Re: Inconclusive Decisions
Posted: Thu Aug 21, 2014 9:18 am
A few thoughts, since this has deviated a little from the original topic.
Inconclusive is an appropriate finding. It is the most transparent and seems to represent a finding that differs in degree (correspondence without reliable predictability) and should be accompanied by a well articulated conflict resolution. (what if there is examiner disagreement between inconclusive and individualization). Lab policies that re-analyze to no value or utilize arbitrary standards, may bias towards individualizations or operating efficiently over thoroughness. I understand not every lab has endless resources, and balancing efficiency and thoroughness has to happen, but it is a reality that should be recognized by the fact that the way your SOP's are written can determine your outcomes and if an outside agency comes in for 'backlog reduction' they might create the appearance of 'errors' because they reversed the findings that your SOP allows for.
Ultimately, the regulating principle of our discipline is the scrutiny of our work by other examiners. Ridge data, in sequential and spatial agreement (or disagreement) don't regulate conclusions, they count as conclusions. It's not just the properties of physical object (friction ridge impressions), but the attitudes of people towards them that regulate conclusions. So if you say "I have never made an erroneous exclusion, individualization, etc" but don't have a 100% verification policy, it's kind of meaningless. The best way to minimize error is to detect it.
RE: exclusion to all others.
There's a little circular reasoning going on here. The thing we call the law of biological uniqueness first uttered byQuetellet is the result of observations both from examinations in the forensic realm as well as the biological sciences. Therefore, we would not appeal to the law to justify the law. (induction never justifies itself). You can't say effectively, "I know this observation is unique because uniqueness is defined as an observation of the unique" Therefore because individualization is not biologically nor observationally determined, the best you can say is "I have never seen, nor would I expect to see, this amount of similarity in friction ridge impressions that came from different sources.”
RE: Sufficiency
Because data in sequential and spatial agreement doesn't determine conclusions asking 'how much is enough' or at what point does an inconclusive become an individualization is not one merely of quantity or quality for that matter, but one of social obligation in the form of 'when would someone else bring this back to me because they'd ID/exclude it?' That is testable. Give it to someone (blindly or not, sometimes one is preferred over the other) and let them scrutinize it. When people converge on the same result, after scrutinizing the data, it strengthens the reliability of the conclusion while admitting uncertainty.
More of a nightmare from a polciy standpoint in my mind is an inconclusive generated from AFIS and whether or not to report it.
Inconclusive is an appropriate finding. It is the most transparent and seems to represent a finding that differs in degree (correspondence without reliable predictability) and should be accompanied by a well articulated conflict resolution. (what if there is examiner disagreement between inconclusive and individualization). Lab policies that re-analyze to no value or utilize arbitrary standards, may bias towards individualizations or operating efficiently over thoroughness. I understand not every lab has endless resources, and balancing efficiency and thoroughness has to happen, but it is a reality that should be recognized by the fact that the way your SOP's are written can determine your outcomes and if an outside agency comes in for 'backlog reduction' they might create the appearance of 'errors' because they reversed the findings that your SOP allows for.
Ultimately, the regulating principle of our discipline is the scrutiny of our work by other examiners. Ridge data, in sequential and spatial agreement (or disagreement) don't regulate conclusions, they count as conclusions. It's not just the properties of physical object (friction ridge impressions), but the attitudes of people towards them that regulate conclusions. So if you say "I have never made an erroneous exclusion, individualization, etc" but don't have a 100% verification policy, it's kind of meaningless. The best way to minimize error is to detect it.
RE: exclusion to all others.
There's a little circular reasoning going on here. The thing we call the law of biological uniqueness first uttered byQuetellet is the result of observations both from examinations in the forensic realm as well as the biological sciences. Therefore, we would not appeal to the law to justify the law. (induction never justifies itself). You can't say effectively, "I know this observation is unique because uniqueness is defined as an observation of the unique" Therefore because individualization is not biologically nor observationally determined, the best you can say is "I have never seen, nor would I expect to see, this amount of similarity in friction ridge impressions that came from different sources.”
RE: Sufficiency
Because data in sequential and spatial agreement doesn't determine conclusions asking 'how much is enough' or at what point does an inconclusive become an individualization is not one merely of quantity or quality for that matter, but one of social obligation in the form of 'when would someone else bring this back to me because they'd ID/exclude it?' That is testable. Give it to someone (blindly or not, sometimes one is preferred over the other) and let them scrutinize it. When people converge on the same result, after scrutinizing the data, it strengthens the reliability of the conclusion while admitting uncertainty.
More of a nightmare from a polciy standpoint in my mind is an inconclusive generated from AFIS and whether or not to report it.