CLPEX Chat Board

So what's up with the latest CTS proficiency test?

No one I've spoken to thought it was any different from previous tests, but the results are shocking.

Is this a result of more labs taking these tests for the first time? Are the test being distributed to investigators instead of latent examiners?

The number of mis identifications is an embarassment to us all.

And some of the comments tell me that the college degree requirement in our field isn't working

And what about the single agency that gave this CTS test to 199 AFIS examiners who never look at prints under a magnifier, but who do all their work on screen? And gave them all the test at the same time in a large classroom or auditorium setting? What was that all about? For folks who don't know what we're talking about, go to http://www.ctsforensics.com/reports/teststatus.aspx and scroll down to the "Latent Prints" boxes on the spreadsheet, then click on "16-May-2013"

As you scroll down the PDF page, keep in mind that the blue boxes on the spreadsheet indicate errors.

The defense attorneys will have a field day with this when they discover it. Grab your butt with both hands and hang on tight!

Let's hear it for YQGRY8-516. He/she got a perfect score -- all 12 latent prints WRONG!!! Several other test takers got 11 out of 12 wrong. Are these folks testifying? Are they morons? What's wrong with the agencies who hire them? If you live in their jurisdiction, be afraid. Be very afraid.

From the information supplied by CTS it is apparent that the Agencies/PD’s/ Bureaus etc may not be accepting their responsibility to ensure the candidates are suitable to take such tests.

This result therefore raises some questions; “Does an AFIS operator = a Latent Print Examiner?”

Does the total reliance on computer based systems provide for sufficient skill building in the latent comparison arena?

Do 2 or 3 years as an AFIS operator enable an individual to fulfill the requirements to apply for certification?

Should AFIS operators be classed differently than Latent Examiners?

More clarity from the agency supplying the 199 individuals may be required to resolve some of these and other questions. Speculation will continue with absence of information.

A bit of number crunching for you guys to get your teeth into.

Total number of declared results = 7281 (607 participants x 12 latents, one participant omitted 3 answers)
Totoal number if ID's possible = 5463 (607 x 9)

Total number of correct answers (ID and NI) = 6527 (out of 7281 = 89.64%)

Number of missed ID's = 319 (out of 5463 = 5.84%)

Number of erroneous ID's = 435 (out of 7281 = 5.97%)

I have a bottle of whisky for the first person to testify to a 10.36% error rate in latent prints!

Graham F wrote:From the information supplied by CTS it is apparent that the Agencies/PD’s/ Bureaus etc may not be accepting their responsibility to ensure the candidates are suitable to take such tests.

I'm not sure that this person even knew they were taking a test that day:

97CP7N-516: Please inform us in time and give us lunch

And this quote leads me to believe that they haven't started to "work AFIS" yet:

RG9WVZ-516: Very interesting more especially for the people that are ready to work AFIS.

Unfortunately this person isn't aware that version 515 is the same test - but digital:

NC7YR7-516: My view is that the assessment should be done through a computer not manual, understand that change is a pain but let us allow the change, and used the computer, when use a computer you can able to zoom the print in many ways. [sic]

G6NHVP-516 With 5A and 5H not being linked, the suspect(sic) might not be their niece and nephew, depending on where the prints were lifted also.

I'm thinking this person completely missed the idea behind the proficiency test..... we're not on a CSI episode here.


Perhaps more control needs to be taken by CTS as to who is allowed to take the test - because when we all face the results of these tests in court, there can be huge mis-interpretations of the results of this stuff.

timbo wrote:Perhaps more control needs to be taken by CTS as to who is allowed to take the test - because when we all face the results of these tests in court, there can be huge mis-interpretations of the results of this stuff.

I'm thinking some simple demographics with which to sort the results would solve the problem. How did CLPEs do? How did non-certified LPEs do who are full time LPEs? How did LPEs do who aren't senior enough to take the CLPE test? How did Detectives who are not full time LPEs do, who only look at their suspects prints a few times a year? How did college forensics students do? \

But when we are all lumped together, then serious LPEs and CLPEs have to answer in court for the errors of the detectives and the students.

CTS does provide a disclaimer on the bottom of the first page of the summary.

Steve Everist wrote:CTS does provide a disclaimer on the bottom of the first page of the summary.

True, Steve - but no breakdown of CLPEs / LPEs / detectives / students / well trained monkeys. I agree with Tazman - this information would be very useful. Also an indication from that data of the number of certified folks who made erroneous IDs, and if two or more erroneously ID'd the same print (ie. an erroneous ID rate for this test for the ACE-V method).

The way I read that page is that when the 199 students(?) are taken out, of the 376 remaining, 7 made erroneous IDs (the eighth just forgot to write item numbers, I guess). That equates to 0.15% erroneous ID rate. A further 23 made misses for a 0.51% miss rate. Note that that is individual examiner error rates only (and only if my math is correct), not error rates for the ACE-V process.

Those number are a lot more palatable - still worrying, but more palatable.

timbo wrote: True, Steve - but no breakdown of CLPEs / LPEs / detectives / students / well trained monkeys. I agree with Tazman - this information would be very useful.

Yes, it would be very useful. However, CTS is not in the business of doing studies via proficiency tests. They just make the proficiency test and report the results. They send them to agencies, who determine how they are going to use them and who they are going to give them to. The results are based on the answer sheets that are returned. Different agencies use the tests differently.
We split our office between the two CTS test sets, and we had 9 LPE's take this test and get them all right. But CTS only receives results from one test from our agency. So there are eight that got them all correct that will go unreported because we don't purchase one test for each of our LPE's, and then send them in.

As useful as it could be for the purpose of having data, I doubt that they'd go so far as to include questionnaires to separate out all of the classes of people taking the test.

@Tazman:

I think the simple answer for these results (if you were ever asked in court) is that you don't control nor do you even know who took the test. Anyone, and I mean anyone that has internet access, could potentially purchase and take the test. If CTS would ever release who actually was a part of the results of this test (and I doubt they ever would), it could go to show that it's not just latent print examiners take this. Perhaps a foreign or domestic entity gave this test as a way to screen potential applicants/examiners/(insert anything in here) and they ended up submitting all of the results without thinking about the consequences to the discipline.

No matter how you word it, I think the simple answer is you (if you took the test) got them all right and this shouldn't be used as a measure of the discipline, the methodology, or the examiners that testify.

Hi Josh

I'm not worried about explaining CTS myself and, as Steve implies, if the defense made an issue of the scores, the witness and the prosecution should both reference the disclaimer. Any LPE familiar with the way CTS works should be able to handle this in court. It's the defense attorneys and researchers who will use this to chip away at our credibility with the larger judicial community that I wonder about, and the LPEs who are not familiar with CTS.

I cannot get the report to pull up from the CTS site anymore. Does anyone have the report saved?

So when practitioners start to discuss the results, CTS suddenly removes them from the web without explanation

Free and open discussion is a hallmark of science.

Even if you don't like the results, they should be available to learn from