Handwriting as Evidence Research Paper - Research Paper Examples

This sample Handwriting as Evidence Research Paper is published for educational and informational purposes only. If you need help writing your assignment, please use our research paper writing service and buy a paper on any topic at affordable price. Also check our tips on how to write a research paper, see the lists of criminal justice research paper topics, and browse research paper examples.

Handwriting is produced by finely coordinated and precisely articulated movements of the hand, finger, and arm, manipulating a tool which leaves a trace on a surface with the ultimate control being responsible – the brain. It is a very complex activity requiring intelligence that distinguishes human being from the rest of the animal species. By way of writing, the author can convey and preserve ideas or information. With the advancement of technology, the writing instrument changes considerably, but handwriting, being the most personal and immediate means of graphic communication, and the movements associated with it remain virtually the same since the creation of the strokes and the alphabets.

Ever since the invention of the “written language,” the need for the verification of authorship started to exist. Nowadays, there is a widespread demand for the examination of handwriting, not only in criminal cases and civil litigations but also in historical and archaeological investigations, and handwriting examination is an important specialist subject in forensic sciences. Basic principles in handwriting identification are given in a number of books authored by authorities in the field (Harrison 1966; Osborn 1978; Hilton 1982; Ellen 1993), while forensic scientists continue to conduct research in the scientific examination of handwriting.

Writer Identification

Physiologically the ability to write is a “motorperception” skill, involving psychological aspects, perception, and muscular coordination on the part of the author who is constantly under the influence of a variety of factors. These comprise both internal (or personal) and external (or circumstantial) influences. Because of the large number of variables affecting the writer, handwriting from one person on different occasions cannot be expected to be identical in each and every aspect of line quality, speed, pen pressure, shading, pen emphasis, slant, arrangement of strokes, proportionality, chronological sequence, and connecting components. This phenomenon of a “nonuniformity” in the handwriting of an individual is commonly referred to as natural variation.

That handwriting varies does not present serious problems to the task of identification. The skill to write is developed and practiced over many years and is largely a subconscious and automatic activity of human beings. Thus, the handwriting of a person, despite not being 100 % precise, conforms to a distinctive and predictable pattern. The verification of handwriting, as to whether or not it was written by a certain individual, is based upon the fact that handwriting embodies various qualities and attributes which in combination are sufficiently unique to be individualized. The joint probability of a compound event comprising a number of mutually independent simple events is the product of the probabilities of the simple events. Hence, if the disputed handwriting contains a variety of personal characteristics that lie within the range of natural variation of the known writing and at the same time not even a single fundamental, or unexplainable, difference from the genuine specimens can be found, then the combined probability of occurrence of this set of characteristics present in the handwriting of another person selected at random from a population will be so small that it can be ignored. The argument is that it would not be reasonable to assume that the same combination of writing attributes, in sufficient numbers and weight, would accidentally occur in the handwriting of another person without showing even a single idiosyncrasy from the control specimens. This hypothesis forms the basis of identification.

Handwriting characteristics are the observable features generated by the movement of the author’s hand. The act of writing involves not only two-dimensional movements of the pen but also variations in the pressure of the writing tip on the substrate. Minutiae of the landing movement when the writing tip approaches and finally “touches down” on the surface of the paper generate a large variety of shapes at the start of a stroke. Likewise the rehearsed motion at the conclusion of a stroke provides specific calligraphic attributes. Details of the habitual manners of utilizing the pen to write contribute toward much of the visual appearance which can be generalized by the terminology of line quality. The spatial relationship of strokes, the formation of the basic components of words or characters, the relative dimensions of two or more alphabets, and the ways of connecting adjacent strokes, in aggregate, constitute the writer’s calligraphic profile. The determination of authorship therefore depends on the discovery, in the disputed script, of personal characteristic features that fall within the range of natural variation of writing habits of the author of the control material.

In practice, the process of identification must include a determination of the extent, kind, and significance of this resemblance as well as of the variation. It is necessary to determine whether the divergence is due to the operation of a different individual or is only the expected and inevitable variation found in the genuine writing of the same writer. It is also necessary to decide whether the resemblance is the result of a skilful imitation or is the habitual and characteristic resemblance which naturally appears in a genuine writing. When these two questions are correctly answered, the whole problem of identification is solved (Osborn 1978). Figure 1 gives a diagrammatic presentation on the process of handwriting examination.

