Ethics and StatisticsEthics and Statistics
What is Ethics?
Is ethics an examination of :
- when an act is right or wrong?
- what kinds of things are good or desirable?
- when a person deserves blame, reward or neither?
- what kinds of virtues are likely to lead to good behavior?
You are right. It is all of the above!
"a set of morally-permissible standards of conduct all members of a group want each other to follow."
Moral literacy is a combination of a number of complex skills and abilities that all individuals should aspire to acquire. These skills and abilities are behaviors can be describe as a person being able to:
- recognize moral problems and to assess the complex issues that they raise,
- identify and appreciate the underlying ethical values,
- evaluate moral problems from amny perspectives
- assess disagreements on and propose responses to moral problems, and
- choose to act with wisdom and responsibility.
Studying and learning about ethics is a very broad and expansive topic. As statisticians, what we can do is focus on recognizing the ethical issues as they appear in statistical practice. And, to help us manage the ethical dilemmas that we do confront we need to develop a framework for how to think about ethical issues.
Components of Moral Literacy
Ethics sensitivity, the ability to recognize more problems when they occur and appreciate the full significance of the situation, ethical reasoning, the ability to evaluate moral problems through an understanding of the major ethical frameworks and moral imagination, the ability to understand and analyze a wide range of disagreements about and proposed responses to these problems; these are all components of moral literacy.
Part of your academic preparation as a statistician should include this awareness and appreciation of the bigger ethical picture of certain scenarios you will no doubt encounter so that you will be able to think through these dilemmas considering a wide array of solutions and implications in a thoughtful manner.
The first step is to be able to recognize ethical issues in the first place. Are you proud or ashamed of an action? Does it feel 'wrong' for some reason? Emotions often help reveal what is important to us, what we value and what effects us. Is your action compatible with your own basic ethical values? Is the action compatible with ethical principles?
Categories of Moral Senses (Jonathan Haidt, Social and Moral Psychologist):
- Loyalty to In-group
- Respect for authority or hierarchy
- Purity / sanctity
What if a client asks you to eliminate data because it doesn’t fit? How do you think about this kind of request?
Start by asking yourself some questions...
Ethical Thinking / Ethical Reasoning Skills
- Are you proud or ashamed of the action? Does it feel wrong?
- Is such an action compatible with your basic ethical values?
- Is the action compatible with ethical principles (these can be gathered by looking at the Statistical Ethics)
The idea is to think through what the issues are, therefore, ethical thinking involves making your value considerations obvious, to bring them out into the open so that you can examine them against our ethical values.
Our ethical values help us to define what we consider right and wrong. Ethical values guide individuals as well as groups and can serve as goals and ideals that we aspire to and by which we measure ourselves, others and societies. Examples include what we think of as fair, being just or impartial. Respect is another example of an ethical value whereby we acknowledge others.
An ethical framework can help to lay out our values. For instance, a virtue-based approach might involve principles related to the development of good habits of character including:
"Practical wisdom cannot be acquired solely by learning general rules. We must also acquire, through practice, those deliberative, emotional, and social skills that enable us to put our general understanding of well-being into practice in ways that are suitable to each occasion."
Formalizing an agreed upon approach to managing ethical dilemmas and stating these principles as a central aspect for how a company or an institution carries itself in day-to-day activities is an example of ethical thinking. These published ethical principles that form the backbone of an institution's perspective on the conduct of business are often labeled as a 'Code of Conduct'. Statisticians at Penn State are guided by two examples of ethical thinking that will be examined in greater depth.
- The Penn State Principles
- The American Statistical Association Ethical Guidelines
What about Academic Integrity? Is this an intrinsic or extrinsic value? Is it only a negative value comprised of a list of things that you are not supposed to do, or does it have is own positive content? As educators it is our responsibility to introduce others to the community of statisticians, to prepare them for its responsibilities, to help them anticipate poor decisions, mistakes and temptations.
Moral imagination is, given an ethical problem, being able to creatively consider a variety of choices, the consequences from a variety of perspectives.
"An ability to imaginatively discern various possibilities for acting in a given situation and to envision the potential help and harm that are likely to result from a given action."
Moral imagination allows us to experience compassion, to imagine ourselves in the situation of another. This is not what we would feel or believe in that situation but what another, given their values, beliefs, and character is undergoing. How well can you 'think outside the box'? How well can you genuinely consider situations from other perspectives? In doing this you are developing a much sensitive attunement to the complexities of the situation.
Work as a statistical consultant touches nearly every field of scientific study. As a result, the statistical consultant may be confronted with potential ethical challenges from any number of contexts and there is simply no way to categorize and prescribe what actions should be taken in each case. Instead, the statistical consultant must have at their familiar disposal a trusted method for evaluating what can be very complicated challenges to guide their ethical decision making. The key phrase here is a method for thinking through what their response should be.
Seven-Step Guide to Ethical Decision-Making
State the problem.For example, "there's something about this decision that makes me uncomfortable" or "do I have a conflict of interest?".
Check facts.Many problems disappear upon closer examination of situation, while others change radically.
Identify relevant factors.For example, persons involved, laws, professional code, other practical constraints ( e.g. under $200).
Develop list of options.Be imaginative, try to avoid "dilemma"; not "yes" or" no" but whom to go to, what to say.
Test options.Use such tests as the following:
Harm Test: Does this option do less harm than alternatives?
Publicity Test: Would I want my choice of this option published in the newspaper?
Defensibility Test: Could I defend choice of option before congressional committee or committee of peers?
Reversibility Test: Would I still think choice of this option good if I were adversely affected by it?
Colleague Test: What do my colleagues say when I describe my problem and suggest this option is my solution?
Professional Test: What might my profession's governing body for ethics committee say about this option?
Organization Test: What does the company's ethics officer or legal counsel say about this?
Make a choice based on steps 1-5.
Review steps 1-6.
What could you do to make it less likely that you would have to make such a decision again?
Are there any cautions you can take as an individual (and announce your policy on question, job change, etc.)?
Is there any way to have more support next time?
Is there any way to change the organization ( for example, suggest policy change at the next departmental meeting)?
(From Davis, Michael (1999) Ethics and the University, New York: Routledge, p. 166-167.)
Moral literacy does not give you a pre-prescribed set of rules to follow for every situation. There are just too many circumstances and complexities that change from situation to situation to make a formal set of rules possible for every situation. Instead, one looks for a morally guided path and working through a process like the one above provides a process for managing your thoughts, perceptions and your decisions.
It Takes Courage!
These are not easy decisions and sometimes there are associated costs or risks. Sometimes we have to admit we are wrong! Generating possible actions is helpful because sometimes the "wrong" answer is obvious and we need to really think about the "right" one.
In the long run, the goal is to do your reasonable best and take responsibility for your actions.
Scholarship and Research IntegrityScholarship and Research Integrity
At Penn State University research integrity is a critical ingredient in any research program. The Schloarship and Research Integrity (SARI) program was launched in fall 2009 to provide graduate students with opportunities to identify, examine, and discuss ethical issues relevant to their disciplines. This program was expanded in 2011 to include faculty, postdoctoral fellows as well as undergraduate students - essentially all individuals that are involved in research that takes place at Penn State.
The SARI @PSU program is composed of two parts: an online course, and an interactive, discussion-based component; and encompasses content that is both interdisciplinary and discipline-specific. The online portion (Part 1), offered through the Collaborative Institutional Training Initiative (CITI), provides a common language and understanding of the history and principles of the responsible conduct of research. The discussion-based component (Part 2) provides an opportunity for in-depth exploration of important issues unique to each field of study.
The Penn State ValuesThe Penn State Values
The Penn State Values are guidelines (not rules) that provide a set of values to help students make ethical choices in their daily lives. Each Value is founded on academic values essential to the maintenance and flourishing of the Penn State community.
The Penn State Values
The Pennsylvania State University is a community dedicated to personal and academic excellence. The Penn State Values were developed to embody the values that we hope our students, faculty, staff, administration, and alumni possess. At the same time, the University is strongly committed to freedom of expression. Consequently, these Values do not constitute University policy and are not intended to interfere in any way with an individual’s academic or personal freedoms. We hope, however, that individuals will voluntarily endorse these common principles, thereby contributing to the traditions and scholarly heritage left by those who preceded them, and will thus leave Penn State a better place for those who follow.
We act with integrity and honesty in accordance with the highest academic, professional, and ethical standards.
We respect and honor the dignity of each person, embrace civil discourse, and foster a diverse and inclusive community.
We act responsibly, and we are accountable for our decisions, actions, and their consequences.
We seek and create new knowledge and understanding, and foster creativity and innovation, for the benefit of our communities, society, and the environment.
We strive for excellence in all our endeavors as individuals, an institution, and a leader in higher education.
We work together for the betterment of our University, the communities we serve, and the world.
ASA Ethics GuidelinesASA Ethics Guidelines
American Statistical Association's Committee on Professional Ethics published the following:
Ethical Guidelines for Statistical Practice
From their website:
The Ethical Guidelines address eight general topic areas and specify important ethical considerations under each topic.
