assert that a source identification is the conclusion that two impressions were made by the same source."


So, now we are supposed to testify that, "It is my conclusion that my observations provide extremely strong support for the proposition that the impressions originated from the same source and extremely weak support for the proposition that the impressions originated from different sources."

REALLY???

In order to hopefully save people some clicking around here is the link to the page with the documents. It's going to take me some time to digest them.

ER wrote: ↑Tue Jul 17, 2018 5:42 pm Second, nowhere in this document does it say that YOUR agency HAS to use or regularly report all 5 levels with the "support for different" conclusion being the one that I haven't yet used. I've known that this was coming for awhile, and I think I'm ready to start using it in situations where there are no similarities, where the exemplars are complete enough, but where the latent doesn't have a delta or core. It'll definitely require new training and new protocols, but I think it's better than lumping this situation in with regular INC and definitely better than excluding without a core or delta.

The "support for different" jumped out at me, also. I can't think of a time I would have used it as a result of the balance of support tipping towards exclusion, but also having some support towards "source identification." The area in your description above [emphasis mine], doesn't sound like it would fit the category as the draft defines it. It sounds more like a "not identified" conclusion, that could be the result of ambiguity in searching. In our office, we refer to this inconclusive determination as "inability to locate corresponding detail due to ambiguity."

The reason I don't think it fits that is based on your comment in the other thread, where you wrote:

In my personal experience in erroneous exclusions, almost all examiners making this error, when shown the proper orientation, location, and a corresponding target group will find sufficient similarities to identify.

The way I'm reading your comment is more of a statement regarding ambiguity, not one that tips the balance supporting exclusion. Like I said, I can't remember a situation where I'm thinking that I wished I had this option.

Thanks for providing that link, NRivera. I meant to do that...

Well, I was hoping to focus on some of the more positive aspects of this standard first, but yes, you've zeroed in on the sentence that causes me the most concern too.

So let's look at the entire section here because I think it helps to look at the entire picture here.

5 Qualifications & Limitations
● An examiner shall not assert that a source identification is the conclusion that two impressions were made by the same source or imply an
individualization to the exclusion of all other sources.
● An examiner shall not suggest that the offered conclusion is an expression of absolute certainty.
● An examiner shall not assert or imply that latent print examination is infallible or has a zero error rate.
● An examiner shall not cite the number of latent print comparisons performed in his or her career as a measure for the accuracy of a conclusion offered in the case at hand.
● An examiner shall not use the expression ‘reasonable degree of scientific certainty’ or similar assertions as a description of the confidence held in his or her conclusion.

I hope that most people reading this agree with bullet points 2-5 above and even the second half of point 1. If my agency decides to adopt this standard, then we will have to change some of our report wording. Overall, that the easy part. Currently, we say, "Latent A was identified as having been made by Tom Jones." We could instead say, "The comparison of Latent B and the FPC bearing the name Sally Smith resulted in an identification decision." Or go to a bullet point structure like "Latent C - FPC bearing the name Sledge Hammer: Source Identification." Semantics, but fairly easy changes. (Quick aside: the full text of the definition of the source identification conclusion would be reported to the customer in some way according to policy.)

Court responses get a little more difficult because we can't really control what questions are asked of us. I'm kind of stream-of-consciousness writing here, but here's a go at responding to court questions. If these aren't the way that attorneys phrase questions in your jurisdiction, then please post on that.

Prosecutor: Did you conduct any fingerprint comparisons in this case?
Me: Yes. I compared 5 latent prints developed on the beer can to a FPC bearing the name Roger Robot.

Prosecutor: And did you reach any conclusions in those comparisons?
(Now here I used to say.)
Me: Yes. I identified the 5 latents prints to the FPC bearing the name Roger Robot. (Or) Yes. I identified the 5 latents prints and the prints on the FPC bearing the name Roger Robot as having come from the same source.
(Now I'll have to learn to say.)
Me: Yes. My conclusions in those 5 comparisons were identifications (or source identifications, still trying to understand the significance/difference/importance of that word).