Handwriting as Evidence Research Paper

For many years, document examiners have been practicing the empirical theory of handwriting examination. The US Supreme Court case, Daubert et al. v. Merrell Dow Pharmaceuticals, alongside with several other cases, namely, US District Court ruling, United States v. Starzecpyzel; US Supreme Court ruling, General Electric Co. v. Joiner; US Supreme Court ruling, Kumho Tire Co. v. Carmichael; and US 11th Circuit Court of Appeals, United States vs. Paul, have prompted the so-called Daubert Standard which refers to the following guidelines for admitting scientific expert testimony: (1) Judge is gatekeeper: The task of “gatekeeping,” or assuring that scientific expert testimony truly proceeds from “scientific knowledge,” rests on the trial judge. (2) Relevance and reliability: The trial judge is required to ensure that the expert’s testimony is relevant to the task at hand and that it rests on a reliable foundation. The judge must find it more likely than not that the expert’s methods are reliable and reliably applied to the facts at hand. (3) Scientific knowledge = scientific method/methodology: A conclusion will qualify as scientific knowledge if the proponent can demonstrate that it is the product of sound “scientific methodology” derived from the scientific method. (iv) Factors relevant: The Court defined “scientific methodology” as the process of formulating hypotheses and then conducting experiments to prove or falsify the hypothesis and provide a test for establishing its “validity”:

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

In 2000, the above rule was amended in an attempt to include the additional provisions which state that a witness may only testify if:

The testimony is based upon sufficient facts or data.
The testimony is the product of reliable principles and methods.
The witness has applied the principles and methods reliably to the facts of the case. Although the Daubert Standard is concerned with legal requirements for scientific testimony in the United States, it is nevertheless important to assess the mandate of handwriting examination as a specialist forensic science subject.

Scientific Method And Handwriting Examination

Science is a branch of knowledge conducted on objective principles involving the systematized observation of, and experiment with, phenomena especially concerned with the material and function of the physical universe (Allen 1990). It stems from the curiosity of an observant mind, which generates impetus to the quest for an explanation. Observation is the primary scientific procedure of comprehending things; based on observation, scientists attempt to provide logical deductions that can plausibly explain the incident. A hypothesis is thus formulated. The proposed explanation is tested by way of further observations and experimentation. The latter simulates the observed events under controlled conditions and the designer of the experiment attempts to draw logical inferences from the results thus obtained. Similar experiments, often using a variety of approaches and methodology, are performed by different workers to verify the results reproducibility. When a hypothesis cannot live up with the standard or is considered unsatisfactory, it is either modified or discarded. On the other hand, a hypothesis passing stringent scientific tests will be adopted into the framework of a theory which not only explains the observations in a reproducible way but also makes meaningful predictions.

Application of the scientific method often involves measurement which facilitates a more accurate description of the observed phenomenon in numerical terms and quantifiable units. Phenomena presented in the form of numerical data can be compared and correlated, and their numerical representations treated by mathematical and/or statistical procedures to generate more information which objectively explains the observations in a more detailed and profound manner. Handwriting embodies many parameters that can be measured, and in the Latinized writing – slant or slope – the angle of inclination of the axis of letters relative to the baseline (Hilton 1982) is an important measurable parameter. It varies between writers from backward slanting to a sharply inclined forehand slope. The wide range of variation in slant gives rise to good discriminating power among individual writers.

Figure 2 illustrates the variability of slant in the handwriting of four individuals: Writer A exhibits a habit of occasionally writing alphabets with a backward slant, for example, the l of Sherlock, the p of Vampire, the first d of deduction, and the g of original. But many of the other letters exhibit a near-vertical or slightly forward slant. On the other hand most, if not all, of the characters in sample B are forward slanting, for example, the d and the t, although a few of the letters are more vertically oriented, for example, the p of the two points and independent, and the y of the by’s. Author C produces consistently forward-slanting alphabets in many of the d’s and t’s. While displaying natural variation, specimen D appears to be more consistently forward slanting. The diversity of slant in the handwriting of different individuals can therefore be utilized for the identification of authorship. Additionally, the slant of different alphabets in a sample of handwriting can vary considerably in magnitude, for example, as evidently shown in the writing of C, the j of subjective and objective and the y of say are less forward slanting than the other letters. It is generally observed that (1) while the slant of an alphabet in the handwriting of an individual generally conform with one another, different alphabets of the writings of the same individual can vary to a greater extent; (2) in some writings, the position of an alphabet in different words can display different slanting angles, for example, the initial t of the word that in sample D is much more slanting than the terminal t of the same word; and (3) the slant of an alphabet can also vary in accordance with the letters adjacent to, or flanking, it. Leung (Leung et al. 1987) noted the same phenomenon in Chinese handwriting and found that measurable parameters of the handwriting of a writer do not conform to a uniform value but may vary in accordance with the structure of, and the position of the strokes in, the character. The concept of “internal variability” was suggested which is attributed to the divergence of writing habits in relation to the position of the writing element and the overall structure of the written words or characters. It was maintained that this calligraphic property can be utilized as an additional tool for handwriting identification (Leung et al. 1987; Leung 1994). To conclude, the numerical expression of depicted features enables the document examiner to characterize handwriting objectively and accurately and is especially useful when the similarities and/or differences are inconspicuous.

