Friction Ridge Process Map

[ER]

To those that are holding up the OSAC 5 Conclusion scale because...
- They want to replace the word "Identification".

Go with the compromise that was reached in the OSAC document. The evils of the word "identification" are grossly overstated and unsupportable. Progress is in changes to the definition and examiner testimony. Seek progress there. Not on this word.

[ER]

To those that find value in the compromises that were made in writing the OSAC Conclusion standard...

Speak up, and let's move this forward.

[anwilson]

This is a standard for conclusion terms and definitions. Something that DOES NOT EXIST currently. Your arguments would only hold water if the new standard was being compared to an existing standard. It is not. The new standard is being compared against nothing. It obviously wins.

See SWGFAST Standards for Examining Friction Ridge Impressions and Resulting Conclusions (Latent/Tenprint) released on 4/27/2013. First issue was released as a standard on 9/13/2011.

[ER]

By all definitions of the word, that was not a standard. It was not universally adopted by the field (not even close), nor does it come close to the definition of a standard established by ANSI.

Obviously, universal adoption won't happen without significant changes to statutes and funding.

The new OSAC Conclusion standard would be the first STANDARD (as defined by ANSI, who's definition is the only one that really counts) for friction ridge conclusions.


But even if you want to view that document as a standard. Fine. Where were the validation studies before it was published? How does that standard prevent errors? These are obviously ridiculous questions.

[rsc55]

I am at a loss as to why we have so many people dragging their heels on these conclusions.

At least four of these conclusions are currently in use by many (if not most) agencies, just not necessarily with the same words (hence the need for standardized wording).

Source Identification = Identification/Individualization (this semantic debate has been going on for ages. let's just choose one and move on)
Support for same source = Inconclusive with similarities/cannot exclude (which is very vague)/etc.
Inconclusive = nothing to debate here.
Source Exclusion = No change. If you're still using "no ID," I think you're behind the times.

IMHO, the only "new" one is Support for Different Source. Which I could argue is the same as saying "unable to locate" or "inconclusive due to lack of focal point." I'd be willing to bet that there are many labs using particular wording to *imply* this conclusion without any official definition written down. This is the only one where I could *see* a potential argument for needing instructions on how to use it... but how about we agree what to call it first...?

If we can't agree on the same WORDS, how on earth can we expect for someone to draft instructions and take the time to validate said conclusions? If we are waiting for EVERYTHING to be done before we all adopt the same conclusions, we will be waiting a very long time.

Instructions would certainly be valuable, as would validation studies on the conclusions. But the fact remains that we currently do not have either of those, so I don't see why we can't just all agree on the words to describe them, at least that will be an improvement over the nothing we have now. This process will have to go step-by-step. So let's just move step one along.

If you don't like the conclusions, can you propose better ones? Because I haven't heard alternate proposals, just complaints about these. If you're simply dragging your heels because of one word, can you accept that there doesn't exist a single way to make *every* examiner happy, so let's just compromise and move on with bettering our field.

[Michele]

There is a lot of misinformation going around, which only fuels the debate.

First, 4 of the 5 conclusions are not the same as:
ID
Consistent but not sufficient
Inconclusive and
Exclusion

Glenn mentioned that with the new ID 'there is a subtle difference in its meaning'. Do practitioners understand the subtle difference? If not, maybe they should look into it before accepting them.

As I read Glenn's interpretation, Support for Different Source is NOT 'fail to locate'. This shows that the definitions are not clear enough that we all understand them the same. I think with ID, Inconclusive, and Exclusion, at least everyone understands the current use.

I think it would create a lot of issues to accept something that clearly isn't understood by everyone.

Second, it seems like people just want to accept the new conclusions and move on but that's not how the SDO process works. The SDO hasn't even put this out to the public yet. When they do, everyone will get a chance to state the issues they see, but we're not even close to that point yet.

Third, alternatives have been proposed. Instead of expanding inconclusive, there is a bigger need to expand the weight of ID's (Basic, Advanced, and Complex). Agencies are using this and they find it easy to understand and transparently represents information to prosecutors, defense attorneys and the courts. Along with this, it is valuable to not only state something as inconclusive but also state the reason for the inconclusive (that's not a separate conclusion, it's a different reason for the same conclusion).

