User:QDE-can/Sandbox

This is my personal sandbox - it contains only tentative drafts of information that may or may not ever be proposed or added to wikipedia. RB Ostrum (talk) 16:48, 28 December 2009 (UTC)

---

Admissibility 'standards' and FDE

Most court systems have guidelines for the [Scientific evidence (law) | admissibility of scientific evidence] which applies to all forms of this type of evidence. These all apply to QDE evidence and, while they vary in some specifics, there are common themes and concerns.

Various Court Rulings

• Daubert v. Merrell Dow Pharmaceuticals (1993) USA: Court defined "scientific methodology" as the process of formulating hypotheses and then conducting experiments to prove or falsify the hypothesis, and provided a nondispositive, nonexclusive, "flexible" test for establishing its "validity":

1. Empirical testing: the theory or technique must be falsifiable, refutable, and testable.
2. Subjected to peer review and publication.
3. Known or potential error rate.
4. The existence and maintenance of standards and controls concerning its operation.
5. Degree to which the theory and technique is generally accepted by a relevant scientific community.

• R. v. Mohan (1994) Canada: Expert evidence, stated Justice Sopinka, should be admitted based on four criteria:

1. relevant,
2. necessary to assist the trier of fact,
3. should not trigger any exclusionary rules, and
4. must be given by a properly qualified expert.

Other Rules and Guidelines

• Federal Rules of Evidence (USA), in particular Rule 702 which states "If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise, if (1) the testimony is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case."

FDE activities relating to Admissibility

It should be noted that, for the most part, the guidelines listed above were and are not intended to be an 'inclusive checklist'. That is, it is the responsibility of the trial judge to assess the situation using these factors as examples of things to be given consideration in their assessment. Furthermore, no clear guidelines have been given as to what would fulfill these 'requirements' in any given case.

• Studies relating to
• Studies relating to
• Studies relating to
• Studies relating to

Scope of document examination

The section references ASTM so needs to be updated.

Presently it reads:

The American Society for Testing and Materials, International (ASTM) publishes standards for many methods and procedures used by FDEs. E30.02 was the ASTM subcomittee for Questioned Documents, and ASTM Standard E444-09 (Standard Guide for Scope of Work of Forensic Document Examiners) indicates there are four components to the work of a forensic document examiner. ...

Suggested wording:

The Scientific Working Group for Questioned Document Examination has published standards for many methods and procedures used by FDEs. Copies of SWGDOC standards can be obtained from the SWGDOC website.

The SWGDOC 'Standard for Scope of Work of Forensic Document Examiners' indicates there are four components to the work of a forensic document examiner.

Conclusions

The page does not presently mention or discuss conclusions that may be offered by QDEs for various examination types.

The SWGDOC standard should be presented. Clearly, the alternative using a logical approach (see below) should be presented since many labs (particularly European) are using this and it will form the basis of the 'new' SWGDOC standard.

Could structure the section to first present the general conclusion scale with an explanation of its interpretation. This could be further clarified for different types of examinations where the scale does not work as well (eg. ink differentiation).

Identification / Individuality

The present discussion is adequate as it reflects the traditional approach to FDE work (and other forensic disciplines). But it doesn't delve into the challenges being made to this approach — nor the logical issues with it.

This ties in with the 'Bayesian Approach' below.

Logical (Bayesian) Approach to Evidence Evaluation

There is a clear need to explore 'alternatives' to the traditional approach to both examinations and the resulting expression of conclusions.

The opinion, or conclusion, of an ‘expert’ witness is the outcome of the evaluation process done by that person. The most common conclusion wording speaks directly about a proposition (ie., a possible explanation or scenario) in probabilistic or definite terms (the latter being an end-point where probability equals 1 or 0). For example, SWGDOC has terminology definitions for up to 9 'levels' of conclusions relating to authorship.

Examples include the following (with the proposition element of each statement underlined):

• It is my opinion [or conclusion] that the writer of the known material wrote the questioned material.
• It is my opinion (or conclusion or determination) that the writer of the known material probably wrote the questioned material.
• There is evidence which indicates (or suggests) that the writer of the known material may have written the questioned material but the evidence falls far short of that necessary to support a definite conclusion.
• Etc.

While this type of wording for conclusions is very common in all forensic disciplines, including FDE work, it is not without critics who believe the expert is not in a good position to make such assertions about the propositions based solely on the evidence.