Handwriting as Evidence Research Paper

Another important facet of the scientific method is classification which simplifies the approach to a given problem by narrowing its broader elements and ramifications on the basis of categorical distinctions. Classification methods can be effectively applied to find out the relationship, if any, of certain observations so that general principles can be derived. It forms the basis of natural sciences such as biology, chemistry, and taxonomy. Apart from writing attributes which can be characterized quantitatively, there are other qualitative features that are equally useful for identifying the writer. Eldridge reported a scheme for the classification of a selected set of cursively handwritten letters comprising d, f, h, k, p, and t (Eldridge et al. 1984). These letters were treated in terms of three major structural components: the staff, the base, and the crossbar. Twelve features were chosen for classification, although not all features are applicable to all letters. They are as follows: (1) construction order, the order of construction of the base and the staff; (2) number of strokes; (3) base direction, the direction of the curvature from the uppermost part of the base to the beginning of the upstroke to the next letter; (4) base closure; (5) top of staff closure; (6) top of staff direction; (7) bottom of staff closure; (8) bottom of staff direction; (9) preceding letter join; (10) following letter join; (11) crossbar position; and (12) crossbar curvature. Each feature has at least two, and possibly as many as seven, distinct forms. It was found that the letter k best discriminates between writers in the sample of 61 subjects.

A systematic classification scheme not only provides valuable information on the variability of handwriting but also inspires objective interpretation on examination results. Leung conducted statistical studies into 98 selected Chinese characters written by 438 subjects (Leung et al. 1993). A classification system on the qualitative features of pen emphasis, sequence of strokes, stroke connections, and structural detail was established. Diversities of writing habits depicted among the subjects confirmed the belief that many of the classifiable handwriting attributes contribute toward the science of authorship verification. Figure 3 illustrates four chronological writing sequences in the Chinese character Live which comprises one slanting stroke, one vertical stroke, and three horizontal strokes. The possibility of applying the concept of classification by means of stroke sequencing on the capital letter “E” is demonstrated.

Handwriting as Evidence Research Paper

Individuality Of Handwriting

The principles of handwriting examination have been laid down in classical treatise that the identification of authorship depends on the discovery in the questioned script of a combination of personal characteristics which lie within the range of variation of the control specimens accompanied by the absence between them of any idiosyncrasies that cannot be satisfactorily explained. However, as pointed out by Sugiyama and Kurauchi, the problem of the methods in practice relies on the basic task of determining and extracting a sample’s distinguishing features, which have often been questioned for dependence of subjective judgment and intuition rather than on objectivity and empirical evidence (Sugiyama and Kurauchi 1986). The rapid advances in computer technology enable the practical application of statistical classification techniques to handwriting examination. Many forensic scientists reported positive results; to name just a few, Hardcastle, Thornton, and Totty reported a computer-based system for the classification of block handwriting on checks (Hardcastle et al. 1986), and Eisermann and Hecker described the Forensic Information System for Handwriting (FISH), which utilized an image-processing technique with statistical classification analysis for the identification of handwriting (Eisermann and Hecker 1986).

Motivated by several rulings in US courts concerning expert testimony in general, and handwriting testimony in particular, Srihari undertook a study to objectively validate the hypothesis that handwriting is individual (Srihari et al. 2002). They maintained that there are two variances of concern in handwriting comparison, within the handwriting of the same individual and between the handwritings of two individuals, and that the within-writer variance is less than the between-writer variance. The goal of their study was to objectively verify this common observation. A source document in English of 156 words capturing all characters (letters and numerals) and certain character combinations of interest was designed for each participant to copy three times. Handwriting samples of 1,500 individuals, representative of the US population with respect to gender, age, ethnic groups, etc., were obtained. Analyzing differences in handwriting was done by using computer algorithms for extracting features from scanned images of handwriting specimens. Two types of features characteristic of the handwriting were obtained: conventional and computational. Conventional features are the handwriting attributes that are commonly used by forensic document examiners which comprise 21 discriminating elements, namely, (1) arrangement, (2) class of allograph, (3) connections, (4) design of alphabets and their construction, (5) dimensions (vertical and horizontal), (6) slant or slope, (7) spacings (intraword and interword), (8) abbreviations, (9) baseline alignment, (10) initial and terminal strokes, (11) punctuation, (12) embellishments, (13) legibility or writing quality, (14) line continuity, (15) line quality, (16) pen control, (17) writing movement, (18) natural variations or consistency, (19) persistency, (20) lateral expansion, and (21) word proportions (Huber and Headrick 1999). Macro- and micro-computational features are those that can be determined algorithmically by software operating on scanned image of handwriting. The following set of 11 macro-computational features were used: (1) entropy of gray values, (2) gray-level threshold, (3) number of black pixels, (4) number of interior contours, (5) number of exterior curves, (6) number of vertical slope components, (7) number of horizontal slope components, (8) number of negative slope components, (9) number of positive slope components, (10) slant, and (11) height, while micro-computational features are computed at the allograph, character, or shape level.

Statistical analysis was conducted using the identification model and the verification model. In the identification model, given a questioned handwriting sample x and samples of handwriting of n known writers, the classification task was to pick out the writer of x among n writers. In verification model, given two unknown handwriting samples x1 and x2 and samples of handwriting of n writers, the objective was to determine whether x1 and x2 were written by the same person or by two different people among the n writers. It was found that writer identification accuracy was close to 98 % for two writers while the verification accuracy was about 96 %. Using global attributes of handwriting and very few characters in the writing, the ability to determine the writer with a high degree of confidence was thus established.

Another study was made to show that the handwriting of an individual is so unique that by applying discriminant analytical methods on the parameters of width-height ratio, symmetry factor, slant, and tilt, the writer could be objectively indentified (Cheung and Leung 1989). In the above experiment, 25 Chinese characters were selected. A total of 376 handwriting specimens were collected from 21 subjects and after 3 months, 6 of the subjects were randomly selected to write once again the 25 characters which were then used as the “questioned samples”; each of the handwriting specimens was characterized by one or more of the 4 measurable parameters. Applying linear discriminant analysis, although the six “questioned” handwritings could not be correctly associated with their corresponding authors by using only one of the parameters, a perfect recognition rate was recorded using all four parameters. The findings are consistent with the hypothesis that verification of authorship is based upon the discovery of a number of divergent characteristic features which in combination become sufficiently personal for an identification to be made.

The above-mentioned two pieces of research have clearly demonstrated the uniqueness of handwriting, and it can be inferred that consideration of both measurable and qualitative parameters in handwriting gives rise to a powerful theoretical foundation for the scientific identification of authorship. Using this theoretical framework, identification of handwritings can be made, with objectivity occupying the center position.

The Art And Science Of Handwriting Examination: An Epilogue

Successes in computational classification experiments on handwritings provide scientific support for the long-established (but not vigorously tested) practices of handwriting examination. However, these experiments were conducted on natural handwritings but were not directed to forgery, or any other examinations involving such complications as disguise, handedness, and influence of health and drugs. It is envisaged that for a long time, handwriting identification would have to be made by specialist document examiners instead of relying on computational software for verifying authorship. As such, the entailment of a certain degree of subjectivity would be inevitable in handwriting examination. The objective component of handwriting examination provides the most easily acquired skills of a document examiner, while the subjective reasoning processes require years of training to fully develop (Whiting 1986).

The initial stage of handwriting analysis is basically a comparison of morphological features between the questioned writing and the control exemplars. It depends on the ability, on the part of the document examiner, to find out the more subtle, personal writing attributes in the script. This exercise is by and large objective. Some degree of subjectivity is involved because the procedure relies on the expert’s cognitive selection of handwriting features which he or she regards as important personal characteristics. While it is the author’s personal experience that well-trained document examiners generally pick out similar assortment of characteristics from the same handwriting specimens, there are deviations, both in terms of numbers and the ways of how the depicted characteristics are described and comprehended.