- Professionalism points out the need for competence, judgment, diligence, self-respect, and worthiness of the respect of other people.
- Responsibilities to Funders, 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 funders understand the capabilities and limitations of statistics in addressing their problem, and that the funder's confidential information is protected.
- 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.
- 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.
- Responsibilities to Research Team Colleagues addresses the mutual responsibilities of professionals participating in multidisciplinary research teams.
- 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, participating in peer review, and respecting differing professional opinions.
- Responsibilities Regarding Allegations of Misconduct addresses the sometimes painful process of investigating potential ethical violations and treating those involved with both justice and respect.
- 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.
P-STAT AccreditationP-STAT Accreditation
The practice of Statistics is a job for skilled professionals. Accredited statisticians have been recognized by their peers as combining education, experience, competence, and commitment to ethics at a level that labels them as professionals. Accreditation provides a measure of assurance to employers, contractors and collaborators of statisticians, and a mark of accomplishment to society at large.
The ASA's accreditation program is modeled after programs in Australia, Canada, and the United Kingdom. Accreditation is a portfolio-based rather than an examination-based credential and is renewable every five years. Accreditation is also voluntary; applicants seek accreditation because they believe the credential is worthwhile to them, but it is not a requirement for practice.
Accreditation applicants will submit materials to be reviewed by members of the ASA Accreditation Committee, peers who will evaluate submissions based on the ASA's Guidelines for Accreditation. Those who meet these guidelines will be awarded the designation "accredited professional statistician."
ASA accreditation is a voluntary credential offered to ASA members that provides peer recognition for all of the following:
- Having advanced statistical training and knowledge
- Having experience in applying statistical expertise competently
- Maintaining appropriate professional development
- Agreeing to abide by ethical standards of practice
- Being able to communicate effectively
There is a fee to apply for and an annual fee to maintain accreditation.
Ethics ScenariosEthics Scenarios
The following scenarios provide a few practice scenarios for you to apply the ethical considerations discussed in this section.
As the person in charge of transportation, you research new transportation techniques for the Penn State University Park campus and find that there is a low cost alternative to the CATA buses. Unfortunately, there are some troubling safety issues related to this alternative but you find several other colleges that have implemented this new technique. You ask each college for data on the safety numbers of the new technique. You analyze the data and realize that if you eliminate one of the colleges there is no difference in the safety of the old vs. new techniques. And, your budget will be balanced if you implement the new technique. What should you do?
- Can you identify the elements of the American Statistical Association Ethical Guidelines that come into question in the example?
- How do the guidelines help you considering the action you might take?
While you are working on a problem on an exam you accidentally notice that the student next to you has a different answer. You decide to go back and check your work. After realizing that you made a mistake, you change your answer.
- Is this an act of academic integrity?
- What did the student do wrong?
- How could this have been avoided?
A student is accused of plagiarism in a consulting report. While most of the report appears to be original, the professor finds several paragraphs that are directly copied from a journal article, which is cited in the students reference list. The student, convinced that he did not plagiarize, produces all of the notes he used when writing his paper. it is discovered that the copying was a result of sloppy note taking, because the student did not distinguish between his own notes and the words of the article's author in his notes.
- Is this plagiarism?
- Is this student responsible for a violation of academic integrity, or is this merely a case of 'bad study habits'?
Alf Josefson studied benthic organisms from 14 areas of Skagerrak. His study determined that the area is eutrophic, based on a trend indicating increased biomass (weight per m2) that he found when combining data. The slope of the regression lines of biomass was positive at 12 of 14 areas. According to the researcher, this change was not due to chance and he argued that the conclusions could be extended to the whole of Skagerrak; he claims there was a general increase in biomass and that the changes in biomass are caused by "human generated input of nutrients to the sea and that there is evidence of eutrophication" (p. 404). John gray responded, arguing that there is no definitive evidence of eutrophication in Skagerrak. He claimed that Josefson improperly used statistical methods when combining data and of using the precautionary principle instead of statistic norms. He accused Josefson of "crying wolf" (p. 405) to the public on an issue of great political importance. In his reply to Gray, Josefson asks if scientists should not publish results until certainty has been obtained or if they can "warn on the basis of less security" (p. 405).
This disagreement is about (at least) two questions:
- Is it legitimate to combine data the way Josefson does?
- When is it appropriate to communicate? How certain should one be before going public?
- What should this statistical consultant do?
- What actions should they take, if any? Why?