The most obvious such alternative to this approach is a proper probabilistic approach for evaluation of evidence that focuses on the evidence, and does not extend further to any statement about the propositions under consideration.

For example, the proposed page will explain, contrast, and compare:

1. the 'classical' approach (non-Bayesian with conclusions expressed as posterior odds — this is the approach used in most disciplines in North America - and introduced above),
2. a 'Full Bayesian' approach (Bayesian processes with conclusions expressed as posterior odds - requiring some 'assessment' of priors for the propositions and/or a decision heuristic for the conclusion threshold), and
3. the logical (‘LR’, ‘BF’ or other) approach (results expressed as a “likelihood ratio” or equivalent concept).

Obviously, discussion will have to distinguish between these approaches and explain the pros/cons of each.

Also worth emphasizing that many proponents (including myself) differentiate strongly between the "Bayesian" approach (meaning the full application including priors and posteriors) and the more limited/constrained "logical" approach (meaning expression only of the LR aspect) with the argument being that the latter isn't "Bayesian", per se. This comes from the idea that the concept of the LR, or at least likelihood, is used in many ways that don't involve Bayes Theorem at all. In addition, it makes no demands on anyone to use or understand the theorem in full in order to make use of the LR information provided by the expert.

Comparison process - Examination

I would also like to expand the area the Examination section to more fully explain how a comparison leads to a conclusion. This ties into the previous 2 topics but it is separate.

Could distinguish the traditional model that classifies features as being 'class' versus 'accidental/personal' to a more modern (?) approach that focuses more upon 1) complexity of the graphic movement and 2) rarity of forms (movement combinations). Pros and cons as well as commonalities between these two could be explored.

In addition, the concepts of evaluation and weighing of evidence ties back (again) to the topic of Bayes/LR.

Research topics

Outline topics being researched -- IGS, Unil, NIJ-sponsored, etc.

Examiner competency

This should include issues relating examiner competency such as methods of assessment, limitations in testing, test results and their interpretation, and controversies.

This should also include a more extensive and frank discussion of testing that has been done to date (Kam, Found, others) as well as a review of the criticisms (both valid and invalid). This would likely lead to discussion of conclusion scales and alternatives used in different jurisdictions (see info re. Bayesian philosophy/evaluation). It could also relate to various court rulings (if only to highlight the manner in which courts have viewed this information).

This also leads, in my mind, to a more in-depth discussion of 'error rate' in forensic science examination procedures. After all, any professed skill should be testable in some manner and, if tested, the results could be used (carefully) as an estimate of potential 'error rate' associated with the tested skill. This is, admittedly, a very simplified view of a very complex problem. Skill testing is difficult even for simple tasks and FDE examinations are anything but simple tasks. It is very difficult to identify and control potential sources of variation and, therefore, difficult to ensure that any given task will produce a truly meaningful estimate of the theoretical error rate inherent in a given process. There is also the issue of a 'method' error rate versus that of a given practitioner; again, not a trivial matter to address.

"Tools of the trade"

This is an existing topic in clear need of a re-work. At present it reads :

Common tools of the trade

• Excellent Eyesight
• Handlens/Loupe
• Stereomicroscope
• Electrostatic Detection Apparatus (ESDA)
• Video Spectral Comparator (VSC)

IMO, this is a ridiculously trivial list that doesn't begin to reflect all of the methodologies and equipment used in QD examination. Proposed (and rejected) in Nov 2008:

Tools of the trade

For Handwriting Authorship Examinations:

• Excellent eyesight/ visual perception

• May be ‘corrected’ vision but requires 20:20 visual acuity as well as good colour discrimination.
• An absence of significant vision defects such as astigmatism or colour blindness.
• An absence of related cognitive conditions such as form blindness

• Handlens or Loupe

• Basic hand magnifier 5x to 8x commonly used

• Stereomicroscope for Optical Microscopy

• Binocular microscope with magnification in the common range of 0.8x to 150x

• Various lighting sources

• Oblique light, variable angle
• Ring light
• Coaxial light

For Non-handwriting Examinations:

• Spectral Imaging: used to examine documents in ultraviolet, visible and/or near-infrared regions of the ElectroMagnetic spectrum allowing documents (or elements such as inks, papers, coatings, etc) to be differentiated. These methods are also useful for decipherment of obliterated entries when different inks are used to produce the document. Some common devices include:

• Foster & Freeman Co. UK: Video Spectral Comparator (VSC)
• Projectina CH: Docucenter or Dragon

• Hyperspectral imaging systems: used to examine spectral response of inks, papers, and other materials allowing these to be differentiated
• Chromatography: various methods used to examine chemical composition of inks, papers, and other materials, particularly dyes used in colour components

• Thin-layer Chromatography (TLC)
• High-performance Liquid Chromatography

• Spectroscopy: used to characterise documents or their components (there are many different methods used for inks, papers, coatings, etc)

• Gas chromatography-mass spectrometry (GC-MS)
• Inductively coupled plasma mass spectrometry (ICP-MS)
• RAMAN Spectroscopy
• Atomic Force Microscopy
• FT Spectroscopy
• X-ray Flourescence (XRF) Microscopy

• Optical Interferometry: used to examine and/or measure 3-dimensional characteristics on documents and/or printing devices
• Print Quality Metrics systems: specialized machine-vision systems used to measure various aspects of printer/fax/copier output to evaluate make-and-model or to compare a suspect device with questioned hardcopy output
• Electrostatic Detection/decipherment Device (EDD): used to detect and/or decipher latent indentations in paper. Common devices include:

• Foster & Freeman Co. UK: Electro Static Detection Apparatus(ESDA)
• Projectina CH: Docustat DS-210
• Kinderprint Indentation Materializer Electrostatic Document Device (IMEDD)
• Other indentation detection/decipherment methods

• Oblique lighting apparatus
• Gelatine lifting (Koeijer et al)
• Microsil lifting

• Typewriter Ribbon Reader: used to generate a legible transcript of typewritten entries on a carbon-film typewriter ribbon

• Ribbon Analysis Workstation (Envisage RAWII)

In Development:

• Cedar-FOX software

Rework of EDD page

Some work now done. Consider the following:

Examples of use

• Suicide notes
• Police notebooks
• Medical records
• Other???

Limitations (better discussion)

• Secondary indentations
• Over-humidification
• Difficult paper types
• Objects (books, magazines)

Other uses

• Footmark visualization

History

General

This could branch to some other wiki containing Tim's work... or cross-link and present the key elements in WP.

Canada

Need dates: inception/creation, cessation, expansion/reduction

Public lab systems: RCMP, CSIS, CRA/CBSA, CFS, Laboratoire judiciaire, other

Societies: CSFS Document Section

Major figures: John Lomax (Montreal), Herbert J. Walter (Winnipeg), Roy Huber (Ottawa)

WP Pages to develop

• Southwestern Association of Forensic Document Examiners (SWAFDE)
• Southeastern Association of Forensic Document Examiners (SAFDE)
• European Network of Forensic Science Institutes (ENFSI) - page created by TimothyPilgrim
  • European Network of Forensic Handwriting Experts (ENFHEX)
  • European Document Experts Working Group (EDEWG)
• Scientific Working Group for Forensic Document Examination (SWGDOC)
• Gesellschaft für Forensische Schriftuntersuchung (GFS) e. V. (German and English versions)
• The Australian and New Zealand Forensic Science Society (ANZFSS)
• Forensic Science Society (FSS)
• La Trobe University Forensic Expertise Profiling Laboratory (FEPL)
• Institut für Schrift- und Urkundenuntersuchung (ISU)

• Sequence of strokes - page created; still needs work

• Lead: Definition in context
• General method: Reference ASTM? (only when standard published - still draft)
• Possible applications with visual/image examples:

• Sequence of writing strokes
• Sequence of ballpoint inks with other materials (bbp ink, toner, etc)
• Sequence of creasing with toner
• Sequence of indentation creation

• Validation of methods

• Physical matching

• Lead: Definition in context (reconstruction or evaluation of common origin)
• General method: Reference ASTM?
• Possible applications with examples:

• Torn paper
• Cut paper
• Glass particles
• Duct tape
• Wood pieces
• Other

• Price markers

Add to Evidence under Bayes' theorem page with discussion of R v T ruling as well as general discussion of why a likelihoodist approach is 'inevitable'.

Done

• Journal of the Canadian Society of Forensic Science
• Journal of the American Society of Questioned Document Examiners
• European Network of Forensic Science Institutes page created

References

This is here only to avoid cite errors on this page: