are case studies ethical

McCombs School of Business

• Español ( Spanish )

Videos Concepts Unwrapped View All 36 short illustrated videos explain behavioral ethics concepts and basic ethics principles. Concepts Unwrapped: Sports Edition View All 10 short videos introduce athletes to behavioral ethics concepts. Ethics Defined (Glossary) View All 58 animated videos - 1 to 2 minutes each - define key ethics terms and concepts. Ethics in Focus View All One-of-a-kind videos highlight the ethical aspects of current and historical subjects. Giving Voice To Values View All Eight short videos present the 7 principles of values-driven leadership from Gentile's Giving Voice to Values. In It To Win View All A documentary and six short videos reveal the behavioral ethics biases in super-lobbyist Jack Abramoff's story. Scandals Illustrated View All 30 videos - one minute each - introduce newsworthy scandals with ethical insights and case studies. Video Series

Case Studies UT Star Icon

Case Studies

More than 70 cases pair ethics concepts with real world situations. From journalism, performing arts, and scientific research to sports, law, and business, these case studies explore current and historic ethical dilemmas, their motivating biases, and their consequences. Each case includes discussion questions, related videos, and a bibliography.

A Million Little Pieces

James Frey’s popular memoir stirred controversy and media attention after it was revealed to contain numerous exaggerations and fabrications.

Abramoff: Lobbying Congress

Super-lobbyist Abramoff was caught in a scheme to lobby against his own clients. Was a corrupt individual or a corrupt system – or both – to blame?

Apple Suppliers & Labor Practices

Is tech company Apple, Inc. ethically obligated to oversee the questionable working conditions of other companies further down their supply chain?

Approaching the Presidency: Roosevelt & Taft

Some presidents view their responsibilities in strictly legal terms, others according to duty. Roosevelt and Taft took two extreme approaches.

Appropriating “Hope”

Fairey’s portrait of Barack Obama raised debate over the extent to which an artist can use and modify another’s artistic work, yet still call it one’s own.

Arctic Offshore Drilling

Competing groups frame the debate over oil drilling off Alaska’s coast in varying ways depending on their environmental and economic interests.

Banning Burkas: Freedom or Discrimination?

The French law banning women from wearing burkas in public sparked debate about discrimination and freedom of religion.

Birthing Vaccine Skepticism

Wakefield published an article riddled with inaccuracies and conflicts of interest that created significant vaccine hesitancy regarding the MMR vaccine.

Blurred Lines of Copyright

Marvin Gaye’s Estate won a lawsuit against Robin Thicke and Pharrell Williams for the hit song “Blurred Lines,” which had a similar feel to one of his songs.

Bullfighting: Art or Not?

Bullfighting has been a prominent cultural and artistic event for centuries, but in recent decades it has faced increasing criticism for animal rights’ abuse.

Buying Green: Consumer Behavior

Do purchasing green products, such as organic foods and electric cars, give consumers the moral license to indulge in unethical behavior?

Cadavers in Car Safety Research

Engineers at Heidelberg University insist that the use of human cadavers in car safety research is ethical because their research can save lives.

Cardinals’ Computer Hacking

St. Louis Cardinals scouting director Chris Correa hacked into the Houston Astros’ webmail system, leading to legal repercussions and a lifetime ban from MLB.

Cheating: Atlanta’s School Scandal

Teachers and administrators at Parks Middle School adjust struggling students’ test scores in an effort to save their school from closure.

Cheating: Sign-Stealing in MLB

The Houston Astros’ sign-stealing scheme rocked the baseball world, leading to a game-changing MLB investigation and fallout.

Cheating: UNC’s Academic Fraud

UNC’s academic fraud scandal uncovered an 18-year scheme of unchecked coursework and fraudulent classes that enabled student-athletes to play sports.

Cheney v. U.S. District Court

A controversial case focuses on Justice Scalia’s personal friendship with Vice President Cheney and the possible conflict of interest it poses to the case.

Christina Fallin: “Appropriate Culturation?”

After Fallin posted a picture of herself wearing a Plain’s headdress on social media, uproar emerged over cultural appropriation and Fallin’s intentions.

Climate Change & the Paris Deal

While climate change poses many abstract problems, the actions (or inactions) of today’s populations will have tangible effects on future generations.

Cover-Up on Campus

While the Baylor University football team was winning on the field, university officials failed to take action when allegations of sexual assault by student athletes emerged.

Covering Female Athletes

Sports Illustrated stirs controversy when their cover photo of an Olympic skier seems to focus more on her physical appearance than her athletic abilities.

Covering Yourself? Journalists and the Bowl Championship

Can news outlets covering the Bowl Championship Series fairly report sports news if their own polls were used to create the news?

Cyber Harassment

After a student defames a middle school teacher on social media, the teacher confronts the student in class and posts a video of the confrontation online.

Defending Freedom of Tweets?

Running back Rashard Mendenhall receives backlash from fans after criticizing the celebration of the assassination of Osama Bin Laden in a tweet.

Dennis Kozlowski: Living Large

Dennis Kozlowski was an effective leader for Tyco in his first few years as CEO, but eventually faced criminal charges over his use of company assets.

Digital Downloads

File-sharing program Napster sparked debate over the legal and ethical dimensions of downloading unauthorized copies of copyrighted music.

Dr. V’s Magical Putter

Journalist Caleb Hannan outed Dr. V as a trans woman, sparking debate over the ethics of Hannan’s reporting, as well its role in Dr. V’s suicide.

East Germany’s Doping Machine

From 1968 to the late 1980s, East Germany (GDR) doped some 9,000 athletes to gain success in international athletic competitions despite being aware of the unfortunate side effects.

Ebola & American Intervention

Did the dispatch of U.S. military units to Liberia to aid in humanitarian relief during the Ebola epidemic help or hinder the process?

Edward Snowden: Traitor or Hero?

Was Edward Snowden’s release of confidential government documents ethically justifiable?

Ethical Pitfalls in Action

Why do good people do bad things? Behavioral ethics is the science of moral decision-making, which explores why and how people make the ethical (and unethical) decisions that they do.

Ethical Use of Home DNA Testing

The rising popularity of at-home DNA testing kits raises questions about privacy and consumer rights.

Flying the Confederate Flag

A heated debate ensues over whether or not the Confederate flag should be removed from the South Carolina State House grounds.

Freedom of Speech on Campus

In the wake of racially motivated offenses, student protests sparked debate over the roles of free speech, deliberation, and tolerance on campus.

