CLPEX Chat Board

Two articles, and then my questions are at the end.
Michele

http://www.chicagotribune.com/news/loca ... 4011.story

State police point finger at prints analyst
By Steve Mills, Chicago Tribune reporter
9:03 p.m. CST, January 6, 2012

The Illinois State Police said Friday that it had reassigned a fingerprint analyst and is reviewing a number of the analyst's cases in Cook and other counties.

The agency was vague on details but said the cases do not involve misidentifications but rather raise questions about quality control. The issue was detected in a routine quality control review, said state police spokeswoman Monique Bond.

Bond said that officials had not determined how many cases might be involved but that the department's division of forensic services has notified prosecutors of the problem.
"Right now what they've found is a potential issue," Bond said. "What that means, we won't know until they complete further analysis."

Bond would not identify the analyst, but said the person works in the Chicago lab and has been reassigned while the inquiry is being conducted.
"We don't know the extent of it yet," said Bond.

Cook County prosecutors were notified of the situation and have begun to review the cases they were told are at issue, said Sally Daly, a
spokeswoman for State's Attorney Anita Alvarez. "We are aware of this and it's something that could be of concern to us," said Daly.

smmills@tribune.com

http://www.aledotimesrecord.com/news/x1 ... s-analysis

The Illinois State Police Division of Forensic Sciences reviews analysis
GateHouse News Service
Posted Jan 09, 2012 @ 11:31 AM

Springfield, Ill. —
The Illinois State Police (ISP), Division of Forensic Services, Forensic Sciences Command employs a strict internal quality control program to ensure all ISP forensic scientists maintain proficiency in their forensic specialty and that the quality of their case work is consistent with our high standards.

Through its Quality Assurance Program, the Forensic Sciences Command has identified a potential quality concern that could impact some latent fingerprint analysis. Specifically, the Forensic Sciences Command has identified a quality concern regarding work performed by one of ISP’s forensic scientists. The ISP is currently performing a comprehensive review to determine the scope of the concern and is taking measures to remedy any effect it may have. The cases under review do not involve any fingerprint misidentification.

The ISP has notified all the State's Attorney offices for which the scientist analyzed evidence to make them aware of the situation. As part of our remediation process, analysis previously performed by the scientist will be reviewed, and reanalyzed, as necessary. Results of any reanalysis will be reported to the agency which originally submitted evidence to the laboratory, per our normal reporting protocol.
Copyright 2012 Aledo Times Record. Some rights reserved

These articles are very vague so I'm just speculating...
... I'm reading this and thinking, (edited to indicate my true meaning) it’s great when quality control measures catch errors after they happen but shouldn’t we strive to have quality assurance measures that prevent errors from occurring, or to find the errors right when they occur instead of after they are reported out? I think using verification as a full review, instead of as someone reproducing the conclusion, might find any problems right when they happened and would avoid having to notify the prosecutor’s office and avoided getting named in the paper. I know it's hard to say since we don't know what the errors are but I can't think of error that wouldn't have been found with verification as a full review process.

I didn't see the word 'errors' in any of those articles.

... and by errors, I mean issues

SThurman wrote:I didn't see the word 'errors' in any of those articles.

I googled news on the situation and found specific mention that NO erroneous identifications were made. However, no hint was given regarding the issue at hand. I have seen similar situations because of a missed identification, either in casework or on a proficiency test, which I hardly think merits an audit, much less a newspaper article.

Pat wrote: I googled news on the situation and found specific mention that NO erroneous identifications were made. However, no hint was given regarding the issue at hand. I have seen similar situations because of a missed identification, either in casework or on a proficiency test, which I hardly think merits an audit, much less a newspaper article.

The article mentions the issue came up as a result of a routine quality control review. There are a lot of layers here. Not knowing their policy, and assuming mis-ID as the issue, it could be that they don't verify non-ID's. If that's the case, there's no way of finding a miss other than a review. If that's the case, how often do these routine reviews occur? If it's once a year, then a problem that could have been identified early could become a larger issue. Thus a decision to conduct an audit.

If this is the case, then the agency may be at fault because they don't have rigorous enough quality control (and verification) procedures in place so as to catch something like this early, when an audit would not be needed. So you may be right that the reason for the audit may not generally have merit, but because of their policies and procedures, an issue has arisen. If that's the case, then the agency would carry some blame, and may have to look into adjusting their procedures as a result.