Prosecutor (or maybe defense on cross?): Why did you reach an identification decision?
Me: In each separate comparison I observed significant correspondence between the latent print and the exemplar print. I also did not observe any significant differences between the latent and the exemplar.

Prosecutor (that's trying to be thorough?): And what does identification mean?
Me: Identification means that the amount of correspondence strongly supports that these two fingerprints came from the same source and that there is very weak support that these fingerprints came from different sources.

Defense: So it's possible that this latent print came from someone else, not Mr. Robot?
Me: The amount of correspondence between these prints, the clarity of the details that correspond, and the lack of significant differences means that I would very much not expect to find this level of correspondence in another person, not Mr. Robot.

Prosecutor (on redirect): Who left that latent print on the beer can?
Me: When I compared the latent print to the card labeled Roger Robot, I reached an identification opinion (or conclusion, or decision, or just identification).

For me, this change seems do-able. It'll take some practice and caution with word choice. Every examiner will have to work out their own version of this for it to sound comfortable. I know that the way I might phrase it isn't how other people feel comfortable speaking. We all make our own word choices in testimony, and we all have to connect with juries from different areas. Overall, I don't think it's going to drastically change my testimony. I'll start off simple by just saying identification and follow up with more clarity if and when there are follow-up questions. I'll encourage the prosecutor to ask about the ID definition, but I can't control exactly how he does it.

@Steve Everist

I get your point, and like I said, "I think I'm ready". As it's new, I'm trying to figure it out. First off, part of the reason behind this level is so that we can better integrate decision terms with other comparison disciplines.

Even so, I can think of comparisons where I "know" that it's not him, but I'm not sufficiently confident in the differences to say exclusion. If that sounds foreign to you, try thinking of a comparison where you "know" that it IS him, but it's just not enough to call an ID. That concept has been around forever. So, when do I think this, should I communicate this to the trier of fact, and how do I develop a protocol on this?

"Should I communicate this?" - Absolutely, yes.

"How do I develop a protocol on this?" That's tomorrow-ER's problem.

"When do I think this?" You're right. Probably not every time where "there are no similarities, where the exemplars are complete enough, but where the latent doesn't have a delta or core." Maybe it's more like when I know orientation and location, but don't have a core or delta? In that situation, I'm pretty confident that it's not him because there's no real ambiguity of where to look but I know I need that core or delta for the sufficiency threshold. If I DO have a delta, but I don't know orientation and I don't know finger or palm, then right now I'm excluding. But maybe it's more appropriate to say that there is support for different?

What do you think?

As examiners start using 'Support for Different Sources' and 'Support for Same Source' conclusions, I wonder how often we will be on the wrong side of the 'actually was not made by - actually was made by threshold.' I predict it will be significantly more often than we are now being on the wrong side of the 'actually was not made by - actually was made by threshold' when we use Exclusion/Inconclusive/Identification conclusions. JohnV

I feel like these conclusions represent a fundamental shift in the discipline's line of thinking which is good. Prior to this I felt that we were playing a language game that used an increasing amount of technical words to describe something with decreasing resolution, all based upon the criticism du jour. It felt comical and that's probably because it's a meme, and it looks like this:
.

What I do like about it is that it seems to be moving into a likert scale which is marrying the two essential components of what we do, the data and the belief of the examiner, it's a psychometric scale essentially.

The three issues I see which are still of concern are as follows:

Likert Scales can be prone to Central Tendency Bias.(This is basically John's point above) Since the weight of the conclusion has been an ongoing concern, will that concern be echoed if the conclusions are under-weighted due to this bias?

I think the central tendency bias may be exacerbated by the fact that there's not enough information about how to rate the data in order to get you to the various buckets. While the document recognizes this, it seems like maybe this is a cart before the horse type scenario.

There's still a vestige of the language game we were playing before that, in essence says a source identification doesn't mean I identified the source of this print.

ER wrote: ↑Thu Jul 19, 2018 10:45 am Even so, I can think of comparisons where I "know" that it's not him, but I'm not sufficiently confident in the differences to say exclusion. If that sounds foreign to you, try thinking of a comparison where you "know" that it IS him, but it's just not enough to call an ID. That concept has been around forever. So, when do I think this, should I communicate this to the trier of fact, and how do I develop a protocol on this?

We were discussing the same thing. The counter to this conclusion is, I'm guessing, the most common inconclusive scenario. It's currently one of our accepted conclusions and requires documentation to support it.

However, having the counter to the "support for different sources" conclusion doesn't mean I've had... um... the counter to the counter. As I mentioned earlier, when looking at that potential conclusion, I don't think of a print that I wish I would have had that conclusion available. However, this may be a result of our exclusion guidelines and having an inconclusive-ambiguity conclusion available.

If I DO have a delta, but I don't know orientation and I don't know finger or palm, then right now I'm excluding. But maybe it's more appropriate to say that there is support for different?

What do you think?

I'm actually surprised at this, especially since "I couldn't find it/target groups" doesn't sound like support for different. And as I quoted you above, if someone (verifier/tech review/etc...) showed you the proper orientation, then you may find sufficiency for identification. And as written, the draft states "Any use of this conclusion shall include a statement of the degree of support and the factor(s) limiting a stronger conclusion." I'm trying to figure out what that documentation would look like.

Regarding "Support for Different Sources," I was also having trouble imagining specific situations where I would feel comfortable using this conclusion. As it happens, I came across a great one today. Imagine a latent impression with an obscured/distorted core and clear ridges above and around it with a pretty tight recurve. The print basically screams "not-an-arch," yet due to the distortion, you may not be confident enough to exclude based on that. Now imagine the suspect has all arches - plain, no tented arches. No focal point, nothing to really set your anchor on, yet you can't find it and it just doesn't seem to fit. I would argue this is the perfect instance to call "Support for Different Sources." We're basically talking about a Level 1 detail conclusion.

I'm curious to see where this takes us, though. Where does our threshold for suitability for comparison fall now? Would I call a blurry whorl with a single Level 2 detail suitable because I can come to a "Support for Difference Sources" conclusion? As it is now, I wouldn't call that suitable, as I'll never ID the thing, and I tend to stay away from excluding something with such limited detail. How many more latents will we be analyzing? What's the utility? Ultimately, I think this will be up to individual agencies to determine based on workload and personnel.

On another note, I too am weary of the first bullet point for "Qualifications & Limitations"

An examiner shall not assert that a source identification is the conclusion that two impressions were made by the same source or imply an
individualization to the exclusion of all other sources.

My concern here is that we're playing a game of semantics (when aren't we?). When we testify, shouldn't we by trying to make the evidence clear for the lay-people in the courtroom? If I say "I have determined that the two impressions were made by the same source," isn't that much clearer than saying "I have come to the conclusion of Identification"?

Going back to ER's sample court questions, it seems as if we are talking in circles as opposed to simply answering the question. Yes, I believe the two impressions were made by the same source. That is the purpose of being able to offer an expert opinion. I feel that as long as we can properly qualify our limitations, there should be nothing inherently wrong with this statement.

When I initially read the bullet point, I was hoping it was just a very confusing way of emphasizing that we can't say ...to the exclusion of all others. Since these are combined into one bullet point, it seems to me that they are saying we can't say "Two impressions were made by the same source to the exclusion of all others" or "individualization to the exclusion of all others." Otherwise, I would have expected this to be split into two separate bullet points.

So my question for anyone on the inside - What exactly did you mean by this statement? And if we are in fact not allowed to say that the impressions were made by the same source, shouldn't the second half of this be completely separated?