Freedom vs. Duty in Clinical Social Work

What should social workers do when their personal values come in conflict with the clients they are meant to serve?

Full Disclosure: Manipulating Donors

When an intern witnesses a donor making a large gift to a non-profit organization under misleading circumstances, she struggles with what to do.

Gaming the System: The VA Scandal

The Veterans Administration’s incentives were meant to spur more efficient and productive healthcare, but not all administrators complied as intended.

German Police Battalion 101

During the Holocaust, ordinary Germans became willing killers even though they could have opted out from murdering their Jewish neighbors.

Head Injuries & American Football

Many studies have linked traumatic brain injuries and related conditions to American football, creating controversy around the safety of the sport.

Head Injuries & the NFL

American football is a rough and dangerous game and its impact on the players’ brain health has sparked a hotly contested debate.

Healthcare Obligations: Personal vs. Institutional

A medical doctor must make a difficult decision when informing patients of the effectiveness of flu shots while upholding institutional recommendations.

High Stakes Testing

In the wake of the No Child Left Behind Act, parents, teachers, and school administrators take different positions on how to assess student achievement.

In-FUR-mercials: Advertising & Adoption

When the Lied Animal Shelter faces a spike in animal intake, an advertising agency uses its moral imagination to increase pet adoptions.

Krogh & the Watergate Scandal

Egil Krogh was a young lawyer working for the Nixon Administration whose ethics faded from view when asked to play a part in the Watergate break-in.

Limbaugh on Drug Addiction

Radio talk show host Rush Limbaugh argued that drug abuse was a choice, not a disease. He later became addicted to painkillers.

U.S. Olympic swimmer Ryan Lochte’s “over-exaggeration” of an incident at the 2016 Rio Olympics led to very real consequences.

Meet Me at Starbucks

Two black men were arrested after an employee called the police on them, prompting Starbucks to implement “racial-bias” training across all its stores.

Myanmar Amber

Buying amber could potentially fund an ethnic civil war, but refraining allows collectors to acquire important specimens that could be used for research.

Negotiating Bankruptcy

Bankruptcy lawyer Gellene successfully represented a mining company during a major reorganization, but failed to disclose potential conflicts of interest.

Pao & Gender Bias

Ellen Pao stirred debate in the venture capital and tech industries when she filed a lawsuit against her employer on grounds of gender discrimination.

Pardoning Nixon

One month after Richard Nixon resigned from the presidency, Gerald Ford made the controversial decision to issue Nixon a full pardon.

Patient Autonomy & Informed Consent

Nursing staff and family members struggle with informed consent when taking care of a patient who has been deemed legally incompetent.

Prenatal Diagnosis & Parental Choice

Debate has emerged over the ethics of prenatal diagnosis and reproductive freedom in instances where testing has revealed genetic abnormalities.

Reporting on Robin Williams

After Robin Williams took his own life, news media covered the story in great detail, leading many to argue that such reporting violated the family’s privacy.

Responding to Child Migration

An influx of children migrants posed logistical and ethical dilemmas for U.S. authorities while intensifying ongoing debate about immigration.

Retracting Research: The Case of Chandok v. Klessig

A researcher makes the difficult decision to retract a published, peer-reviewed article after the original research results cannot be reproduced.

Sacking Social Media in College Sports

In the wake of questionable social media use by college athletes, the head coach at University of South Carolina bans his players from using Twitter.

Selling Enron

Following the deregulation of electricity markets in California, private energy company Enron profited greatly, but at a dire cost.

Snyder v. Phelps

Freedom of speech was put on trial in a case involving the Westboro Baptist Church and their protesting at the funeral of U.S. Marine Matthew Snyder.

Something Fishy at the Paralympics

Rampant cheating has plagued the Paralympics over the years, compromising the credibility and sportsmanship of Paralympian athletes.

Sports Blogs: The Wild West of Sports Journalism?

Deadspin pays an anonymous source for information related to NFL star Brett Favre, sparking debate over the ethics of “checkbook journalism.”

Stangl & the Holocaust

Franz Stangl was the most effective Nazi administrator in Poland, killing nearly one million Jews at Treblinka, but he claimed he was simply following orders.

Teaching Blackface: A Lesson on Stereotypes

A teacher was put on leave for showing a blackface video during a lesson on racial segregation, sparking discussion over how to teach about stereotypes.

The Astros’ Sign-Stealing Scandal

The Houston Astros rode a wave of success, culminating in a World Series win, but it all came crashing down when their sign-stealing scheme was revealed.

The Central Park Five

Despite the indisputable and overwhelming evidence of the innocence of the Central Park Five, some involved in the case refuse to believe it.

The CIA Leak

Legal and political fallout follows from the leak of classified information that led to the identification of CIA agent Valerie Plame.

The Collapse of Barings Bank

When faced with growing losses, investment banker Nick Leeson took big risks in an attempt to get out from under the losses. He lost.

The Costco Model

How can companies promote positive treatment of employees and benefit from leading with the best practices? Costco offers a model.

The FBI & Apple Security vs. Privacy

How can tech companies and government organizations strike a balance between maintaining national security and protecting user privacy?

The Miss Saigon Controversy

When a white actor was cast for the half-French, half-Vietnamese character in the Broadway production of Miss Saigon , debate ensued.

The Sandusky Scandal

Following the conviction of assistant coach Jerry Sandusky for sexual abuse, debate continues on how much university officials and head coach Joe Paterno knew of the crimes.

The Varsity Blues Scandal

A college admissions prep advisor told wealthy parents that while there were front doors into universities and back doors, he had created a side door that was worth exploring.

Providing radiation therapy to cancer patients, Therac-25 had malfunctions that resulted in 6 deaths. Who is accountable when technology causes harm?

Welfare Reform

The Welfare Reform Act changed how welfare operated, intensifying debate over the government’s role in supporting the poor through direct aid.

Wells Fargo and Moral Emotions

In a settlement with regulators, Wells Fargo Bank admitted that it had created as many as two million accounts for customers without their permission.

Stay Informed

Support our work.

Annual Review of Ethics Case Studies

What are research ethics cases.

For additional information, please visit Resources for Research Ethics Education

Research Ethics Cases are a tool for discussing scientific integrity. Cases are designed to confront the readers with a specific problem that does not lend itself to easy answers. By providing a focus for discussion, cases help staff involved in research to define or refine their own standards, to appreciate alternative approaches to identifying and resolving ethical problems, and to develop skills for dealing with hard problems on their own.

