Ateneo Graduate School of Business
Rockwell Drive, Makati City
Statistics and Ethics
In Partial Fulfillment
Of the Requirements for
Principles of Management
Master in Business
Administration
Standard Program
Submitted by:
Ana Marie Canaliza
Rochelle Leonor
Ma. Araceli Pono
Submitted to:
Prof. Jim Pacheco
Introduction
Statistics: A Background
Statistics
is a branch of applied mathematics that deals with the collection and
interpretation of quantitative data. It is also concerned with the use of
probability theory to estimate population parameters.
Statistics
goes back as far as history itself. As early as 3800 BC, populations in
Babylonia were already accounted for. Census was used to count adult male
citizens for military and administrative purposes.
It
was Achenwall who first used the word ‘statistiks’ sometime in the 1700s. But
it was Zimmerman and John Sinclair who popularized the term ‘statistics’ in
their publication. However, the most prominent figure in the field was Sir
Ronal Fisher who made important contributions from 1912 to 1962, many of which
have significant impact on contemporary statistical procedures.
The field of ethics involves systematizing, defending
and recommending concepts of right and wrong behavior. Also called moral
philosophy, the study of ethics deals with doing the right thing.
Just like the field of statistics, the study of
ethics goes a long way back in history, even during the time of Plato and
Aristotle. Philosophers have always discussed the difference between right and
wrong,
Of course, the field is subject to much debate. Since
ethics is subjected to much relativity, what is ethical to one may not
necessarily be ethical to another. Thus, the debate is still ongoing.
Unethical Use of Statistics
Just like in the field of
accounting, where figures can be manipulated to push a company’s agenda, the
field of statistics is also subject to such unethical use. While balance sheets
are sometimes overstated to show the market that a company is healthy,
statistical data are also manipulated to promote one’s ulterior plan.
In the New York Times June
17 1995 edition, Dick Teresi—editor of a magazine for Harley Davidson
bikers—wrote that helmets may actually kill users. In the article “The case for
no helmets,” he argued that although helmets may protect the head, it can also
break the spine in collisions at speeds greater than 15 miles an hour. Teresi
cited a study indicating that, “Nine US states without helmet laws had a lower
fatality rate (3.05 deaths per 10,000 motorcycles) than those mandated helmets
(3.38). He further supported his argument that in a survey of 2,500, 98 percent
opposed such laws.
While the data is not
exactly unethical, it still misleads the readers to believing that not wearing
helmet is the best way to go. Helmet laws were passed because it was proven
that wearing these headgears save more lives.
Another classic example of
an unethical use of statistics was presented in the ABC television program
“20/20” by Barbara Walters. The program tackled whether survey results are
facts or works of fiction.
In its March 31, 1995
episode, ABC correspondent John Stossel investigated a survey comparing the
discipline problems experienced by teachers in the 1940s and those experienced
today. Results show that teachers in the 1940s mostly worried about students
talking in class, chewing gum and running in the halls. Today, however,
teachers worry about being assaulted.
This was highly publicized
in the print media that even then-First Lady Barbara Bush and former Education
secretary William Bennett were worried.
But Stossel doubted the
reliability of this data. With the assistance of a Yale professor, Stossel
found the original source of the survey and discovered it wasn’t a survey at
all. Texan oilman T. Colin Davis, the source, has simply identified certain
disciplinary problems encountered by teachers. The list was not obtained from a
statistical survey but was based on Davis’ personal knowledge of the problems
in the 1940s since he was a student at that time. Several more misleading (and
possibly unethical) surveys were presented on the ABC program.
Clair Felbinger of the
American University wrote a book dealing with unethical use of statistics.
Entitled “Lying with statistics,’ she discussed the how practitioners use
statistics to lie. She cautions statisticians to be cautious in presenting data
since many people rely on statistics to determine trends and to judge public
opinion.
In her book, she cited one
example of violating statistical assumptions, which resulted in an incorrect
result. In the 1936 Literary Digest magazine, its presidential preference poll
predicted that Alf Landon would defeat Franklin Roosevelt in a landslide.
Roosevelt won.
What went wrong? Pollsters
at the Literary Digest used a biased sample. However, statistics indicate that
one can only make predictions if he has a random sample of the population—in
this case, all eligible voters. In a random sampling, each person in the
population must have an equal and non-zero chance of being included in the
sample.
In the case of the Literary
Digest poll, the potential respondents were readers of the magazines and those
who had telephones. The poll was also taken during the Depression period where
most poor people did not have phones. Therefore, the poll was biased against
poor people. It was partial to people with higher educations since they read
the literary magazine. Wealthy people and those better educated also tended to
be Republicans, which was Landon’s party.
Felbinger proposes five
simple questions people may pose when confronting the veracity of statistics.
First, she says people must ask, “Who says
so?” It must be determined who generated the statistics and whether they
stand behind the data. “How does
he or she know?” People must ask whether the sample was biased? The
N must be large enough to allow a reliable conclusion. “What’s missing?” The number of cases should be reported, as well as the standard error.
