|
G o o d
M o r n i n g !
via
THE WEEKLY DETAIL
Monday">
|
_________________________________________ |
__________________________________________
Breaking NEWz you can UzE... |
|
compiled by Jon Stimac |
|
Fingerprints Reveal Injured Suspect's Identity –
FT WORTH STAR TELEGRAPH, TX - July 6, 2007
...prints tipped off police to the real name of a wounded man who was
dropped off at a hospital...
Fingerprint Science Centenary Marked
–
YORKSHIRE POST, UK
- July 5,
2007 ...a century of forensic science is being celebrated
at a museum with a new exhibition...
Judge OKs Request for Fingerprints of Slaying Suspect –
SOUTH
BEND TRIBUNE, IN
-
July 2,
2007 ...judge granted permission to
prosecutors to fingerprint accused murderer...
FBI Vet: Fingerprint Matched Suspect's
–
CHICAGO SUN TIMES, IL
- July 2, 2007
...retired FBI fingerprint analyst testified that a fingerprint from
reputed mobster appeared on an application... |
__________________________________________
Recent CLPEX Posting Activity |
Last Week's
Board topics
containing new posts
Moderated by Steve Everist |
|
Statistics and Misidentifications - The
weeks Detail
Michele Triplett 9300 Sun Jul 08, 2007 3:53 pm
The Lockerbie Connection.
Iain McKie 4188 Sun Jul 08, 2007 3:43 pm
Hypothetical Scenario
Charles Parker 334 Sat Jul 07, 2007 8:13 pm
SOP for crime scene work
Wayne Reutzel 473 Sat Jul 07, 2007 4:01 pm
Interesting Tidbit 5
Charles Parker 412 Sat Jul 07, 2007 2:45 am
General Electric v. Joiner
Charles Parker 502 Wed Jul 04, 2007 3:42 pm
tell all job description for LPE in Colorado
sandra wiese 650 Wed Jul 04, 2007 2:52 am
Surgically Altered Fingerprints
Pat A. Wertheim 527 Tue Jul 03, 2007 6:06 pm
Suggeested Reading
Bill 255 Tue Jul 03, 2007 3:08 pm
Highlands County Sheriff's Office---Sebring, FL---Hiring CST
J Fennell 122 Mon Jul 02, 2007 3:39 pm
Latent Print Examiner Positions - CONUS/OCONUS
wkpetroka 1019 Mon Jul 02, 2007 7:32 am
2007 CLPEX.com T-shirt Slogan Contest Entries
clpexco 330 Mon Jul 02, 2007 2:37 am
SWGFAST Standing Tenprint Committee
clpexco 112 Sun Jul 01, 2007 1:23 pm
(http://clpex.com/phpBB/viewforum.php?f=2)
|
UPDATES ON CLPEX.com
Updated the Detail Archives.
_________________________________________
We continued a series on latent print reporting,
the process, conclusions and error. We discussed milestones along the
path of crime scene to the Analysis phase of ACE-V that can negatively
affect what would have been a conclusive outcome.
We conclude the series with
the second half of a discussion of errors in the latent process.
_________________________________________
Errors in Latent Print
Examination - Part 2
by Kasey Wertheim
The final exercise in our series on latent print process is to consider
error for all possibilities involving sufficient similarity or sufficient
dissimilarity for a conclusion to 2 possible ground truth states: matching
or non-matching prints. To accomplish that, we have arrived at this
point in the process having determined the following state:
A latent print is considered suitable for comparison by that examiner. Upon
a comparison of the latent print, it may be determined that in fact the
impression is not suitable for a conclusion. However, the definition
of "suitable for comparison" is set upon the principle that the act of
comparing a print establishes that it is comparison quality. Whether
similarity or dissimilarity is suitable for a conclusion is the subject of
this week's detail.
The known print is considered suitable for comparison by that examiner.
Generally, the vast majority exemplar prints are suitable for comparison.
Suitable common area and orientation has been achieved.
For the 8 possible scenarios below, it is critical that we have already
determined these 3 conclusions. In some cases we will proceed to
comparison and evaluation and cycle back to our hypothesis involving
orientation and common area.
Sometimes during the comparison itself we encounter additional information
that changes our evaluation. All
subsequent steps in the latent process would be repeated and the new
information would be incorporated into the hypothesis and decision making
process.