Michele, I love you. You're a friend. But I gotta get this off my chest, since you are the first to criticize soooo many of these other agencies that land in the papers (and involve lots of my other friends and colleagues in this business, e.g. FBI, USACIL, Vermont, Illinois (a couple of times), etc.).

1. Perhaps scientists shouldn't speculate so much until they have full data/information to properly make an assessment. That tends to prevent errors (I mean issues) too....

2. King Co. better have the best damn SOPs in the land that prevent all errors and always reach consistent (and accurate) opinions amongst scientists. And everything should be so clear and spelled out in them that every scientist in the agency knows exactly what is expected of them and can always meet that with no ambiguity. [which i know is not the case since I have read posts where KCSO folks have said "here's what we do, but it's not actually in any of our policies"]. And those KCSO SOPs should clearly state when they are following SWGFAST standards and when they are NOT.

3. How do you know that ISP DOESN'T do full technical reviews in case reviews and looks at notes, worksheets and how the scientist reaches a conclusion?

4. I can think of PLENTY of instances where a full technical review (as you describe) would NOT catch some errors until audits are done (which for the record ARE quality assurance measures--spot checks, audits, external/internal reviews are all part of the process to catch different types of errors). To give an example during an AUDIT here, we found a scientist was incorrectly making a chemical. It still worked fine (it was a dye stain) so all control tests and worksheets and batch numbers were fine and unremarkable during the case review...it wasn't til we went to a reagent log book during an annual audit did we see the error. There are other instances that I suspect, even scientists at KCSO during audits, will find errors that were not caught during a full technical review.

5. The notion that a full technical review will catch all errors is unwarranted, overidealistic, and just plain silly. The notion that full tech reviews and looking at the basis for conclusions, etc. will prevent erroneous decisions OR as you've claimed before, lead to consistent (consensus) justifiable results is also unwarranted. Please present the data that shows this.

6. Glass houses. Stones. Just sayin'...

g.

Glenn,

Doesn’t the fact that they notified the prosecutor’s office, transferred the employee, and gave a press conference indicate that ISP considers this significant? As experts in this field, isn’t it proactive to speculate on problems and try to come up with solution before errors happen? Why is speculating seen as criticizing instead of being seen as noticing something that can be improved? Does it matter if people are our friends? Does that mean that we’ll overlook things and only scrutinize the people who aren’t our friends?

I don’t see this as an issue of ‘my sop’s can beat up your sop’s’. I think it’s an issue of constantly looking at how to improve.

If ISP does complete reviews on everything (which they might), wouldn’t the reviewer bear some responsibility for not finding the errors? That will be interesting to hear. You’re absolutely right, it may not be an issue that casework review would have found. I was basing that on the articles that mentioned casework, but you bring up another issue that we can all learn from. When an error could be considered significant, maybe the agency should have audits more often (you mentioned an annual review).

I don’t understand why SOP’s need to say if the agency follows SWGFAST or not. If an agency has this in their sop’s then they have to commit to reviewing and changing their sop’s every 6 months.

Number 5, “The notion that a full technical review will catch all errors is unwarranted, overidealistic, and just plain silly.” You’re right. That’s why I said “might have found”... If I’ve indicated that it WILL prevent ALL errors someplace, “please present the data that shows this”.

My point is that regardless what this issue is about (in Illinois), we can use the article to learn from and speculate about how to improve our practices pro-actively instead of waiting. We don’t need to sit back and wait for the final results before we implement improved policies, we may never know the entire story. Good QA should look for errors (issues) before they become big problems. Good QA should also looks at the agency’s responsibility in the error, instead of just blaming the examiner.

How can you even comment on the situation without any facts?

From Steve Everist post; Not knowing, assuming , it could be, If that's the case, If this is the case, again, If this is the case. Some real critical thinking skills here.

From Michele’s most recent post:

‘isn’t it proactive to speculate?’ What, to assume it to be true without conclusive evidence?

‘we can use the article to learn from and speculate about how to improve our practices pro-actively instead of waiting.’ Without any factual data to examine, I guess we could read just about any article and speculate on how we can learn from it. I am reading the The Musgrave Ritual, and I speculate the Butler did it.

If their QA found the issues, then their QA is working.

That does not necessarily mean that they have problems at the review or policy level.

We'll probably find out soon enough.

I'm going to speculate that this thread is done.

Zippity do da day.