“Did somebody change the subject?” Did the
incidence of a condition increase over time or are the data gathered more
carefully now? “Does it make sense?”
The American Statistical Association formulated
Ethical Guidelines for Statistical Practice. These guidelines serve a framework
for reliable statistical practice to guide practitioners and readers from
possible misunderstanding, mistreatment, misuse and misinterpretation of data
gathered. Since such information provides significant foundation for
development in the economic, demographics, social and environmental fields, it
should be confined to ethical considerations.
Statisticians, therefore, have a public obligation to
uphold integrity and righteousness in the application of their expertise to
contingencies where private interests may inappropriately influence the
development of their statistical knowledge and judgment.
Statistical work must be discernable and exposed to
appraisal with respect to quality and pertinence in order to advance knowledge.
Such evaluation may comprise rationale of the postulation, approach, and data
processing utilized by another person. However, it must also be understood that
all data presented initially by statisticians should not be misleading and had
been verified to be true and accurate.
The ethical guidelines for statistical practice
ascertain the identity of ethical alliance with the public, government,
clients, or employers and other professionals. For the statisticians to uphold
the integrity of their work, they have a responsibility to divulge and are
subject to the following:
1.
Honesty and objectivity
manifest their findings and interpretations.
2.
Avert incorrect,
deceptive, or undocumented statements.
3.
Disclose any financial
or other interests that may affect their professional statements.
4.
Collect only data
necessary for the purpose of their inquiry.
5.
Advise each potential
respondent concerning the general nature and sponsorship of the research and
use of the data.
6.
Delineate the
boundaries of the inquiry as well as the boundaries of the statistical
inferences, which can be acquired from it.
7.
Be prepared to document
data sources utilized in an inquiry; known erroneousness in the data; and steps
taken to modify or to refine the data, statistical procedures applied to the
data, and the assumption is required for their usage.
8.
Make the data available
for analysis by other responsible parties with pertinent preventives for
privacy concerns.
9.
Perceive that the
choice of a statistical procedure may to some point be a matter of judgment and
that other statisticians may select alternative programs.
Since other individuals, who are unaccustomed with
statistical exercises might be dependent on the statistician’s expertise and
recommendations; statisticians should put effort to furnish not only reliable
data but also dependable directions. By this, they should strive to make clear
their credentials to commence the statistical study. They should also employ
statistical procedures without concerns for a favorable outcome/results.
Statisticians should only participate in work/study, which corresponds to
accepted ethical standards and which they are qualified and skilled to perform.
The Ethical Guidelines also address eight general
topic areas and specify important ethical considerations under each topic:
A. Professionalism points out the need for competence,
judgment, diligence, self-respect, and worthiness of the respect for other
people.
B. Responsibilities to
Founders, Clients, and Employers discusses the practitioner’s responsibility for assuring
that statistical work is suitable to the needs and resources of those who are
paying for it, that founders understand the capabilities and limitations of
statistics in addressing their problem, and that the founder’s confidential
information is protected.
C. Responsibilities in
Publications and Testimony addresses the need to report sufficient information to give readers,
including other practitioners, a clear understanding of the intent of the work,
how and by whom it was performed, and any limitations on its validity.
D. Responsibilities to
Research Subjects describes
requirements for protecting the interests of human and animal subjects of
research -- not only during data collection but also in the analysis,
interpretation, and publication of the resulting findings.
E. Responsibilities to
Research Team Colleagues addresses the mutual responsibilities of professionals participating in
multidisciplinary research teams.
F. Responsibilities to
Other Statisticians or Statistical Practitioners notes the interdependence of professionals
doing similar work, whether in the same or different organizations. Basically,
they must contribute to the strength of their professions overall, by sharing
nonproprietary data and methods, by participating in peer review, and by
respecting differing professional opinions.
G. Responsibilities
Regarding Allegations of Misconduct addresses the sometimes painful process of investigating
potential ethical violations and treating those involved with both justice and
respect.
H. Responsibilities of
Employers, Including Organizations, Individuals, Attorneys, or Other Clients
Employing Statistical Practitioners encourages employers and clients to recognize the highly
interdependent nature of statistical ethics and statistical validity. Employers
and clients must not pressure practitioners to produce a particular “result”
regardless of its statistical validity. They must avoid the potential social
harm that can result from the dissemination of false or misleading statistical
work.
Practitioners should be
cautious when presenting statistical data since these data can be a major
factor in formulating rules and regulations. Readers are also advised to
exercise caution when interpreting data. They must be analytical and not easily
believe everything they read. In the words of Cynthia Crossen, author of
‘Tainted Truth,” an exposé on misleading and biased surveys, “If everybody is misusing numbers and scaring us with
numbers to get us to do something, however good [that something] is, we’ve lost
the power of numbers.
www4.gvsu.edu/robbinsd/courses/
pa611/readings/felbinger.
http://calcnet.cst.cmich.edu/faculty/lanfear/ReadingStats.html
www.dartmouth.edu/~chance/course/Syllabi/
97Dartmouth/day-5/hiv.pdf