For the 2 hypothesis sets of evaluation listed as Hs6, 8 scenarios are
possible:
For matching prints:
A1) Correct match
A2) Incorrect match
A3) Correct Inconclusion - similarity
A4) Incorrect Inconclusion - similarity
For non-matching prints:
B1) Correct exclusion
B2) Incorrect exclusion
B3) Correct Inconclusion - dissimilarity
B4) Incorrect Inconclusion - dissimilarity
For the presence of
similarity, the examiner's hypothesis and null hypothesis are:
Hs6B: Sufficient Q/Q of similarity (agreement) of detail does not exist to
establish individualization.
H0: Sufficient Q/Q of similarity (agreement) of detial exists to establish
individualization.
For prints that
match, if the
examiner accepts the null hypothesis by concluding that there is agreement
to individualize, two scenarios are present: Either the examiner was A1)
correct or was A2) incorrect. If the examiner rejects the null hypothesis
by concluding that there is not agreement to individualize, then two
scenarios are present: Either the examiner was A3) correct or was A4)
incorrect.
A1) Correct Acceptance of Agreement to Individualize - The examiner
has correctly identified a print that matches. There is no technical error
- the null hypothesis was correctly accepted. The examiner acted in an
acceptable manner. Real world result: The right bad guy is convicted.
Agency action: none necessary. No recommendations.
A2) Incorrect Acceptance of Agreement to Individualize - The examiner
has incorrectly determined there was agreement to individualize. The print
could NOT and should NOT have been identified by that examiner based on
his/her ability level, but it was. The ground truth was that the prints
were made by the same source, but in reality the examiner was not able to
confidently arrive at this conclusion. There is technically no error
because the prints actually were made by the same source. The key is that
the latent print in question was compared and was therefore considered of
comparison quality by that examiner, but the final evaluation should have
resulted in an inconclusive determination instead of a match determination
based on the ability of the examiner. If the examiner had more ability,
this might become scenario A1, but the examiner did not have that ability
because accept of the null hypothesis was in error. A possible
scenario might be the only print in a major case and the examiner "pushed
the envelope" and guessed correctly. Another possible scenario might be an
examiner who knowingly verifies an identification that is outside his/her
comfort level. In statistics, this would be considered the more serious
"Type I" error of commission. The examiner incorrectly accepted the null
hypothesis when it was actually false. However, in reality this type of
error would not necessarily be discovered. Furthermore, this scenario would
be difficult to prove, since no valid and reliable methods are widely used
to quantify and compare examiner ability and latent print difficulty.
Besides, who would say anything... it was, in reality, a correct "guess".
There is scientific and statistical (type I) error. The examiner acted in
an unacceptable (and in fact an unethical) manner. Real world result: The
right crook goes to jail. Agency action: difficult to discover, document,
and take action – so most likely, none would occur. But it is
recommended that agencies be aware that this problem can exist and take
measures to address this scenario in policies and procedures with defined
methods of discovery through QA/QC with appropriate corrective action.
Disciplinary action may be considered in this scenario because the examiner
knowingly acted in an unethical and unprofessional manner.
A3) Correct Rejection of Agreement to Individualize - The examiner
has correctly determined that there was not a sufficient volume of
similarity present in sequence to individualize. The print could not and
should not have been identified by that examiner based on his/her ability
level, and in fact it was not. There is no error because the identification
was not within the ability range of that examiner. Technically the print
was made by the same area of the same source being compared, but the
similarity that was present was not sufficient and therefore the match was
not 'called' by the examiner. THIS SHOULD NOT BE CALLED A 'MISS'. A miss
would occur when that examiner could have and should have made the match but
did not orient the print correctly or otherwise compare the area in
question. In this scenario, it is true that the prints were made by
the same source. It is also true that the examiner was looking at the
correct area of the known print of value, and they saw similarity that was
present. However, in the opinion of the examiner, based on his/her ability
including training, experience, etc., it was simply not enough to
individualize. This is OK!! The print may have been valuable for
exclusion, and may have even been excluded to other fingers in the case by
the same examiner. Although of value for comparison, the print in
question should have been deemed unsuitable for purposes of
individualization, and it was - it was not identified. Distortion may also
have co-existed for which that the examiner could not or would not account.
In the field of latent prints, there is no error; the print is considered
not suitable for individualization by that examiner and
the examiner made a
correct determination for their ability level.
