Assignment On Role of Statistics in Law: Submitted By: Albert Benjamin 1473 Sem IV

You might also like

Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 8

Assignment

On
Role of Statistics in Law

Submitted by: Albert Benjamin


1473
Sem IV

The purpose of statistical science is to provide an objective basis for the analysis of problems
in which the data depart from the laws of exact causality. A general logical system of
inductive reasoning has been devised, is applicable to data of this kind, and is now widely
used in scientific research. Some understanding of its principles is therefore important for . . .
those whose interests lie in the employment of technological advances resulting from
research1.
The important area of statistics can no longer be ignored by lawyers, judges, or legal scholars.
They cite two factors causing this. Because of the increasing intrusion of "federal government
into the economy and scientific research, courts and legislatures have confronted larger and
larger numbers of technical disputes involving problems such as air and water quality, the
safety and economics of nuclear generators, the deregulation of natural gas, and the presence
of carcinogenic agents in the workplace and in the kitchen." 2 A second factor compelling the
use of statistics and quantitative methods in the legal arena is the proliferation of numbers
available to society. Examples are crime rates, health statistics, economic data, educational
levels and achievements, migration patterns, and employment statistic.

In this assignment, we will deal with the following:


● Role of Statistics in various facts of legal field
● Role of Statistician in litigation
● Examples of Statistical Applications in Law
● Limitations in the use of Statistics in Law
● Illustrative Cases

ROLE OF STATISTICS IN VARIOUS FACETS OF LEGAL FIELD


Role in Employment Discrimination cases
Courts generally have accorded statistical evidence great, if not decisive, weight. The
relevance of statistical evidence is premised upon the legal principle that a statistical
demonstration of the unequal consequences of a particular employment practice or selection
process establishes a prima facie case of discrimination.

Role in using a model for Jury Selection


Writing in Trial, Brams and Davis suggest implementation of a mathematical or statistical
model of jury selection. They state: "Since trial lawyers must play averages in an uncertain

1 The Heron II (Koufos v. C. Czarnikow, Ltd) [1967] 3 AU E.R. 694-95 (H.L.)


2 L. Brilmayer and L. Kornhauser, Review: Quantitative Methods and Legal Decisions, 46 U. Chi. L. Rev. 116
(1978).
world, we believe that the optimal challenge thresholds given by our model may be quite
useful, given that its assumptions reasonably approximate actual jury selection procedures.”3

Role in Demonstrating Reliability of Expert Evidence


The testimony in a court of law is based on sufficient facts or data and is the product of
reliable principles and methods that have been reliably applied. Using examples from latent
fingerprint matching and trace evidence (bullet lead and glass), scientists often consider in
assessing the trustworthiness of evidence to enable courts to better distinguish between
“trustworthy” and “questionable” evidence.

ROLE OF STATISTICIAN IN LITIGATION


The unique function of statistics is the measurement of the degree of uncertainty in drawing
inferences from sample data. This involves statistical inference extracting information about
populations from samples and quantifying the risks of error associated with the inferences.
Thus the role of the statistician in litigation is to perform such tasks as determining sample
size needed to provide the desired level of precision to the outcome of an investigation,
compute intervals around sample calculations to allow for uncertainty inherent in the
investigative methods, and to evaluate the significance of differences between independent
samples. Further, in "planning of investigations, the statistician is asked to advise what steps
would lead to results that best fit into the available analytical models, and in the analytical
stage what may be learned from data according to what can be examined in terms of
precision.”4
In short, the use of statisticians and statistics in litigation depends on the legal relevance of
the available data in the case. Where numbers exist and can be treated statistically, the litigant
has in the statistician a witness who can testify objectively as to the quality of the inferences.
Finally, it also should be noted that, in utilizing statisticians as expert witnesses, much
depends on the attitude of the presiding judge toward statistics and quantitative methods.
Since almost any statistical presentation depends upon data which are subject to criticism by
a skillful opponent, a judge who takes a very cautious view of evidentiary rules may be
expected to sustain enough objections by opposing counsel to wipe out such testimony.
Further, it should be obvious that statistical proof can be expected to be more readily digested

3S. Brams and M. Davis, A Game-Theory Approach to Jury Selection, 12 Trial 47 (1976).
4 7J. Dawson, Scientific Investigation of Fact - The Role of the Statistician, 11 Forum 897 (1976).
in a bench trial than in a trial to a jury. Such a presentation should also be better received if it
relates to something other than liability e.g., damages.5

EXAMPLES OF STATISTICAL APPLICATIONS IN LAW


One of the earliest reported instances of proof by statistical probabilities occurred in the
infamous trial in 1899 of Alfred C. Dreyfus.6 To identify writing in a document as that of
Dreyfus, the prosecution's witnesses reported a number of close matches between the lengths
of certain words and letters in a document and the length of certain words and letters in
correspondence taken from Dreyfus's home. As later noted by the panel of experts appointed
to review the evidence in the Dreyfus case, there was nothing statistically remarkable about
the existence of close matches in some word lengths between the disputed document and
Dreyfus's correspondence. Thus, one of the first examples on introducing probabilities as
proof ended on an unsatisfactory note.
Today the state of knowledge has advanced such that statistics are readily used in court cases.
For instance, statistics have been allowed as proof in jury discrimination cases, 7 as well as in
school,8 voting,9 housing,10 and age discrimination cases.11 In Alabama v. United States, the
court stated that "statistics often tell much, and courts listen." In Jones v. Lee Way Motor
Freight, Inc.,12 the court observed that "statistics often demonstrate more than the
testimony of many witnesses."