The second stage of authorship determination is the interpretation of the depicted writing characteristics. The document examiner has to evaluate the relative significance of the writing attributes which he or she has discovered and consider the information related to the case, for example, the physical or mental condition of the alleged writer, the presence or absence of any external influences, and the possibility of disguise. This is a mental exercise which encompasses subjectivity to a larger extent. Handwriting experts can diverge considerably, especially for difficult cases, in their assessment of the relative importance of the depicted handwriting features. It is during this stage of the examination process that divergence of opinion among handwriting experts sometimes arises.

An element of subjectivity also relates to the expression of an opinion in that after assessing the evidence and forming an opinion, one has to transform the conceptual inclination into written words. Any findings or observations will be useless unless they are comprehended and their significance appraised. Based on the assessment of the examination results, the expert will have to come up with an opinion on the authorship. Inaccuracy and variability can occur. While a wholly subjective opinion will be the personal view on the subject matter without any evidential support which is undoubtedly inacceptable, an entirely objective opinion would be restricted to facts and figures only. In handwriting examination, a 100 % objective “opinion” will not meet the expectations of the parties concerned because the so-called opinion will only be a list of the similarities or differences that the expert has discovered without any interpretation as to what these findings indicate.

Opinion expressions used by handwriting experts vary among different schools of document examiners, both in terms of opinion expressions and meaning (Leung and Cheung 1989). Some of the more popular opinion expressions being adopted are as follows: wrote or written by, highly probable or very high probability, probable or high probability, very likely, likely, could well have, consistent with, could have, inconclusive, no evidence, consistent with (not being written by), unlikely, highly unlikely, probably not, could not have, and did not write. While not all of the above-mentioned expressions are used by a single document examiner, the choice of terminology depends on the reasoned preference of the individual expert. The levels of opinion also vary among different practitioners. A few document examiners, which can be described as the “all-or-none” school, believe that only three classes of opinion, namely, identification, elimination, and inconclusive, should be given (McNally 1979). However, the majority of document examiners who maintain that since facts of handwriting examination will vary in number and significance, the language selected to summarize different sets of facts must vary, some being in definite terms and some in qualified terms (Cole 1980). Nowadays, it is widely accepted that in order to reflect the degree of certainty or belief, a scale consisting of different levels of expressions should be used. Ellen described a scale of five ranges of opinion and proposed that a small number of divisions of a scale are advantageous in that the significance of each point on the scale can be easily defined (Ellen 1979), but an 11-point opinion scale was also suggested (Leung 1994). In 1991, the Questioned Documents Section of the American Academy of Forensic Sciences and the American Board of Forensic Document Examiners have adopted recommended guidelines in report and testimony which consist of a 9-point opinion scale (identification, strong probability, probable, indications, no conclusion, indications did not, probably did not, strong probability did not, elimination) based on the concept that probability, as used in handwriting opinions, is not a statistical measurement but a measurement of the examiner’s confidence based on scientific principles and experienced judgment (McAlexander et al. 1991).

The divergence of the number of opinion scales and terminology used by handwriting experts clearly show that the process of expressing opinions does involve a certain degree of subjectivity on the part of the expert although the expressed opinion is based on scientific observations and experienced judgment. After examining the handwritings or signatures, the expert has to assign the opinion in ordinary language terms that he or she believes to be the most appropriate, and subjectivity comes into play during the transformation from abstract thinking – a degree of certainty or confidence in one’s mind – into written words which is susceptible to vagueness and controversy when the readers of the report attempt to convert in their minds these same written words back to the abstract certainty of belief. The inherent difficulty in handwriting and signature examination is due to the fact that contrary to DNA profiling, basic statistical data in relation to frequencies of occurrence of handwriting characteristics, especially for signatures which are unique calligraphic entities of their owners, are unknown. By comparison, the chance match of DNA grouping of, say, one in a billion reported by two laboratories embodies the same meaning because the entirely objective opinions from both laboratories are the results derived from statistical data. Handwriting experts cannot provide explicit probability figures routinely appearing in DNA grouping reports but instead have to use such ordinary language terms as “highly probable,” “probable,” or “may” to represent the degree of confidence, or certainty.