In statistics, there is no error - the examiner correctly accepted their
hypothesis From an ethical perspective, the examiner acted in an
acceptable manner. Whether or not this level of performance is
acceptable is either an issue that should be addressed in agency
policy/procedure, or a decision that supervisory personnel should be
prepared to discuss and decide on a case-by-case basis. Real world result:
The right crook does not go to jail. Agency action: either accept current
performance level or provide additional training to increase performance
level. Also, either report the comparison as inconclusive or assign it to
another examiner within or outside the agency, depending upon agency policy.
If procedures are in place to deal with this scenario and the impression is
identified by other examiners, the real world result may change to the right
crook going to jail. Disciplinary action for the examiner in this
scenario is not warranted. Corrective action may include additional
training or assignment to other duties depending on the history of the
individual.
A4) Incorrect Rejection of Agreement to Individualize - The examiner
has incorrectly determined that there was not a sufficient volume of
similarity present in sequence to individualize. The print could and should
have been identified by that examiner based on his/her ability level, but it
was not. Possible scenario: the suspect was a friend, acquaintance or
family member and the examiner was not objective. Another possible
scenario: the examiner was asked by an officer or co-worker not to identify
a print that matched. THIS IS AN INTENTIONAL ERROR. In statistics, this
would be considered the less serious "Type II" error of omission - the
examiner incorrectly did not accept the true null hypothesis when they
should have. The examiner acted in an unacceptable and unethical
manner. Real world result: The right crook does not go to jail.
Agency action: hopefully verification or QA/QC would reveal this error.
Proving that an
examiner intentionally called a print "no value" or "inconclusive" when they
knew it matched would be difficult to prove, since no valid and reliable
methods are widely used to quantify and compare examiner ability and
comparison difficulty.
It is recommended that
agencies be aware that
this problem can exist and take measures to address this scenario in
policies and procedures with defined methods of discovery through QA/QC with
appropriate corrective action. Disciplinary action may be considered
in this scenario because the examiner knowingly acted in an unethical and
unprofessional manner.
For the presence of dissimilarity, the
examiner's hypothesis and null hypothesis are:
Hs6B: Sufficient Q/Q of dissimilarity (disagreement) of detail does not
exist to establish exclusion.
H0: Sufficient Q/Q of dissimilarity (disagreement) of detail exists to
establish exclusion.
If the examiner accepts the
null hypothesis by concluding that there is disagreement to exclude, two
scenarios are present: Either the examiner was B1) correct or was B2)
incorrect. If the examiner rejects the null hypothesis by concluding that
there is not sufficient disagreement to exclude, then two scenarios are
present: Either the examiner was B3) correct or was B4) incorrect.
B1) Correct Acceptance of Disagreement to Exclude - The examiner has
correctly excluded a print that in fact does not match. There is no error.
Real world result: contact may or may not have been made with that item of
evidence, but contact is not proved. The examiner and agency should be
aware that this conclusion may be used in an attempt to disprove contact,
but this conclusion alone does not disprove contact. Agency action: none
necessary. No recommendations necessary.
B2) Incorrect Acceptance of Disagreement to Exclude - The examiner
has incorrectly determined there was sufficient dissimilarity present in the
impression to exclude. The print could NOT and should NOT have been
excluded by that examiner based on his/her ability level, but it was. There
is technically no error because the prints did not actually match. A
possible scenario might be an unethical defense witness who wants to cast
doubt on the case, so he/she chooses an impression that is not actually of
value and proceeds to exclude the defendant as a possible donor of that
impression. A
casework scenario would be an examiner who kept a borderline impression that
was not suitable for comparison just because it was the only latent in the
case, but reported an exclusion.
It is noteworthy that
this scenario would have to also coincide with an error in Analysis Hs2,
Comparison Hs3, or Hs4. It is impossible based upon an examiners
ability that the non-matching impression be correctly considered suitable
for comparison against suitable known prints in a common area, yet the
examiner cannot correctly find sufficient disagreement to exclude. One
of the first 3 states would have to not exist: that the impression is
suitable for comparison, that the area is present in the known print, or
that the known print quality is suitable. In statistics,
incorrect acceptance of the null hypothesis is considered the more serious
"Type I" error of commission. The examiner incorrectly accepted the null
hypothesis that was actually false. Real world result: contact may or may
not have been made with that item of evidence, but contact is not proved.