Research Ethics Cases for Use by the NIH Community

• Theme 23 – Authorship, Collaborations, and Mentoring (2023)
• Theme 22 – Use of Human Biospecimens and Informed Consent (2022)
• Theme 21 – Science Under Pressure (2021)
• Theme 20 – Data, Project and Lab Management, and Communication (2020)
• Theme 19 – Civility, Harassment and Inappropriate Conduct (2019)
• Theme 18 – Implicit and Explicit Biases in the Research Setting (2018)
• Theme 17 – Socially Responsible Science (2017)
• Theme 16 – Research Reproducibility (2016)
• Theme 15 – Authorship and Collaborative Science (2015)
• Theme 14 – Differentiating Between Honest Discourse and Research Misconduct and Introduction to Enhancing Reproducibility (2014)
• Theme 13 – Data Management, Whistleblowers, and Nepotism (2013)
• Theme 12 – Mentoring (2012)
• Theme 11 – Authorship (2011)
• Theme 10 – Science and Social Responsibility, continued (2010)
• Theme 9 – Science and Social Responsibility - Dual Use Research (2009)
• Theme 8 – Borrowing - Is It Plagiarism? (2008)
• Theme 7 – Data Management and Scientific Misconduct (2007)
• Theme 6 – Ethical Ambiguities (2006)
• Theme 5 – Data Management (2005)
• Theme 4 – Collaborative Science (2004)
• Theme 3 – Mentoring (2003)
• Theme 2 – Authorship (2002)
• Theme 1 – Scientific Misconduct (2001)

For Facilitators Leading Case Discussion

For the sake of time and clarity of purpose, it is essential that one individual have responsibility for leading the group discussion. As a minimum, this responsibility should include:

• Reading the case aloud.
• Defining, and re-defining as needed, the questions to be answered.
• Encouraging discussion that is “on topic”.
• Discouraging discussion that is “off topic”.
• Keeping the pace of discussion appropriate to the time available.
• Eliciting contributions from all members of the discussion group.
• Summarizing both majority and minority opinions at the end of the discussion.

How Should Cases be Analyzed?

Many of the skills necessary to analyze case studies can become tools for responding to real world problems. Cases, like the real world, contain uncertainties and ambiguities. Readers are encouraged to identify key issues, make assumptions as needed, and articulate options for resolution. In addition to the specific questions accompanying each case, readers should consider the following questions:

• Who are the affected parties (individuals, institutions, a field, society) in this situation?
• What interest(s) (material, financial, ethical, other) does each party have in the situation? Which interests are in conflict?
• Were the actions taken by each of the affected parties acceptable (ethical, legal, moral, or common sense)? If not, are there circumstances under which those actions would have been acceptable? Who should impose what sanction(s)?
• What other courses of action are open to each of the affected parties? What is the likely outcome of each course of action?
• For each party involved, what course of action would you take, and why?
• What actions could have been taken to avoid the conflict?

Is There a Right Answer?

Acceptable solutions.

Most problems will have several acceptable solutions or answers, but it will not always be the case that a perfect solution can be found. At times, even the best solution will still have some unsatisfactory consequences.

Unacceptable Solutions

While more than one acceptable solution may be possible, not all solutions are acceptable. For example, obvious violations of specific rules and regulations or of generally accepted standards of conduct would typically be unacceptable. However, it is also plausible that blind adherence to accepted rules or standards would sometimes be an unacceptable course of action.

Ethical Decision-Making

It should be noted that ethical decision-making is a process rather than a specific correct answer. In this sense, unethical behavior is defined by a failure to engage in the process of ethical decision-making. It is always unacceptable to have made no reasonable attempt to define a consistent and defensible basis for conduct.

This page was last updated on Friday, July 7, 2023

• Browse All Articles
• Newsletter Sign-Up

• 30 Apr 2024

When Managers Set Unrealistic Expectations, Employees Cut Ethical Corners

Corporate misconduct has grown in the past 30 years, with losses often totaling billions of dollars. What businesses may not realize is that misconduct often results from managers who set unrealistic expectations, leading decent people to take unethical shortcuts, says Lynn S. Paine.

• 23 Apr 2024
• Cold Call Podcast

Amazon in Seattle: The Role of Business in Causing and Solving a Housing Crisis

In 2020, Amazon partnered with a nonprofit called Mary’s Place and used some of its own resources to build a shelter for women and families experiencing homelessness on its campus in Seattle. Yet critics argued that Amazon’s apparent charity was misplaced and that the company was actually making the problem worse. Paul Healy and Debora Spar explore the role business plays in addressing unhoused communities in the case “Hitting Home: Amazon and Mary’s Place.”

• 15 Apr 2024

Struggling With a Big Management Decision? Start by Asking What Really Matters

Leaders must face hard choices, from cutting a budget to adopting a strategy to grow. To make the right call, they should start by following their own “true moral compass,” says Joseph Badaracco.

• 26 Mar 2024

How Do Great Leaders Overcome Adversity?

In the spring of 2021, Raymond Jefferson (MBA 2000) applied for a job in President Joseph Biden’s administration. Ten years earlier, false allegations were used to force him to resign from his prior US government position as assistant secretary of labor for veterans’ employment and training in the Department of Labor. Two employees had accused him of ethical violations in hiring and procurement decisions, including pressuring subordinates into extending contracts to his alleged personal associates. The Deputy Secretary of Labor gave Jefferson four hours to resign or be terminated. Jefferson filed a federal lawsuit against the US government to clear his name, which he pursued for eight years at the expense of his entire life savings. Why, after such a traumatic and debilitating experience, would Jefferson want to pursue a career in government again? Harvard Business School Senior Lecturer Anthony Mayo explores Jefferson’s personal and professional journey from upstate New York to West Point to the Obama administration, how he faced adversity at several junctures in his life, and how resilience and vulnerability shaped his leadership style in the case, "Raymond Jefferson: Trial by Fire."

• 02 Jan 2024

Should Businesses Take a Stand on Societal Issues?

Should businesses take a stand for or against particular societal issues? And how should leaders determine when and how to engage on these sensitive matters? Harvard Business School Senior Lecturer Hubert Joly, who led the electronics retailer Best Buy for almost a decade, discusses examples of corporate leaders who had to determine whether and how to engage with humanitarian crises, geopolitical conflict, racial justice, climate change, and more in the case, “Deciding When to Engage on Societal Issues.”