Another area in which statistics have been relied upon is in anti-trust cases to establish the
relevant market and the defendant's share thereof. To the courts, the degree of concentration
serves as a proxy for the degree of competition in a particular market, an important element
of many antitrust cases and antitrust policy in general.

In United States v. United Shoe Machinery Corp, the court attempted to shorten the
hearing by encouraging the use of sampling device. Statistical presentations have also been
used in deceptive advertising and trademark infringement cases (Bristol-Myers Co. v.
5 Id. at 899, 900
6 L. Tribe, Trial by Mathematics: Precision and Ritual in the Legal Process, 84 Harv. L. Rev. 1332 (1971).
7 Turner v. Fouche, 396 U.S. 346, 360 (1970)
8 United States v. Hines County Board of Education, 417 F.2d 852, 858 (5th Cir. 1969)
9 Alabama v. United States, 304 F.2d 583, 586 (5th Cir.), aff’d per curiam, 371 U.S. 37 (1962).
10 Kennedy Park Homes Ass'n v. Lackawanna, 436 F.2d 108, 113 (2d Cir. 1970), cert, denied, 401 U.S. 1010
(1971).
11 Hodgson v. First Fed. Sav. And Loan Ass'n, 455 F.2d 818, 828 (5th Cir. 1972).
12 3431 F.2d 245 (10th Cir. 1970), cert, denied, 401 U.S. 954 (1971).
FTC). The use of statistical analysis in personal injury and wrongful death cases has been
well documented, particularly life table analysis used in estimating the expected years of
life remaining but for the injury.13

LIMITATIONS IN THE USE OF STATISTICS


Even though statistical evidence has gained widespread acceptance within recent years, the
courts have placed limitations on the use of statistics. As a possible explanation, Whitten
writes that "such limitations demonstrate the innate distrust courts have of statistical evidence
as opposed to direct testimony." Some of the more important limitations as delineated by the
courts appear below.
Sample Size
It is a well-established fact among statisticians that a sample must be representative of the
population from which it is drawn. Otherwise, valid inferences cannot be drawn. Among
other things, the sample must be of adequate size. When the sample size is too small, the
courts may rule that it cannot be used to support any significant conclusion.
Relevance
Statistical data must be relevant, and any comparisons made must be among relevant groups.
When plaintiffs have advanced statistics which courts consider irrelevant to their contentions,
the courts have refused to credit them. In Robinson v. City of Dallas14 a Negro policeman
challenged a rule which called for the disciplining of employees who do not pay their "just
debts." The court concluded that statistics showing that Negroes tended to be less affluent
than Caucasians and tended to have more debts as a group were not relevant to the
case.
Judicial Discretion
Although courts have found statistical evidence necessary and useful, considerable judicial
mistrust of statistics remains. Such skepticism probably arises from the great potential for
misuse and distortion of statistical evidence. In Keely v. Westinghouse Electric Corp.15, the
court stated that "too many use statistics as a drunk man uses a lamppost- for support
and not for illumination”

ILLUSTRATIVE CASES

13 For an application of this concept, see W. Curtis and L. Wilson Determining Loss of Earnings From
Impairment or Death, 37 Ala. Law. 221 (1976)
14 2514 F.2d 1271
15 M04 F. Supp. 543 (E.D. Mo. 1975).
Example 1
This example illustrates the application of statistical techniques in determining economic
losses encountered in the early 1970s by a vertically integrated poultry firm.
There were reduced levels of egg production in all layer flocks controlled by the firm.
Production levels among the 25 flocks, comprising about 500,000 layers, ranged from 9.9
dozen to 19 dozen eggs per year with a mean of 17.88 dozen and a standard deviation of
1.73 dozen. Extensive investigation of the possible causes of reduced production led to the
breeder flock. Eggs produced by the breeder flock were the source of birds in the layer flocks.
It was hypothesized that a genetic fault existed in the breeder flock at the time it was
purchased by the firm. With the problem supposedly originating in the breeder flock, the first
tier, and not significantly revealing itself until the egg production stage, all tiers of the
production processes controlled by the firm were affected. Thus suit was brought against
the breeder to recover damages sustained because of the lower level of egg production.
The initial role of the statistician in this case was determine if the difference between the
expected production per layer of 20 dozen eggs per year (the national norm) and the
average annual output of the flock in question of 17.88 dozen could have occurred due
to chance. Using the i-test for differences between the population mean of 20 and the
sample mean of 17.88, the difference was found to be significant at the .001 level. That is,
the probability is less than .001 (1 in 1000) that a difference occurring by chance
supported the firm's position that genetic factors contributed to reduced egg output.
Lost production was defined as the difference between the actual lay and the expected
norm of 20 dozen eggs per layer per year. Once the extent of production loss was
calculated, appropriate egg prices existing at the time were used in determining the value of
the lost production. This was used as the basis for the award in this case.

