CSFS Document Section Position on the Logical Approach to Evidence Evaluation and Corresponding Wording of Conclusions

Abstract

This article presents the position of the Document Section of the Canadian Society of Forensic Science (CSFS) regarding the use of an evaluation and reporting scheme often referred to as “the logical approach to evidence evaluation.” The section’s position is the logical approach to evidence evaluation and reporting, and is an appropriate and effective option for forensic document examination (FDE) work when implemented as outlined in this paper.

Keywords:

• evaluation
• conclusions
• likelihood-ratio
• forensic document examination

View correction statement:

Correction

View translated version:

La position de la Section des documents de la SCSJ sur l'approche logique de l'évaluation de la preuve et le libellé des conclusions

Acknowledgement

This position paper was reviewed by CSFS QD section members prior to finalization. The final position represents a consensus opinion by section members.

Notes

1 The term “logical approach” derives from the approach outlined by Robertson and Vignaux[86]. It is used in preference to the “Bayesian approach” or “likelihood-ratio approach” both of which also appear in the literature. The end product of the logical approach is an expression, either quantitative or qualitative in nature, conforming to the likelihood-ratio in both form and intent. However, the logical approach can be applied in the absence of quantified data and does not require or invoke the use of Bayes Theorem. Rather it focuses solely on the likelihood-ratio component or some expression equivalent to it in meaning and intent.

2 In general, “The mission of the C.S.F.S. is to promote the study, raise the standards and enhance the stature of forensic science as a distinct discipline…” and, in particular, “to see, receive, marshall, assess, record, report and disseminate information emanating from forensic and associated studies…” Available online at http://www.csfs.ca/about/mission-statement-csfs/

3 The literature makes reference to both “findings” and “evidence”. These are similar terms but it has been observed that “Findings are the result of observations, measurements and classification that are made on items of interest. They can be qualitative (nominal or ordinal) or quantitative (discrete or continuous). No result is also a finding” (ENFSI Guide, 2015). In essence, findings refer to any outcome from analyses and comparisons conducted on materials submitted for forensic examination. For the most part, the present document will preferentially use “findings” rather than “evidence”.

4 See https://www.nist.gov/forensics/organization-scientific-area-committees-forensic-science for more information.

5 It should be noted that each of these bodies discusses the topic using different terms; however, the essential message is the same in each of them.

6 DocSAG is a Special Advisory Group established under the auspices of the Senior Managers Australia and New Zealand Forensic Laboratories (SMANZFL). It is managed by and reports to the Australia and New Zealand Policing Advisory Agency National Institute of Forensic Science (ANZPAA NIFS) Forum.

7 The use of the descriptors “prosecution” and “defence” does not mean this approach applies only to matters of a criminal nature. The approach applies to all types of evidence evaluation regardless of which judicial system is considered. In the civil realm more appropriate terms might be “plaintiff”, “respondent”, “opposing party”, etc.

8 With respect to the basis for probability values ENFSI Guidance Note 3 states: “Relevant and appropriate published data will be used wherever possible.” However, the Guide also notes that “Forensic practitioners often experience difficulty in assigning and justifying probabilities when the assignments are based on expert knowledge.” In those situations “… likelihood ratios can be informed by subjective probabilities using expert knowledge.” This is similar to the manner in which examiners presently assess probabilistic elements of their work.

9 Examples of “direct probability” include “the findings are much more probable if the writer of the specimen material wrote the questioned signatures, than if someone else wrote the signatures” or “it is slightly more probable to observe this degree of correspondence between the specimen and questioned signatures if someone other than the writer of the specimen material wrote the questioned signatures than if the writer of the specimen material wrote them.”

10 It must be noted that issues of terminology, including specific aspects of the conclusion scale, are the subject of ongoing research and debate. The scale presently recommended for conclusions is that shown in Table 1, subject to change based on future research into the optimal number of levels as well as appropriate descriptive terms for each category.

11 The most common acronym used to describe the “traditional” process involved in comparative pattern evidence is ACE-V, referring to Analysis, Comparison, Evaluation and Verification, following the procedures outlined by RA Huber in Expert Witness (The Criminal Law Quarterly 1959–1960, 1959;2:276–295). Huber used ACE to describe the philosophy of science and the correct use of the scientific method in any comparative forensic science.