• 12 Dec 2023

Can Sustainability Drive Innovation at Ferrari?

When Ferrari, the Italian luxury sports car manufacturer, committed to achieving carbon neutrality and to electrifying a large part of its car fleet, investors and employees applauded the new strategy. But among the company’s suppliers, the reaction was mixed. Many were nervous about how this shift would affect their bottom lines. Professor Raffaella Sadun and Ferrari CEO Benedetto Vigna discuss how Ferrari collaborated with suppliers to work toward achieving the company’s goal. They also explore how sustainability can be a catalyst for innovation in the case, “Ferrari: Shifting to Carbon Neutrality.” This episode was recorded live December 4, 2023 in front of a remote studio audience in the Live Online Classroom at Harvard Business School.

• 11 Dec 2023
• Research & Ideas

Doing Well by Doing Good? One Industry’s Struggle to Balance Values and Profits

Few companies wrestle with their moral mission and financial goals like those in journalism. Research by Lakshmi Ramarajan explores how a disrupted industry upholds its values even as the bottom line is at stake.

• 27 Nov 2023

Voting Democrat or Republican? The Critical Childhood Influence That's Tough to Shake

Candidates might fixate on red, blue, or swing states, but the neighborhoods where voters spend their teen years play a key role in shaping their political outlook, says research by Vincent Pons. What do the findings mean for the upcoming US elections?

• 21 Nov 2023

The Beauty Industry: Products for a Healthy Glow or a Compact for Harm?

Many cosmetics and skincare companies present an image of social consciousness and transformative potential, while profiting from insecurity and excluding broad swaths of people. Geoffrey Jones examines the unsightly reality of the beauty industry.

• 09 Nov 2023

What Will It Take to Confront the Invisible Mental Health Crisis in Business?

The pressure to do more, to be more, is fueling its own silent epidemic. Lauren Cohen discusses the common misperceptions that get in the way of supporting employees' well-being, drawing on case studies about people who have been deeply affected by mental illness.

• 07 Nov 2023

How Should Meta Be Governed for the Good of Society?

Julie Owono is executive director of Internet Sans Frontières and a member of the Oversight Board, an outside entity with the authority to make binding decisions on tricky moderation questions for Meta’s companies, including Facebook and Instagram. Harvard Business School visiting professor Jesse Shapiro and Owono break down how the Board governs Meta’s social and political power to ensure that it’s used responsibly, and discuss the Board’s impact, as an alternative to government regulation, in the case, “Independent Governance of Meta’s Social Spaces: The Oversight Board.”

• 24 Oct 2023

From P.T. Barnum to Mary Kay: Lessons From 5 Leaders Who Changed the World

What do Steve Jobs and Sarah Breedlove have in common? Through a series of case studies, Robert Simons explores the unique qualities of visionary leaders and what today's managers can learn from their journeys.

• 03 Oct 2023
• Research Event

Build the Life You Want: Arthur Brooks and Oprah Winfrey Share Happiness Tips

"Happiness is not a destination. It's a direction." In this video, Arthur C. Brooks and Oprah Winfrey reflect on mistakes, emotions, and contentment, sharing lessons from their new book.

• 12 Sep 2023

Successful, But Still Feel Empty? A Happiness Scholar and Oprah Have Advice for You

So many executives spend decades reaching the pinnacles of their careers only to find themselves unfulfilled at the top. In the book Build the Life You Want, Arthur Brooks and Oprah Winfrey offer high achievers a guide to becoming better leaders—of their lives.

• 10 Jul 2023
• In Practice

The Harvard Business School Faculty Summer Reader 2023

Need a book recommendation for your summer vacation? HBS faculty members share their reading lists, which include titles that explore spirituality, design, suspense, and more.

• 01 Jun 2023

A Nike Executive Hid His Criminal Past to Turn His Life Around. What If He Didn't Have To?

Larry Miller committed murder as a teenager, but earned a college degree while serving time and set out to start a new life. Still, he had to conceal his record to get a job that would ultimately take him to the heights of sports marketing. A case study by Francesca Gino, Hise Gibson, and Frances Frei shows the barriers that formerly incarcerated Black men are up against and the potential talent they could bring to business.

• 04 Apr 2023

Two Centuries of Business Leaders Who Took a Stand on Social Issues

Executives going back to George Cadbury and J. N. Tata have been trying to improve life for their workers and communities, according to the book Deeply Responsible Business: A Global History of Values-Driven Leadership by Geoffrey Jones. He highlights three practices that deeply responsible companies share.

• 14 Mar 2023

Can AI and Machine Learning Help Park Rangers Prevent Poaching?

Globally there are too few park rangers to prevent the illegal trade of wildlife across borders, or poaching. In response, Spatial Monitoring and Reporting Tool (SMART) was created by a coalition of conservation organizations to take historical data and create geospatial mapping tools that enable more efficient deployment of rangers. SMART had demonstrated significant improvements in patrol coverage, with some observed reductions in poaching. Then a new predictive analytic tool, the Protection Assistant for Wildlife Security (PAWS), was created to use artificial intelligence (AI) and machine learning (ML) to try to predict where poachers would be likely to strike. Jonathan Palmer, Executive Director of Conservation Technology for the Wildlife Conservation Society, already had a good data analytics tool to help park rangers manage their patrols. Would adding an AI- and ML-based tool improve outcomes or introduce new problems? Harvard Business School senior lecturer Brian Trelstad discusses the importance of focusing on the use case when determining the value of adding a complex technology solution in his case, “SMART: AI and Machine Learning for Wildlife Conservation.”

• 14 Feb 2023

Does It Pay to Be a Whistleblower?

In 2013, soon after the US Securities and Exchange Commission (SEC) had started a massive whistleblowing program with the potential for large monetary rewards, two employees of a US bank’s asset management business debated whether to blow the whistle on their employer after completing an internal review that revealed undisclosed conflicts of interest. The bank’s asset management business disproportionately invested clients’ money in its own mutual funds over funds managed by other banks, letting it collect additional fees—and the bank had not disclosed this conflict of interest to clients. Both employees agreed that failing to disclose the conflict was a problem, but beyond that, they saw the situation very differently. One employee, Neel, perceived the internal review as a good-faith effort by senior management to identify and address the problem. The other, Akash, thought that the entire business model was problematic, even with a disclosure, and believed that the bank may have even broken the law. Should they escalate the issue internally or report their findings to the US Securities and Exchange Commission? Harvard Business School associate professor Jonas Heese discusses the potential risks and rewards of whistleblowing in his case, “Conflicts of Interest at Uptown Bank.”

• 17 Jan 2023

Good Companies Commit Crimes, But Great Leaders Can Prevent Them

It's time for leaders to go beyond "check the box" compliance programs. Through corporate cases involving Walmart, Wells Fargo, and others, Eugene Soltes explores the thorny legal issues executives today must navigate in his book Corporate Criminal Investigations and Prosecutions.

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• HHS Author Manuscripts

Introduction: Case Studies in the Ethics of Mental Health Research

Joseph millum.

Clinical Center Department of Bioethics/Fogarty International Center, National Institutes of Health, Bethesda, MD

This collection presents six case studies on the ethics of mental health research, written by scientific researchers and ethicists from around the world. We publish them here as a resource for teachers of research ethics and as a contribution to several ongoing ethical debates. Each consists of a description of a research study that was proposed or carried out and an in-depth analysis of the ethics of the study.

Building Global Capacity in Mental Health Research

According to the World Health Organization (WHO), there are more than 450 million people with mental, neurological, or behavioral problems worldwide ( WHO, 2005a ). Mental health problems are estimated to account for 13% of the global burden of disease, principally from unipolar and bipolar depression, alcohol and substance-use disorders, schizophrenia, and dementia. Nevertheless, in many countries, mental health is accorded a low priority; for example, a 2005 WHO analysis found that nearly a third of low-income countries who reported a mental health budget spent less than 1% of their total health budget on mental health ( WHO, 2005b ).

Despite the high burden of disease and some partially effective treatments that can be implemented in countries with weaker healthcare delivery systems ( Hyman et al., 2006 ), there exist substantial gaps in our knowledge of how to treat most mental health conditions. A 2007 Lancet Series entitled Global Mental Health claimed that the “rudimentary level of mental health-service research programmes in many nations also contributes to poor delivery of mental health care” ( Jacob et al., 2007 ). Its recommendations for mental health research priorities included research into the effects of interactions between mental health and other health conditions ( Prince et al., 2007 ), interventions for childhood developmental disabilities ( Patel et al., 2007 ), cost-effectiveness analysis, the scaling up of effective interventions, and the development of interventions that can be delivered by nonspecialist health workers ( Lancet Global Mental Health Group, 2007 ). All of these priorities require research in environments where the prevailing health problems and healthcare services match those of the populations the research will benefit, which suggests that research must take place all around the world. Similarly, many of the priorities identified by the Grand Challenges in Mental Health Initiative require focus on local environments, cultural factors, and the health systems of low- and middle-income countries. All the challenges “emphasize the need for global cooperation in the conduct of research” ( Collins et al., 2011 ).

Notwithstanding the need for research that is sensitive to different social and economic contexts, the trend of outsourcing to medical research to developing countries shows no sign of abating ( Thiers et al., 2008 ). Consequently, a substantial amount of mental health research will, in any case, take place in low- and middle-income countries, as well as rich countries, during the next few years.

The need for local research and the continuing increase in the international outsourcing of research imply that there is a pressing need to build the capacity to conduct good quality mental health research around the world. However, the expansion of worldwide capacity to conduct mental health research requires more than simply addressing low levels of funding for researchers and the imbalance between the resources available in rich and poor countries. People with mental health disorders are often thought to be particularly vulnerable subjects. This may be a product of problems related to their condition, such as where the condition reduces the capacity to make autonomous decisions. It may also result from social conditions because people with mental disorders are disproportionately likely to be poor, are frequently stigmatized as a result of their condition, and may be victims of human rights abuses ( Weiss et al., 2001 ; WHO, 2005a ). As a result, it is vitally important that the institutional resources and expertise are in place for ensuring that this research is carried out ethically.

Discussion at a special session at the 7th Global Forum on Bioethics in Research revealed the perception that many mental health researchers are not very interested in ethics and showed up a lack of ethics resources directly related to their work. This collection of case studies in the ethics of mental health research responds to that gap.

This collection comprises six case studies written by contributors from around the world ( Table 1 ). Each describes a mental health research study that raised difficult ethical issues, provides background and analysis of those issues, and draws conclusions about the ethics of the study, including whether it was ethical as it stood and how it ought to be amended otherwise. Three of the case studies are written by scientists who took part in the research they analyzed. For these cases, we have asked scholars independent of the research to write short commentaries on them. It is valuable to hear how the researchers themselves grapple with the ethical issues they encounter, as well as to hear the views of people with more distance from the research enterprise. Some of the ethical issues raised here have not been discussed before in the bioethics literature; others are more common concerns that have not received much attention in the context of international research. The case studies are intended to both expand academic discussion of some of the key questions related to research into mental health and for use in teaching ethics.

Case studies are an established teaching tool. Ethical analyses of such cases demonstrate the relevance of ethics to the actual practice of medical research and provide paradigmatic illustrations of the application of ethical principles to particular research situations. Concrete cases help generate and guide discussion and assist students who have trouble dealing with ethical concepts in abstraction. Through structured discussion, ethical development and decision-making skills can be enhanced. Moreover, outside of the teaching context, case study analyses provide a means to generate and focus debate on the relevant ethical issues, which can both highlight their importance and help academic discussion to advance.

People working in mental health research can benefit most from case studies that are specific to mental health. Even though, as outlined below, many of the same ethical problems arise in mental health research as elsewhere, the details of how they arise are important. For example, the nature of depression and the variation in effectiveness of antidepressive medication make a difference to how we should assess the ethics of placebo-controlled trials for new antidepressants. Moreover, seeing how familiar ethical principles are applied to one's own research specialty makes it easier to think about the ethics of one's own research. The cases in this collection highlight the commonalities and the variation in the ethical issues facing researchers in mental health around the world.

The current literature contains some other collections of ethics case studies that may be useful to mental health researchers. I note four important collections here, to which interested scholars may want to refer. Lavery et al.'s (2007) Ethical Issues in International Bio-medical Research provides in-depth analyses of ethically problematic research, mostly in low- and middle-income countries, although none of these cases involve mental health. Cash et al.'s (2009) Casebook on Ethical Issues in International Health Research also focuses on research in low- and middle-income countries, and several of the 64 short case descriptions focus on populations with mental health problems. Two further collections focus on mental health research, in particular. Dubois (2007) and colleagues developed short and longer US-based case studies for teaching as part of their “Ethics in Mental Health Research” training course. Finally, Hoagwood et al.'s (1996) book Ethical Issues in Mental Health Research with Children and Adolescents contains a casebook of 61 short case descriptions, including a few from outside the United States and Western Europe. For teachers and academics in search of more case studies, these existing collections should be very useful. Here, we expand on the available resources with six case studies from around the world with extended ethical analyses.

The remainder of this introduction provides an overview of some of the most important ethical issues that arise in mental health research and describes some of the more significant ethics guidance documents that apply.

Ethical Issues in Mental Health Research

The same principles can be applied in assessing the ethics of mental health research as to other research using human participants ( Emanuel et al., 2000 ). Concerns about the social value of research, risks, informed consent, and the fair treatment of participants all still apply. This means that we can learn from the work done in other areas of human subjects research. However, specific research contexts make a difference to how the more general ethical principles should be applied to them. Different medical conditions may require distinctive research designs, different patient populations may need special protections, and different locations may require researchers to respond to study populations who are very poor and lack access to health care or to significant variations in regulatory systems. The ethical analysis of international mental health research therefore needs to be tailored to its particularities.

Each case study in this collection focuses on the particular ethical issues that are relevant to the research it analyzes. Nevertheless, some issues arise in multiple cases. For example, questions about informed consent arise in the context of research with stroke patients, with students, and with other vulnerable groups. To help the reader compare the treatment of an ethical issue across the different case studies, the ethical analyses use the same nine headings to delineate the issues they consider. These are social value, study design, study population, informed consent, risks and benefits, confidentiality, post-trial obligations, legal versus ethical obligations, and oversight.

Here, I focus on five of these ethical issues as they arise in the context of international mental health research: (1) study design, (2) study population, (3) risks and benefits, (4) informed consent, and (5) post-trial obligations. I close by mentioning some of the most important guidelines that pertain to mental health research.

Study Design

The scientific design of a research study determines what sort of data it can generate. For example, the decision about what to give participants in each arm of a controlled trial determines what interventions the trial compares and what questions about relative safety and efficacy it can answer. What data a study generates makes a difference to the ethics of the study because research that puts human beings at risk is ethically justified in terms of the social value of the knowledge it produces. It is widely believed that human subject research without any social value is unethical and that the greater the research risks to participants, the greater the social value of the research must be to compensate ( Council for International Organizations of Medical Sciences [CIOMS], 2002 ; World Medical Association, 2008 ). However, changing the scientific design of a study frequently changes what happens to research participants, too. For example, giving a control group in a treatment trial an existing effective treatment rather than placebo makes it more likely that their condition will improve but may expose them to adverse effects they would not otherwise experience. Therefore, questions of scientific design can be ethically very complex because different possible designs are compared both in terms of the useful knowledge they may generate and their potential impact on participants.

One of the more controversial questions of scientific design concerns the standard of care that is offered to participants in controlled trials. Some commentators argue that research that tests therapeutic interventions is only permissible if there is equipoise concerning the relative merits of the treatments being compared, that is, there are not good reasons to think that participants in any arm of the trial are receiving inferior treatment ( Joffe and Truog, 2008 ). If there is not equipoise, the argument goes, then physician-researchers will be breaching their duty to give their patients the best possible care ( Freedman, 1987 ).

The Bucharest Early Intervention Project (BEIP) described in the case study by Charles Zeanah was a randomized controlled trial comparing foster care with institutional care in Bucharest, Romania. When designing the BEIP, the researchers wrestled with the issue of whether there was genuine equipoise regarding the relative merits of institutional and foster care. One interpretation of equipoise is that it exists when the professional community has not reached consensus about the better treatment ( Freedman, 1987 ). Childcare professionals in the United States were confident that foster care was superior, but there was no such confidence in Romania, where institutional care was the norm. Which, then, was the relevant professional community?

The equipoise requirement is justified by reference to the role morality of physicians: for a physician to give her patient treatment that she knows to be inferior would violate principles of therapeutic beneficence and nonmaleficence. As a result, the equipoise requirement has been criticized for conflating the ethics of the physician-patient relationship with the ethics of the researcher-participant relationship ( Miller and Brody, 2003 ). According to Miller and Brody (2003) , provided that other ethical requirements are met, including an honest null hypothesis, it is not unethical to assign participants to receive treatment regimens known to be inferior to the existing standard of care.

A subset of trial designs that violate equipoise are placebo-controlled trials of experimental treatments for conditions for which proven effective treatments already exist. Here, there is not equipoise because some participants will be assigned to placebo treatment, and ex hypothesi there already exists treatment that is superior to placebo. Even if we accept Miller and Brody's (2003) argument and reject the equipoise requirement, there remain concerns about these placebo-controlled trials. Providing participants with less effective treatment than they could get outside of the trial constitutes a research risk because trial participation makes them worse off. Moreover, on the face of it, a placebo-controlled trial of a novel treatment of a condition will not answer the most important scientific question about the treatment that clinicians are interested in: is this new treatment better than the old one? Consequently, in situations where there already exists a standard treatment of a condition, it has generally been considered unethical to use a placebo control when testing a new treatment, rather than using the standard treatment as an active-control ( World Medical Association, 2008 ).

Some psychiatric research provides scientific reasons to question a blanket prohibition on placebo-controlled trials when an effective intervention exists. For example, it is not unusual for antidepressive drugs to fail to show superiority to placebo in any given trial. This means that active-control trials may seem to show that an experimental drug is equivalent in effectiveness to the current standard treatment, when the explanation for their equivalence may, in fact, be that neither was better than placebo. Increasing the power of an active-control trial sufficiently to rule out this possibility may require an impractically large number of subjects and will, in any case, put a greater number of subjects at risk ( Carpenter et al., 2003 ; Miller, 2000 ). A 2005 trial of risperidone for acute mania conducted in India ( Khanna et al., 2005 ) was criticized for unnecessarily exposing subjects to risk ( Basil et al., 2006 ; Murtagh and Murphy, 2006 ; Srinivasan et al., 2006 ). The investigators' response to criticisms adopted exactly the line of argument just described:

A placebo group was included because patients with mania generally show a high and variable placebo response, making it difficult to identify their responses to an active medication. Placebo-controlled trials are valuable in that they expose the fewest patients to potentially ineffective treatments. In addition, inclusion of a placebo arm allows a valid evaluation of adverse events attributable to treatment v. those independent of treatment. ( Khanna et al., 2006 )

Concerns about the standard of care given to research participants are exacerbated in trials in developing countries, like India, where research participants may not have access to treatment independent of the study. In such cases, potential participants may have no real choice but to join a placebo-controlled trial, for example, because that is the only way they have a chance to receive treatment. In the Indian risperidone trial, the issue of exploitation is particularly stark because it seemed to some that participants were getting less than the international best standard of care, in order that a pharmaceutical company could gather data that was unlikely to benefit many Indian patients.

This is just one way in which trial design may present ethically troubling risks to participants. Other potentially difficult designs include washout studies, in which participants discontinue use of their medication, and challenge studies, in which psychiatric symptoms are experimentally induced ( Miller and Rosenstein, 1997 ). In both cases, the welfare of participants may seem to be endangered ( Zipursky, 1999 ). A variant on the standard placebo-controlled trial design is the withdrawal design, in which everyone starts the trial on medication, the people who respond to the medication are then selected for randomization, and then half of those people are randomized to placebo. This design was used by a Japanese research team to assess the effectiveness of sertraline for depression, as described by Shimon Tashiro and colleagues in this collection. The researchers regarded this design as more likely to benefit the participants because for legal reasons, sertraline was being tested in Japan despite its proven effectiveness in non-Japanese populations. Tashiro and colleagues analyze how the risks and benefits of a withdrawal design compare with those of standard placebo-controlled trials and consider whether the special regulatory context of Japan makes a difference.

Study Population

The choice of study population implicates considerations of justice. The Belmont Report, which lays out the ethical foundations for the United States system for ethical review of human subject research, says:

Individual justice in the selection of subjects would require that researchers … should not offer potentially beneficial research only to some patients who are in their favor or select only “undesirable” persons for risky research. Social justice requires that distinction be drawn between classes of subjects that ought, and ought not, to participate in any particular kind of research, based on the ability of members of that class to bear burdens and on the appropriateness of placing further burdens on already burdened persons. ( National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, 1978 )

Two distinct considerations are highlighted here. The first (“individual justice”) requires that the researchers treat people equally. Morally irrelevant differences between people should not be the basis for deciding whom to enroll in research. For example, it would normally be unjust to exclude women from a phase 3 trial of a novel treatment of early-stage Alzheimer disease, given that they are an affected group. Some differences are not morally irrelevant, however. In particular, there may be scientific reasons for choosing one possible research population over another, and there may be risk-related reasons for excluding certain groups. For example, a functional magnetic resonance imaging study in healthy volunteers to examine the acute effects of an antianxiety medication might reasonably exclude left-handed people because their brain structure is different from that of right-handed people, and a study of mood that required participants to forego medication could justifiably exclude people with severe depression or suicidal ideation.

The second consideration requires that we consider how the research is likely to impact “social justice.” Social justice refers to the way in which social institutions distribute goods, like property, education, and health care. This may apply to justice within a state ( Rawls, 1971 ) or to global justice ( Beitz, 1973 ). In general, research will negatively affect social justice when it increases inequality, for example, by making people who are already badly off even worse off. The quotation from the Belmont Report above suggests one way in which research might violate a requirement of social justice: people who are already badly off might be asked to participate in research and so be made worse off. For example, a study examining changes in the brain caused by alcohol abuse that primarily enrolled homeless alcoholics from a shelter near the study clinic might only put at further risk this group who are already very badly off. An alternative way in which research can promote justice or injustice is through its results. Research that leads to the development of expensive new attention deficit hyperactivity disorder medication is likely to do little, if anything, to make the world more just. Research on how to improve the cognitive development of orphaned children in poor environments (like the BEIP) is much more likely to improve social justice.

This last point suggests a further concern about fairness—exploitation—that frequently arises in the context of international collaborative research in developing countries. Exploitation occurs, roughly, when one party takes “unfair advantage” of the vulnerability of another. This means that the first party benefits from the interaction and does so to an unfair extent ( Wertheimer, 1996 ). These conditions may be met in international collaborative research when the burdens of research fall disproportionately on people and institutions in developing countries, but the benefits of research, such as access to new treatments, accrue to people in richer countries. A number of case studies in this collection raise this concern in one way or another. For example, Virginia Rodriguez analyzes a proposed study of the genetic basis of antisocial personality disorder run by US researchers but carried out at sites in several Latin American countries. One of the central objections raised by one of the local national research ethics committees with regard to this study was that there appeared to be few, if any, benefits for patients and researchers in the host country.

Risks and Benefits

Almost all research poses some risk of harm to participants. Participants in mental health research may be particularly susceptible to risk in several ways. First, and most obviously, they may be physically or psychologically harmed as a result of trial participation. For example, an intervention study of an experimental antipsychotic may result in some serious adverse effects for participants who take the drug. Less obvious but still very important are the potential effects of stopping medication. As mentioned above, some trials of psychoactive medications require that patients stop taking the medications that they were on before the trial ( e.g ., the Japanese withdrawal trial). Stopping their medication can lead to relapse, to dangerous behavior (like attempted suicide), and could mean that their previous treatment regimen is less successful when they attempt to return to it. Participants who were successfully treated during a trial may have similar effects if they do not have access to treatment outside of the trial. This is much more likely to happen in research conducted with poor populations, such as the Indian mania patients.

The harms resulting directly from research-related interventions are not the only risk to participants in mental health research. Participation can also increase the risks of psychosocial harms, such as being identified by one's family or community as having a particular condition. Such breaches of confidentiality need not involve gross negligence on the part of researchers. The mere fact that someone regularly attends a clinic or sees a psychiatrist could be sufficient to suggest that they have a mental illness. In other research, the design makes confidentiality hard to maintain. For example, the genetic research described by Rodriguez involved soliciting the enrollment of the family members of people with antisocial personality disorder.

The harm from a breach of confidentiality is exacerbated when the condition studied or the study population is stigmatized. Both of these were true in the case Sana Loue describes in this collection. She studied the co-occurrence of severe mental illnesses and human immunodeficiency virus risk in African-American men who have sex with men. Not only was there shame attached to the conditions under study, such that they were euphemistically described in the advertisements for the research, but also many of the participants were men who had heterosexual public identities.