Example 2
Forty-one-year-old John Branion was a practicing physician in Chicago in the 1960s. On the
afternoon of December 22, 1967, he and his son came home to find his wife, Donna,
murdered by persons unknown. John called the police. Six months later, he found himself on
trial for murder. However he claimed innocence for the same. Today, he is serving a twenty-
to-thirty year sentence for her murder. During the 1960s, John became close to Martin Luther
King Jr., and in 1966, he was his personal physician. He even helped with medical services
for the Black Panthers. This caused him to have some problems with the Chicago police.
The part of the case which is relevant for us is the plea of Branion for Habeas Corpus. John
Branion's petition for a writ of habeas corpus relied on the defense of factual impossibility.
He argued that he could not have murdered his wife within the relevant time-period, for
even under the state's theory of the crime, that would have required him to drive from
the hospital (where he was working until 11:30) to his home (making two stops en
route), to strangle and shoot his wife, to clean his hands, to dispose of his gun, and to
summon his neighbors - all by 11:57 (when the police logged a call reporting the death).
To show that he could not complete all these tasks in these twenty-seven minutes, Branion
invoked some elementary principles of probability and statistics. Branion's brief to the
Seventh Circuit argued that if the impossibly short time of six minutes were allowed for all
the events other than pure driving and strangling, he still would have had to drive from the
hospital and complete the strangling in twenty-one minutes - a feat that could not be
reconciled with the evidence.
6 min. ...............DRIVING TIME .................. 12 min.
15 min. ...............GARROTING TIME ............... 30 min.
At (a) 15 minutes and (b) 6 minutes, the probabilities are less than 1%. Moreover, since the
two time-sequences are independent of each other, the probabilities must be multiplied to find
the chances that both were applicable to Dr. Branion. Hence the probability that the time
between garroting and shooting was 15 minutes and the probability that the driving
time was 6 minutes is .01 X .01 = .0001.... A reasonable doubt probability of 90%, or 0.9,
is 9,000 times larger in magnitude than the joint probability of .0001. In other words, if
there were 9,000 cases similar to Dr. Branion's where the only two facts known were the
garroting time and the driving time, then 8,999 defendants would be innocent and only one
guilty. In short, the chance that Dr. Branion is guilty, on the basis of the garroting time and
the driving time alone, is one in 9,000.53 As previously indicated, the 1/9000 figure for the
probability of Branion's guilt did not impress the court, at least not favorably. First, noticing
that the "jury could have found that 30 minutes lapsed be- tween Branion's leaving the
Hospital and his call to the police," the court concluded that "[w]e therefore should like to
know the probability that the combination of travel and murder times came to 30 minutes or
less" rather than the twenty-one minutes or less discussed in Branion's brief.

This criticism suggests that Branion set out to solve the wrong problem, but it does not
challenge the method he used to attach a probability to the combined time being less
than some critical figure. The appropriate number may be twenty-one minutes, thirty
minutes, or some other value, but this issue need not detain us. In the face of unavoidable
uncertainty in the times for miscellaneous acts other than driving and garroting, the best
procedure would be to compute the pertinent probabilities for all the plausible numbers in the
twenty-one to thirty minute range.

In a Nutshell

Problems posed in the courts result from the conditions and needs of society. Because of
technical and societal changes, a growing proportion of legal issues involves the use of
numerical data. Statistics, which deals with the analysis and interpretation of numerical data,
is a useful tool to aid the courts in decision making. Purpose of this article was to examine the
application of statistics in the litigation process. Using samples drawn from a population or
universe, the statistician employs a system of logical reasoning through which statistical
inferences can be made the risks of error associated with the inferences can be quantified.
This approach helps to reduce uncertainty or translate uncertainty into a form understandable
to the courts. However, statistics cannot guarantee the correct inference and does not
establish guilt. Establishment of guilt is beyond the competence of the statistician. Frustration
or reluctance on the part of the courts in accepting the application of statistical tools is
understandable. Lack of familiarity with these techniques, however, should not result in a
condemnation of statistics and statisticians. Technical language surrounding the tool creates a
communications problem; initially, the problem may be fanned by overestimation of the
contribution of statistics by both proponents and critics. Ultimately, the validity of the
application of appropriate techniques developed within the profession will become apparent
and the accompanying technical language will be accepted by the courts.

You might also like