This conclusion
may be incorrectly perceived as disproving contact,
but this conclusion alone does not prove or disprove contact. Agency
action: this
scenario would be difficult to prove, since no valid and reliable methods
are widely used to quantify and compare examiner ability and latent print
difficulty. Besides, who would say anything... it was, in reality, a
correct "guess" and did not involve a match. There is scientific and
statistical (type I) error. The examiner acted in an unacceptable (and in
fact an unethical) manner by retaining and effecting a conclusion on a print
that was not suitable. It
is recommended that
agencies be aware that
this problem can exist and take measures to address this scenario in
policies and procedures with defined methods of discovery through QA/QC with
appropriate corrective action. Disciplinary action may be appropriate
dependent upon the circumstances in the case and the history of the
employee.
B3) Correct
Rejection of Disagreement to Exclude - The examiner has correctly
determined that there was not a sufficient volume of dissimilarity to
exclude. The print could not and should not have been excluded by that
examiner based on his/her ability level, and it was not. There is no error
because the print was not suitable for comparison purposes. Technically the
print was not made by the same area of the exemplar being compared, but
there was not sufficient dissimilarity to exclude and therefore it was not
excluded. The print in question was deemed suitable for comparison purposes
at Analysis Hs2, was kept as suitable for comparison, and it was compared by
the examiner to the correct area of a known print of suitable quality. By
reaching this point the examiner has concluded that the print is actually
not able to be excluded from this finger, and should report it as "unable to
exclude". Some similarity may also co-exist between the latent print
and the exemplar for which the examiner cannot account. Real world
result: wasted time and contact not proved or disproved. Agency action:
none necessary as long as the impression is reported as unable to exclude.
Recommendation: use this scenario as an example that keeping and comparing
prints not suitable for exclusion wastes time and proves nothing.
B4) Incorrect Rejection of Disagreement to Exclude - The examiner has
incorrectly determined that there was not a sufficient volume of difference
to exclude. The print could and should have been excluded by that examiner
based on his/her ability level, but it was not. Possible scenario: the
examiner works for an agency that won't allow them to keep any impression
that falls short of containing a sufficient volume of detail for
individualization. In this scenario, the examiner may see a whorl pattern
in blood on the murder weapon, but does not exclude the only major suspect
with no whorls, because they have been instructed that it is not an
acceptable practice to keep impressions you cannot identify, even if you can
exclude them. In the field of latent prints, there has been recent heated
discussion on this concept. Scientifically, this would be considered a Type
II error of omission because the examiner failed to accept a null hypothesis
that was true. The examiner knew the print didn't match the subject, yet
he/she did not report the exclusion. Some would even argue that
reporting "no latents of value for identification were developed" instead of
"latent A was not made by person B" would fall into this scenario. The
real world result is that an exclusion could have happened that did not
happen, but touch to the item would have neither been proved nor disproved
either way. However, some examiners would argue that not reporting
this type of evidence is the same as withholding exculpatory evidence from
jury consideration. Although the exclusion would not disprove touch,
it may be one of many factors the jury would consider in a determination of
guilt or innocence. Agency action: it depends on how the agency views
their responsibility to report exculpatory evidence. The examiner did
not report a conclusion that was possible - actions may include anything
from counseling to revision of policies and procedures to corrective or even
disciplinary action.
There is often administrative controversy
surrounding cases where the examiner's report is unclear as to whether his
result was conclusive. This goes back to our discussion of verbiage
stating that "no identification was effected" representing both exclusions
and inconclusive results. If follow-up comparisons to complete
exemplars reveals a match, and it was assumed that the prior phrase was
exclusionary, the two reports would appear to many officers to conflict.
This is the reason why many agencies have moved to or are considering
exclusionary wording for latent prints that were compared to complete
friction ridge exemplars. Even if no palm prints are present, some
latent prints are obviously made by the end joint of the finger and can be
excluded as being the fingerprints of the subject.
In reality, the definition of exclusionary report verbiage and incorporation
of the practice into the identification unit ultimately holds the examiners
to a higher granularity of conclusions. This forces the examiner to go
beyond just finding the matches and instead further define whether
exhaustive comparisons excluded the impression from having been made by the
subject. With increasing scrutiny of the latent process both within
and outside of our community, going are the days that some examiners hide
behind the veil of confusion surrounding whether phrases like "no
identification was effected" are conclusive. On this topic, see also
The Detail # 78: Inconclusive Opinions,
The Detail # 80: Inconclusives.
In verification, all elements of ACE are objectively repeated by another
competent examiner. The same errors apply to the second examiner.
It is noteworthy that two examiners may arrive at correct conclusions and
differ in opinion. For example, if one examiner correctly determines
that a comparison is a match based on a high level of ability (A1), but
another examiner correctly determines that the comparison is inconclusive
(A3) this situation is not considered a major problem in the latent print
community. The agency or administrator may determine the root cause of
the issue and recommend additional training, for example. Likewise, if
one examiner correctly determines that a comparison is an exclusion based on
a high level of ability (B1) but another examiner correctly determines that
the comparison is inconclusive (B3) and therefore of no value for
comparison, this is not considered a major problem. Differing levels
of ability exist within our discipline. However, the situation cannot
exist where one examiner correctly determines agreement for
individualization (A1) and another examiner determines disagreement for
exclusion (B1). In this case, either the agreement being used by
examiner A is not truly agreement or the disagreement being considered by
examiner B is actually distortion. Given recent high profile erroneous
identifications, this is an area that should be further studied and
discussed within the community. What is our level of tolerance as a
community for an examiner that erroneously considers similar features to
agree to the level of individualization? What is our level of
tolerance for an examiner that erroneously considers distortion to disagree
to the level of exclusion? Is our tolerance of erroneous
individualization less than our tolerance for erroneous exclusion? Is
erroneous exclusion acceptable in our discipline?
As we conclude our discussion of errors associated with the latent print
process, it is appropriate to state that for every error, there is a reason
it occurred, a responsible person(s) and corrective action that should be
taken to avoid the error in the future. The goal of any administrator
dealing with an error is to determine the root cause of the issue and change
processes and procedures to prevent the root cause and therefore it's
consequence from occurring in the future. Some root causes may be
extremely distanced from their effect. Take for example the well known
six sigma example: The corrective action to stop disintegration of the
Washington Monument was to turn on the lights later into the evening.
How on earth did they establish that action? Follow their 5 why's:
Why was the monument disintegrating? Because of the use of harsh
chemicals.
Why were harsh chemicals being used? To clean up stains from pigeon...
um... yeah.
Why were there so many pigeons? Because there were a lot of spiders to
eat.
Why were there so many spiders? Because there were a lot of gnats to eat.
Why were there so many gnats? They were attracted there by the monument
lights at dusk.
For more on the 5 why's, visit
http://www.isixsigma.com/library/content/c020610a.asp.
Applying this principle to a conflict, we might ask 5 why's:
Why did the examiner incorrectly conclude exclusion? Because of one
dissimilarity.
Why did the examiner view this event as a dissimilarity? Because they did
not recognize it as distortion.
Why did the examiner not recognize distortion? Because they had never seen
that type of distortion before.
Why had they not seen that type of distortion? Because they had never been
to a course in distortion.
Why had they never been to a distortion course? Because a good one had not
existed prior to Alice Maceo.
We could continue, but you get the main point - without Alice where would we
be? just kidding. Seriously... there is a root cause to
everything. Once we establish the correct conclusion, our reaction to
conflict and error needs to always go beyond the immediate situation to the
root cause with the intent of constructive and corrective action. In
our discipline, the opportunity to deal with different types of conflict
arises from time to time. Usually, this deals only with inconclusive
results that are acceptable in the discipline. Once in a while ethical
issues are uncovered. And very rarely do we have differing conclusions
on opposite ends of the spectrum. But regardless of where the error
occurs, through a good understanding of milestones in the latent print
process, administrators will be in a better position to assess the urgency
of conflict as well as define and respond appropriately to error.
_____________________________________________________________________
Feel free to pass The Detail along to other
examiners. This is a free newsletter FOR latent print examiners, BY
latent print examiners.
There are no copyrights on The Detail, and the website is open for all
to visit.
If you have not yet signed up to receive
the Weekly Detail in YOUR e-mail inbox, go ahead and
join the list now so you don't miss out! (To join this free e-mail
newsletter, enter your name and e-mail address on the following page:
http://www.clpex.com/Subscribe.htm
You will be sent a Confirmation e-mail... just click on the link in that
e-mail, or paste it into an Internet Explorer address bar, and you are
signed up!) If you have problems receiving the Detail from a work
e-mail address, there have been past issues with department e-mail filters
considering the Detail as potential unsolicited e-mail. Try
subscribing from a home e-mail address or contact your IT department to
allow e-mails from Topica. Members may unsubscribe at any time.
If you have difficulties with the sign-up process or have been inadvertently
removed from the list, e-mail me personally at
kaseywertheim@aol.com and I will try
to work things out.
Until next Monday morning, don't work too hard or too little.
Have a GREAT week!
|
|
|
|
|