ACE vs ACE-V errors

[C. Coppock]

REF: "Recent work by Wertheim, Langenburg, and Moenssens (2006) demonstrated that experts with more than 1 year of training had an erroneous individualization (i.e., accusing an innocent person) rate of 0.034 percent out of all individualizations performed." Examiner 12 July.


At testimony, the public and courts are interested in methodology ACE-V error rates, rather than ACE errors. Essentially, what is the error rate of the completed methodology (ACE-V)? Do we have good error stats on our methodology (ACE-V) errors relevant to sufficiently trained examiners...of say (at least) one year?

[g.]

Craig,

Per se, I wouldn't use that study for calculating traditional error rates (primarily it is useful to show that under those conditions, erroneous id's were few, but the lack of non-matching sources in that study is a problem for calculating standard error rates). You can look to Langenburg, "Method Performance Study" from 2009 (May?, JFI) for traditional error rates using ACE and ACE-V. Although you are only looking at one agency's results (N = 6). I think the FBI's recent black box study (submitted for publication) and a study I just finished, to be presented tomorrow at the IAI (and soon to be submitted for publication), will have some very relevant data for error rates and be helpful for those calculations.

Best to you,

g.

[Neville]

Sorry to burst your bubble but if you honestly believe one year of training is sufficient then it is not surprising there is a problem in the world of fingerprint identification.

I think an error rate of zero is the only option. If there is an error made by a trained person then the person making it should no longer be in the job. Nothing less, whether or not it is a so called 'unsafe' ident (to quote the Scottish heads) or an incorrect ident. I am sure the courts and defence would agree.

[C. Coppock]

Yep, one year is not much, especially in lower volume agencies, but there is rarely a solid line anywhere in this field.

[Pat A. Wertheim]

. . . if you honestly believe one year of training is sufficient then it is not surprising there is a problem in the world of fingerprint identification.

Yep, one year is not much, especially in lower volume agencies, but there is rarely a solid line anywhere in this field.

Craig, you may not be familiar with the training program Neville's country, New Zealand. It takes five years to complete training and become a fingerprint examiner. There are thirteen modules of training. Each module contains voluminous reading material and concludes with a detailed examination on the material in that module. You must pass that exam to move on to the next module. If you stick with the training program for five years, complete all thirteen modules, and do not wash out, then you will have read virtually every book and article of significance ever published on fingerprint development, comparison, and identification. At the end of the thirteen modules, there is a final comprehensive examination that makes the IAI Latent Print Certification Exam look like a kindergarten exercise (I know -- I tried to take the comparison portion and it far exceeds in difficulty the IAI Cert exam). You have, I believe, three tries at the final exam. If you can't pass it, you are back to reading the "Help Wanted" ads in the newspaper, just like you did when you were five years younger.

To become a "Senior Examiner," you have to submit a research proposal to the board that overseas that program. If they approve your proposal, you have to conduct your research and turn in a 10,000 word paper detailing your research and conclusions. It is not a "give away," nor an honor bestowed for years in service. It is a serious process and must be taken seriously by the applicant, as is most assuredly taken seriously by the review board. Neville is a senior examiner.

Neville, you should know that there are few training programs in the world that approach the thoroughness of yours. Certainly in the US, I have seen nothing that comes close. Perhaps if we in the United States had one national police force that served the entire country as you do, instead of the patchwork of thousands of ID Units and Crime Labs of varying sizes with varying training requirements, we might be able to better control the training. But unfortunately, there is no binding national standard here. Between ASCLD accreditation, strong defense attacks, and the NAS report, there is tremendous pressure for us to improve and upgrade our training and our practices. That is happening and we are improving. Hopefully, we will be closer to the standards in New Zealand before too many more years have passed.

Neville, if I am incorrect in my recollections of your training program, please correct me.

[Neville]

Pat it was 10 years ago you were here, with a memory like that you are no dinosaur you must be an elephant. I'm sure I have forgotten more in those 10 yrs than you have remembered.

As I understand it it is now 7-8000 words for a Senior Fingerprint Officer (Sergeant) and 10,000 for a Principle Fingerprint Officer (Senior Sergeant) but a higher standard of submission is expected.

I am back out in the field now as a SOCO and away from the office, so am no longer involved in the idententification side of things and happy as a Kuni kuni in the Tutai (Maori lesson 101 Kuni kuni = Pig, Tutai = Feaces).

I have to say our training is similar to the English and Australian System and like you there is no National Police Force in those countries; also Australia like the USA has different States that run independently, i.e. different parliaments as I understand it.

I am sure if your Police Commissioners really wanted to up the anti in forensics, it could be done. Just need the will to make the way.

[Pat A. Wertheim]

I have to say our training is similar to the English and Australian System and like you there is no National Police Force in those countries; also Australia like the USA has different States that run independently, i.e. different parliaments as I understand it.

I am sure if your Police Commissioners really wanted to up the anti in forensics, it could be done. Just need the will to make the way.

Neville, you will recall in my class there were sixteen levels of difficulty, Level 1 being almost inked to inked comparisons and Level 16 being latents that pushed hard on the line of "no value." I was priveleged to have been invited to train in New Zealand, and as you no doubt recall, every active latent print examiner in your beautiful country attended my class. I have also taught in Australia, and in England and Wales.

The Aussies and the Brits, as groups, were capable of performing at higher levels of difficulty in the exercises than Americans. But if there were a World Cup of Latent Print Examination, Team Kiwi would take the victory.

The biggest impediment to training in the US is the fact that because there is no national standard for training, and because there are around 22,000 independent police agencies, each police agency that wants latent print examiners is free to train as much or as little as they choose. In the large agencies and in accredited crime laboratories, the training programs are usually thorough (if not up to the Kiwi standard). But in many smaller agencies such as the one where I started in the 1970s, a two week course was considered sufficient. That is because in the US, the majority of fingerprint examiners are civilian employees. For the most part, they are paid significantly less than the police officers in their departments, and their job benefits are lower. In most of these departments, the civilians are second rate not only in salary and benefits, but in training and equipment. As long as the system of independent police departments exists and there is no enforced national standard for latent print training, there will be those departments where training for civilians remains one of the lowest priorities.

By the way, I will be back down in Wellington in September for a week for your fingerprint symposium, following a week in Sydney for ANZFSS. Hope to see you and all my other Kiwi friends at one or the other.

[Neville]

Apart from anything I will be on holiday in Europe in September, As much as I would love to there to see you when it comes down to it I think I will stay with my plans, windy Wellington in winter v sunny Croatia coast in summer sorry!

Our Civies are on the same pay and very similar benefits.