CLPEX Chat Board

Please vote and comment.

Hello, all. I am taking a survey.

This is a hypothetical situation. A department has two echelons of employees in its latent print unit: (1) Latent Print Examiners and (2) Latent Print Specialists, also called AFIS Techs; the pay scales are different for each of the two positions. The unit has an AFIS system hooked into a statewide network. The AFIS entries are done primarily by two Latent Print Specialists/AFIS Techs. According to the SOPs and Quality Manual, AFIS Techs do NOT make identifications—identifications must be made by the Latent Print Examiners. Reports are written by the Latent Print Examiners.

These hypothetical ‘preliminary identifications’ would be released with a ‘disclaimer’ saying that these are preliminary results, subject to confirmation by Examiners, and that the results are strictly based on the ‘computer comparison’ and in no way constitute a confirmatory identification.

Where to start, where to start? Okay, according to ASCLD regs, formal reports must be written to document comparison results. Are these "Preliminary Identifications" formal reports, with everything required by ASCLD? If not, then they cannot be allowed. Surely your SOP's follow ASCLD regs with regard to formal reporting of conclusions.

You say in your post that your SOP's do NOT allow for AFIS Techs to make identifications. In the Latent Print world, there are no preliminary identifications, subject to verification by Latent Print Examiners.

And...what if they're wrong? I know no one wants to admit that that can happen, but it does. Especially among those who are not formally trained to competency. Do your AFIS Techs take proficiency tests every year? I mean Latent Print Proficiency tests? If not, they cannot report latent print conclusions. (ASCLD, again).

All in all, this is a BAD IDEA!

Why is there any need to refer to them as "identifications" at all? It's really just the results of the tech screening machine output. Which brings me to...

I presume you mean they report their result when they can throw out prospects that are obviously not the latent source while one appears to them to match and report a completely negative result when all the prospects are cast out. Which brings up the question of how decisions are made to cast those out. They're probably right, but how do you answer a question about the qualifications to decide there is no match among the candidates? Can you say someone is qualified to elliminate one or more of the records resulting from an AFIS search but also say they are not qualified to make a conclusion of identification?

If I play the knowledgable defense attorney or their expert in a case in which an AFIS run produced results that were all thrown out by the tech, (in other words, no print evidence for court) I would ask how we know the real actor isn't one of those elliminated by the tech. If the expert knows his business, he will challenge the effective ellimination and will use our own rules of practice to do it. Ellimination is also a conclusion.

And, of course, if they're reporting all the AFIS product as "preliminary identifications," is sounds even worse, although it probably isn't as bad, since there is no conclusion, merely poor language. There's quite enough confusion among legal commentators between AFIS and LPE identification, as demonstrated by the article in Judicature last year.

You can play all sorts of scenarios, but it all comes down to the fact that the techs are just presenting the examiners with something to analyse, and that's probably just the way they should be represented, as something they got by operating the machinery and as something about which they make no conclusion aside from that fact.

Quite apart from the training and qualifications required to examine the prints, this scenario seems to be proposing the release of identifications to investigators without verification.

I say no on those grounds alone.

I wear both hats in my position.... I'm the 'AFIS tech' when I'm searching prints on behalf of the "Forensic Investigator ( Sworn members trained in Identification Philosophy/Principles/Methodology.. )" , and I'm the 'Latent Examiner' when I search prints on behalf of the "Scenes of Crimes Officer (Sworn members trained only in basic print recovery)" where I would perform all subsequent work (except Verification) on the submission including Analysis, (AFIS if required), Comparison...Evaluation..Reporting....Charting...Presentation at court.

When I'm acting as the AFIS Tech, I get the word out to investigators by submitting a report stating words to the effect that ..." a search(es) has been performed and that, as a result, I furnished the Forensic Investigator with a 'potential person of interest' for hard-copy comparison...." . This way the Detectives working the case can know that there ARE results, and they can liaise with the Forensic Unit to ensure that the case is completed by the F.I. in a timely fashion. If there is urgency to the case and the F.I. is unavailable, the Supervisor could have the latent reassigned to another member trained to competency.

If the case ever goes to court, my statement would say that "at no time have I compared latent print..(R1)... to any hard-copy set of fingerprints.." which is generally enough to have me excused from the case.

Gerald Clough wrote:You can play all sorts of scenarios, but it all comes down to the fact that the techs are just presenting the examiners with something to analyse, and that's probably just the way they should be represented, as something they got by operating the machinery and as something about which they make no conclusion aside from that fact.

This is my understanding, though I'm not familiar with ASCLD regulations, I am familiar with this type of procedure.

The LPS/AFIS Tech might work the older cases (years of backlog) or minor felonies. The latent prints are evaluated by the LPE's and then analyzed and submitted into AFIS by the LPS. The LPS would compare the candidate list and identify potential candidates. The LPS would finish their evaluation and forward it back to the LPE as a "possible candidate" who would then verify the results and submit the final report.

From my experience with AFIS, any user can view the candidate list of any other user. So the LPS is essentially completing the labor involved by searching the prints through the local and state database and if the LPE has a question or concern, they have access to review the LPS work.

Maybe it's the terminology used but I don't understand why the results are 'strictly based on the computer comparison' when AFIS requires an examiner to manually compare and effect a decision. I have faith in LPS who know how to properly and accurately compare and identify latent prints. Therefore I believe it is acceptable for them to identify "possible candidates" and make their recommendations. That is if I'm understanding the hypothetical question correctly.

The point was that, if someone is looking at two images and making a decision that they do not match, they are expressing a conclusion, ellimination, that is reserved for latent print examiners. I know that, in practice, any that are disregarded are almost certainly disimilar from the latent in every way except that the system's algorythm places them on a list of possibilities.

My question was how would one answer the question, "How do you know the true actor was not one of those included by AFIS but discarded as apparently not matching." To illustrate, (and Charles can correct me if I misremember) in think of FIG 006 on the Fingerprint Interest Group Page. I think he told me he nearly rejected the suspects record print because of the severe distortion in the latent. If someone who has been doing this work since just after opposable thumbs came out thought it was easy to miss, how can I answer the question by citing the conclusion of someone who would not be allowed to render any official conclusion?

It's simply a matter of who renders official opinion on comparison conclusions. "Knowing how" will not stand up in court. I "know how" to compare shoe marks, but I do not possess the range of technical knowledge of shoe sole manufacture and other matters nor a body of training and experience to be an expert. I've been right every time I believed I had identified a suspect shoe with the scene mark, but I have no business either testifying or using my conclusion in a warrant affidavit.

Now, I think it would take an obsessive and knowledgable defense attorney to raise the issue when there were latents from the scene, candidate from AFIS discarded, and no other suspect identified by fingerprints. But the rarity of challenge doesn't validate the procedure.

And I should point ouf that I have never worked directly with AFIS and can't speak with any authority to day-to-day practice. What I described seemed, though, to me to be what the original post described.

Gerald Clough wrote:I presume you mean they report their result when they can throw out prospects that are obviously not the latent source while one appears to them to match and report a completely negative result when all the prospects are cast out. Which brings up the question of how decisions are made to cast those out. They're probably right, but how do you answer a question about the qualifications to decide there is no match among the candidates? Can you say someone is qualified to elliminate one or more of the records resulting from an AFIS search but also say they are not qualified to make a conclusion of identification?

If the answer is 'Yes' to eliminate and 'No' to identify 'possible candidates', you've set up a double standard. Shouldn't the AFIS Techs display enough technical if not practical experience and aptitude before they're hired to perform said duties?

Gerald Clough wrote:The point was that, if someone is looking at two images and making a decision that they do not match, they are expressing a conclusion, ellimination, that is reserved for latent print examiners. I know that, in practice, any that are disregarded are almost certainly disimilar from the latent in every way except that the system's algorythm places them on a list of possibilities.

That's why it would be prudent for the LPE to check the candidate list of the LPS and the AFIS 21 makes that conveniently possible. But how is the conclusion reserved for LPE only? Maybe on paper and the final report the conclusion will come directly from the LPE but who effects the conclusion first? I understand the concern but isn't this just a nice way of saying that LPS's are not as experienced or as good as an LPE's.

I think the bottom line is: no conclusions should be released outside the laboratory setting until they are FINAL conclusions...the case has been thoroughly examined according to the lab's SOP's and the conclusions are final, and have been verified.

As a practical matter, of course AFIS Techs come to elimination conclusions and identification conclusions. But those conclusions should never be reported to the Investigator until the case has been thoroughly analysed. AFIS searching provides probable cause for an LPE to make a manual comparison of the latent prints to a specific person. When the LPE comes to the conclusion of identification, he/she provides probable cause for the Investigator to focus on one specific person.

I don't believe any ASCLD or ISO-certified laboratory would ever release "preliminary identifications." And yes, Justice Pie, AFIS Techs are usually NOT as proficient or technically capable in LATENT PRINT EXAMINATION as LPE's. No one is "better than" any other...they simply have different skill levels and do different things.

I kind of like the idea of letting AFIS technicians tell the investigators their preliminary conclusions before a qualified LPE has a look. That way, the gung ho investigators could make earlier arrests. What a bonus that would be for me, as a defense expert. First question I would tell my attorney to ask the police expert: "Do you follow the accepted procedure for fingerprint examination in the United States?"

And with those cases where the preliminary conclusion is wrong, think of all the extra money I can make from the civil suits!

First question I would tell my attorney to ask the police expert: "Do you follow the accepted procedure for fingerprint examination in the United States?"

Name one accepted procedure for the entire United States?! HA! That makes me laugh. Yes there are ASCLD and ISO Labs that follow their guidelines and there are suggested guidelines by SWGFAST. But there is nothing beyond ACE-V that is used throughout the United States, in every agency, or in every lab.

Yes our industry wants to act like every id is performed exactly the same, every time an id is performed, but the truth is - they aren't. Especially when different agencies are involved, with each agency having their own SOPs.

It's simply a matter of who renders official opinion on comparison conclusions. "Knowing how" will not stand up in court.

Are you familiar with Rule 702 of the Federal Rules of Evidence? It states that knowledge, skill, experience training and education has to be more than the common person.
A judge is hesitant to deny someone from being an expert.

Additionally, Gerald admits to never having worked with an AFIS system in day to day practice. Were you aware that each agency makes their own settings as to what scores will and will not be reviewed by their examiners (or technicians, or whatever their ten-print operators are called)? This is referred to as their AFIS system's threshold settings. These settings often eliminate a fair amount of missed identifications by the system. Yet we still continue using these AFIS systems. Computers can not testify in court, and judges have already upheld that anyone qualified to be classified as an expert can review the candidates offered by an AFIS system and make the actual determination of id or not.

By the way, I do have over 11 years of experience in our field, so I feel that I am just as qualified as the rest of you to make the statements I have made. And I disagree with you all ganging up on Justice for his/her comments.

I am new to this board and I look forward to the discussions.