I could go on with more examples of mis-information, but that's just distracting from the real issue, and that is that these new conclusions are not clearer and more transparent. They actually increase the amount of subjectivity.

[g.]

First, 4 of the 5 conclusions are not the same as:
ID
Consistent but not sufficient
Inconclusive and
Exclusion

No. rsc55 is correct...these 4 map on the current conclusions sufficiently. And she is using them correctly.

The only real 'unfamiliar one' is Support for Different Source. I agree with Michele that 'fail to locate' should be INCONCLUSIVE (although I think a better term is Alice's INCOMPLETE (need additional exemplars). But in practice, Eric and I have already given a good example: Unable to exclude due to lack of focal points/agency policy etc. No correspondences, but can't exclude. If you compare, find no correspondence, but for an agency that has a policy on exclusions then support for different sources would be appropriate.

Glenn mentioned that with the new ID 'there is a subtle difference in its meaning'. Do practitioners understand the subtle difference?

This is soooooo minor and frankly should not be the reason for hanging up this document. The technical nuance is that the new term is trying to label an identification as an LR (extremely strong support) and not a posterior probability (as it CURRENTLY IS, i.e. the person IS the source of the latent print). If what I just said doesn't make sense to you the reader, don't worry, it's not critical to understanding how to report the conclusion. You'll be fine. You're currently using a posterior probability and NO ONE has ever called you out on it....and likely never will.

Third, alternatives have been proposed. Instead of expanding inconclusive, there is a bigger need to expand the weight of ID's (Basic, Advanced, and Complex). Agencies are using this and they find it easy to understand and transparently represents information to prosecutors, defense attorneys and the courts.

Thank-you for informing us here Michele. This is the first I have heard that an alternative was proposed. Was this proposed by the SDO? Could you tell us which agencies are currently using these? I'd like to reach out to them. I have never seen these in a report. In fact, to my knowledge, I don't know any agencies even using the SWGFAST standard which categorizes COMPLEX and NON-COMPLEX conclusions and reporting those (I do in my reports and reference the SWGFAST standard as well). I agree that parsing out IDs into complexity is a very valuable addition. That should not hold up the OSAC proposed conclusions. That can be an ADD-ON to it.

g.

[MSmith37]

I'm all for standardizing the conclusion and reporting language, but I think a lot of people are fearing the new conclusions are not as closely matched to the current conclusions as the "just go with it" crowd is suggesting here.

Glen, I recall from one of your training classes that Carey explained the OSAC's intention of the new conclusions was for more comparisons to fall into the middle 3 categories and only the strongest conclusions to fall into the Source ID or Source Exclusion category. So, a print that was an ID yesterday would become a Support for Same Source today. The red, yellow, and green SWGFAST sufficiency graph was redrawn with a tiny sliver in the top right (dark green) corner being labeled as Source ID and most of the rest of the graph (red, yellow, and most of the green) was labeled as Support for Same Source. A lot of people took strong issue with that.

How that would be policed? I'm not sure. Maybe proficiency tests where a low quant/qual ID was only correct if labeled Support for Same Source? Either way, the general sense I get is that people are righteously concerned that the new standards are intended to "take our IDs away."

[ER]

a lot of people are fearing the new conclusions are not as closely matched to the current conclusions as the "just go with it" crowd is suggesting here

Let me be clear. The new OSAC conclusions map EXACTLY on to what you're already using. It just does so in a standard way. Currently, no one is using conclusion terms in a standard way. Any suggestion that the new OSAC terms are significantly different from what you're already used to does not understand or is not being truthful. And I can prove that. The people here that are telling you that it's significantly different from what you're used to: Are they or were they on OSAC? Did they have ANY part in writing the standard? No.

OSAC's intention of the new conclusions was for more comparisons to fall into the middle 3 categories and only the strongest conclusions to fall into the Source ID or Source Exclusion category
SWGFAST sufficiency graph was redrawn with a tiny sliver in the top right (dark green) corner being labeled as Source ID and most of the rest of the graph (red, yellow, and most of the green) was labeled as Support for Same Source.