The scientific validity of the methodology for handwriting examination has been well demonstrated and from the “utilitarian” point of view, testimonies in handwriting analysis do help jurors and judges make their own decisions, although occasionally expert opinions are criticized for containing a certain element of subjectivity. In this connection, Evett makes an interesting point “… that a so-called ‘objective test’ can only exist within a framework of assumptions and the validity of those assumptions in an individual case is a matter for subjective judgment. The objectivity is an illusion” (Evett 1996). A good enlightenment about the relationship between objectivity and subjectivity, which is aptly applicable to handwriting evidence, can be found in The Adventure of the Sussex Vampire where Sherlock Holmes said “… It has been a case for intellectual deduction, but when this original intellectual deduction is confirmed point by point by quite a number of independent incidents, then the subjective becomes objective and we can say confidently that we have reached our goal.” Hence, the probabilistic principles of handwriting identification can, perhaps, be treated from a philosophical perspective.

Bibliography:

Allen RE (ed) (1990) The concise Oxford dictionary of current English, 8th edn. Oxford University Press, Oxfor
Cheung YL, Leung SC (1989) A comparative approach to the examination of Chinese handwriting part 4 – identification by statistical classification techniques. J Forensic Sci Soc 29:77–8
Cole A (1980) The search for certainty and the use of probability. J Forensic Sci 25:826–83
Eisermann HW, Hecker MR (1986) FISH-computers in handwriting examinations. In: Proceedings of the 44^th annual meeting of the American Society of questioned document examiners, Savannah, Georgi
Eldridge MA, Nimmo-Smith I, Wing AM, Totty RN (1984) The variability of selected features in cursive handwriting measures. J Forensic Sci Soc 24:179–21
Ellen DM (1979) The expression of conclusions in hand- writing examination. Can Forensic Sci J 12:117–12
Ellen D (1993) The scientific examination of documents, 1st edn. Ellis Horwood, Chicheste
Evett IW (1996) Expert evidence and forensic misconceptions of the nature of exact science. Sci Justice 36:118–12
Hardcastle RA, Thornton D, Totty RN (1986) A computer-based system for classification of handwriting on cheques. J Forensic Sci Soc 26:383–39
Harrison WR (1966) Suspect documents. Their scientific examination, 2nd edn. Sweet & Maxwell, Londo
Hilton O (1982) Scientific examination of questioned documents, revised edition. Elsevier North Holland, New York
Huber RA, Headrick AM (1999) Handwriting identification: facts and fundamentals. CRC Press, Boca Raton
Leung SC (1994) The scientific examination of Chinese handwriting. Forensic Sci Rev 6(2):97–145
Leung SC, Cheung YL (1989) On opinion. Forensic Sci Int 42:1–13
Leung SC, Chung MWL, Tsui CK, Cheung WL, Cheung SYL, Mok MMC (1987) A comparative approach to the examination of Chinese handwriting part 2 – measurable parameters. J Forensic Sci Soc 27:157–173
Leung SC, Cheung WL, Fung HT, Cheung YL (1993) A comparative approach to the examination of Chinese handwriting part 5 – qualitative parameters. J Forensic Sci Soc 33:9–19
McAlexander TV, Beck J, Dick RM (1991) The standardization of handwriting opinion terminology. J Forensic Sci 36:311–319
McNally JP (1979) Certainty or uncertainty in expert testimony. J Police Sci Admin 7:26–27
Osborn AS (1978) Questioned documents, 2nd edn. Patterson Smith, Monclair
Srihari SN, Cha SH, Arora H, Lee S (2002) Individuality of handwriting. J Forensic Sci 47(4):856–874
Sugiyama T, Kurauchi H (1986) Identification of handwriting in Chinese characters using discriminant analysis. Behaviormetrika 19:55–71
S. 11th Circuit Court of Appeals, United States vs. Paul. (97–9302), 999
S. District Court ruling, United States v. Starzecpyzel. 880 F. Supp. 1027 (S.D.N.Y.), 1995
S. Supreme Court ruling, Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 1993, 584–587
S. Supreme Court ruling, General Electric Co. v. Joiner. (96–199) 78F.3d 524
S. Supreme Court ruling, Kumho Tire Co. v. Carmichael. (97–1709) 131 F.3d 1433
Whiting F (1986) The reasoning process: an integral component of questioned document examinations. In: Paper presented at the 44th annual meeting of the American Society of questioned document examiners, Savannah, Georgia