This is flatly untrue.

This has been repeatedly stated by OSAC members at a number of conferences, but let me clearly say it here. If you report ID's now, continue to do things the exact same way, but describe the conclusion using the words in this document. If you report "inconclusive with similarities" or "could not exclude", continue to do things the exact same way, but use the term SSS and the definition in this document. If you don't use "inconclusive with similarities" or "could not exclude", then don't use SSS. This document does not require any agency to change anything but the terms and definitions that they are already using. Anyone who says differently does not understand or is not being truthful.

There is another part of this document that goes into more detail about the Identification decision. It precludes examiners from describing Identifications as "Exclusion of all others", "100% certainty", and "0% error rate". The IAI adopted this over a decade ago. Virtually all latent print examiners across the US have gotten on board with this. There is a very small group of LPEs that have continued to teach these outdated and unsupportable concepts. I hope that no one is surprised that they are the same group that is holding up this standard from moving forward.

To summarize:
The people involved in actually writing the new conclusion document are telling you that it involves no changes to WHEN a conclusion is made but only a STANDARDIZATION in terms and definitions. The people that didn't write the new conclusion document are telling you that all of your IDs will now become SSS. They are also the same people that still support "Exclusion of all others". They criticize a STANDARD document for failing to prevent errors.

Who are you going to believe?

Better question: who has to say that the new conclusion document won't change WHEN a conclusion is made and is only a STANDARDIZATION of terms and definitions for the reader to believe this? Who has to say that if you don't already use SSS, you won't have to use SSS?
Let me know. I'll make a phone call. It'll be the next episode of the podcast.

[Bill Schade]

It’s been an informative discussion and debate. It’s nice to see new people posting on this topic which will affect everyone in some way. It’s important that practitioners hear as many views on these topics as possible. As I try to follow and consider the various points of view. I’m sorry to see the post turn angry and become personal. I see strong voices with strong opinions discussing an important topic. I don’t think anyone is lying and it’s unprofessional to suggest that.

Although I’m not on the OSAC committee I do serve on the Standards Development Organization (SDO) that is reviewing the 5 conclusions document (along with a few other documents) Many OSAC members who did write the documents serve on the SDO as well, so that view is well represented. There have been other points of view expressed in that forum over a long period of time. It has been difficult to arrive at a consensus about the wording of the document but all ideas are being weighed and considered.

I personally disagree with some of the statements in a couple of these posts but I think I’ll save the debate for future Consensus body meetings.

I would remind everyone interested in the outcome that these documents will go out for public comments and you will have that opportunity to express your opinion. The first document put forth by the SDO generated over 250 comments.

[g.]

ER,

Sorry, but I know what MSmith37 is referring to and she is correct. Carey Hall does have a slide as MSmith37 described. And it has been discussed in at least one training class I was in by Carey that a CONSEQUENCE of the new OSAC standard could be that some of the more marginal ID/Inconclusive with features in agreement could fall into the "SUPPORT FOR SAME SOURCE" CATEGORY. This is a consequence now of having a clearer definition of what constitutes a SOURCE ID, namely "EXTREMELY STRONG SUPPORT" for same source.

So yes, conceivably, if you had an impression that you really struggled with to decide if it is an ID; or if the verifier doesn't 'think there is enough there'; or there is some 'on the fence' aspect about the comparison...then this would just fall into the SUPPORT for same source category. It actually makes verification a little easier. If it is not readily agreed upon that it is sufficient for ID, it really by definition will fall into SUPPORT category. Whereas with the current conclusions, there could be more debate or perhaps a push to tip it into the ID category.

This is a CONSEQUENCE of defining all IDs as EXTREMELY STRONG SUPPORT. I don't know if it was the INTENTION of OSAC (we need to hear from Carey or anyone from OSAC on this), but regardless, it is the consequence.

And I know this can be a consequence, because before adopting the new standard, I struggled for some of those marginal ones: which category? ID or INC with features in agreement? And occasionally i had cases where I was 'strong support' for same source but thought.....this one is just on the edge of ID. Is it really INCONCLUSIVE with features in agreement? or is it ID? These were ones that fell in the SWGFAST ZONE B, complex zone, bordering Zone A.

The new OSAC standard makes this an easier decision. If I can't get to EXTREMELY STRONG SUPPORT (the definition of ID) then it is just SUPPORT FOR SAME SOURCE. And as posted previously, this is very simple if I use an explicit LR; it is 10^9 or a billion. If I have a subjective LR greater than 10^9 OR one that is based on a model, then I can justify the ID with EXTREMELY STRONG SUPPORT (but as Boyd points out and is true, the model is still informed by my interpretation of which minutiae to enter into the model and where they are located).

So that is where MSMITH37 is coming from and she has a good question. I don't think it invalidates anything in the past, it helps inform the future. It just says look, if it was 'on the line' before, just put it in SUPPORT FOR SAME SOURCE. I think it makes an ID more justifiable.

I agree also with Boyd's/Michele's point, prior to implementation, an agency can do some basic testing or training to ensure that the new scale is used consistently. But I completely disagree that OSAC needs to do that before creating the standard. Just as I don't expect the SDO to test it before approving it. It would be prudent for an agency to do that testing, but it doesn't have to be all encompassing. Hell, I have plenty of test images I can share from (Langenburg, 2009, JFI "Performance Test of ACE-V") that you could use to test the use of the conclusions prior to implementation.

g.

[ER]

g.,

While that is a consequence of adding in SSS, it is extremely important for the community to understand that the conclusion standard does not require an agency to use that conclusion.

Many agencies have already moved to include that decision as an option for their examiners. So, this new conclusion standard isn't what's changing things. They've already made the change and already see the consequences where the borderline IDs are now being called "inconclusive with similarities". The new conclusion standard only changes the terms and definitions used, not when to use the decision.

Many other agencies have not already included this option for their examiners and will choose not to utilize that option when the new scale is being implemented. That is absolutely acceptable and OSAC members from Henry on down have repeatedly told examiners that agencies should adopt the terms and definitions for the conclusions that they use. Therefore, this new conclusion document won't change when examiners make decisions for these agencies either.

MSmith37 seemed concerned that this standard will change WHEN decisions are made. His concern seemed to be that MOST ID's will now be reclassified as SSS. That is the part that's not true. This standard won't make that change. If an agency chooses to add in the SSS conclusion when they haven't previously had it, then there will be some, small number of ID's changed to SSS. But a) it won't be most ID's and b) it won't be because of this conclusion standard.

[Bill Schade]

The new OSAC conclusions map EXACTLY on to what you're already using

I think there is a difference between “Support for Same Source” and “Inconclusive with similarities”.

I believe there is an implication of “neutrality” in Inconclusive with similarities vs an implication that “its him” in Support for Same Source.

And the fact that “support” comes in four flavors (Extremely Strong, Strong, Moderate and Limited) also contradicts the idea that the conclusions SSS and IWS are exactly the same.

[ER]

1. As an examiner that wrote reports with "Inconclusive with similarities" and testified in a Daubert hearing on that conclusion, they are exactly the same. It is not a neutral statement and never was. From Alice's paper in the JFI through AZ's creation of the "Inconclusive with similarities" phrase to it's use in court. It is an associative statement that falls short of Identified. Those that have not chosen to report of testify to this conclusion may not be the best choice for defining its meaning.

2. It is not a fact that "support" comes in flavors. That is an extremely misleading statement. The standard does not express flavors of SSS. That is a choice that Glenn had made for his reports. His reports are not a part of the standard definition of SSS.

[Bill Schade]

1) noted, but I still think inconclusive is more neutral statement than support for "might be an ID". I agree the nuance might be in the mind of the person hearing the conclusion.

2) I didn't mean to be misleading, but without the "flavors" I think we have a problem.
4 similarities in common is not the same as 12. And the type of similarities would have an impact (as Glenn showed in his example on the other thread.) I think that without defining the flavors of support we are leaving out an important part of the equasion.

It's just my opinion and how I'm looking at the implications of the document. I see your point of view and would like to hear others.