Informed Consent

Many people with mental disorders retain the capacity (ability) and competence (legal status) to give informed consent. Conversely, potential participants without mental problems may lack or lose capacity (and competence). Nevertheless, problems with the ability to consent remain particularly pressing with regard to mental health research. This is partly a consequence of psychological conditions that reduce or remove the ability to give informed consent. To study these conditions, it may be necessary to use participants who have them, which means that alternative participants who can consent are, in principle, not available. This occurred in the study of South African stroke patients described by Anne Pope in this collection. The researcher she describes wanted to compare the effectiveness of exercises designed to help patients whose ability to communicate was compromised by their stroke. Given their communication difficulties and the underlying condition, there would inevitably be questions about their capacity. Whether it is permissible to enroll people who cannot give informed consent into a study depends on several factors, including the availability of alternative study populations, the levels of risk involved, and the possible benefits to participants in comparison with alternative health care they could receive.

In research that expects to enroll people with questionable capacity to consent, it is wise to institute procedures for assessing the capacity of prospective participants. There are two general strategies for making these assessments. The first is to conduct tests that measure the general cognitive abilities of the person being assessed, as an IQ test does. If she has the ability to perform these sorts of mental operations sufficiently well, it is assumed that she also has the ability to make autonomous decisions about research participation. A Mini-Mental State Examination might be used to make this sort of assessment ( Kim and Caine, 2002 ). The second capacity assessment strategy focuses on a prospective participant's understanding and reasoning with regard to the specific research project they are deciding about. If she understands that project and what it implies for her and is capable of articulating her reasoning about it, then it is clear that she is capable of consenting to participation, independent of her more general capacities. This sort of assessment requires questions that are tailored to each specific research project and cannot be properly carried out unless the assessor is familiar with that research.

Where someone lacks the capacity to give consent, sometimes a proxy decision maker can agree to trial participation on her behalf. In general, proxy consent is not equivalent to individual consent: unless the proxy was expressly designated to make research decisions by the patient while capacitated, the proxy lacks the power to exercise the patient's rights. As a result, the enrollment of people who lack capacity is only acceptable when the research poses a low net risk to participants or holds out the prospect of benefiting them. When someone has not designated a proxy decision maker for research, it is common to allow the person who has the power to make decisions about her medical care also to make decisions about research participation. However, because medical care is directed at the benefit of the patient, but research generally is not aimed at the benefit of participants, the basis for this assumption is unclear. Its legal basis may be weak, too. For example, in her discussion of research on South African stroke patients, Pope notes the confusion surrounding the legality of surrogate decision makers, given that the South African constitution forbids proxy decision making for adults (unless they have court-appointed curators), but local and international guidance documents seem to assume it.

Although it is natural to think of the capacity to give consent as an all-or-nothing phenomenon, it may be better conceptualized as domain-specific. Someone may be able to make decisions about some areas of her life, but not others. This fits with assumptions that many people make in everyday life. For example, a 10-year-old child may be deemed capable of deciding what clothes she will wear but may not be capable of deciding whether to visit the dentist. The capacity to consent may admit of degrees in another way, too. Someone may have diminished capacity to consent but still be able to make decisions about their lives if given the appropriate assistance. For example, a patient with mild dementia might not be capable of deciding on his own whether he should move in with a caregiver, but his memory lapses during decision making could be compensated for by having his son present to remind him of details relevant to the decision. The concept of supported decision making has been much discussed in the literature on disability; however, its application to consent to research has received little attention ( Herr, 2003 ; United Nations, 2007 ).

The ability to give valid informed consent is the aspect of autonomy that is most frequently discussed in the context of mental health research, but it is not the only important aspect. Several of the case studies in this collection also raise issues of voluntariness and coercion. For example, Douglas Wassenaar and Nicole Mamotte describe a study in which professors enrolled their students, which raises the question of the vulnerability of student subjects to pressure. Here, there is both the possibility of explicit coercion and the possibility that students will feel pressure even from well-meaning researchers. For various reasons, including dependence on caregivers or healthcare professionals and the stigma of their conditions, people with mental illnesses can be particularly vulnerable to coercion.

Post-Trial Obligations

The obligations of health researchers extend past the end of their study. Participants'data remain in the hands of researchers after their active involvement in a study is over, and patients with chronic conditions who enroll in clinical trials may leave them still in need of treatment.

Ongoing confidentiality is particularly important when studying stigmatized populations (such as men who have sex with men as discussed by Sana Loue) or people with stigmatizing conditions (such as bipolar disorder). In research on mental illnesses, as with many medical conditions, it is now commonplace for researchers to collect biological specimens and phenotypic data from participants to use in future research (such as genome-wide association studies). Additional challenges with regard to confidentiality are raised by the collection of data and biological specimens for future research because confidentiality must then be guaranteed in a long period of time and frequently with different research groups making use of the samples.

Biobanking also generates some distinctive ethical problems of its own. One concerns how consent to the future use of biological specimens should be obtained. Can participants simply give away their samples for use in whatever future research may be proposed, or do they need to have some idea of what this research might involve in order to give valid consent? A second problem, which arises particularly in transnational research, concerns who should control the ongoing use of the biobank. Many researchers think that biological samples should not leave the country in which they were collected, and developing country researchers worry that they will not be allowed to do research on the biobanks that end up in developed countries. This was another key concern with the proposed study in Latin America.

In international collaborative research, further questions arise as a result of the disparities between developing country participants and researchers and developed country sponsors and researchers. For example, when clinical trials test novel therapies, should successful therapies be made available after the trial? If they should, who is responsible for ensuring their provision, to whom should they be provided, and in what does providing them consist? In the case of chronic mental illnesses like depression or bipolar disorder, patient-participants may need maintenance treatment for the rest of their lives and may be at risk if treatment is stopped. This suggests that the question of what happens to them after the trial must at least be considered by those who sponsor and conduct the trial and the regulatory bodies that oversee it. Exactly on whom obligations fall remains a matter of debate ( Millum, 2011 ).

Ethics Guidelines

A number of important policy documents are relevant to the ethics of research into mental disorders. The WMA's Declaration of Helsinki and the CIOMS' Ethical Guidelines for Biomedical Research both consider research on individuals whose capacity and/or competence to consent is impaired. They agree on three conditions: a) research on these people is justified only if it cannot be carried out on individuals who can give adequate informed consent, b) consent to such research should be obtained from a proxy representative, and c) the goal of such research should be the promotion of the health of the population that the research participants represent ( Council for International Organizations of Medical Sciences, 2002 ; World Medical Association, 2008 ). In addition, with regard to individuals who are incapable of giving consent, Guideline 9 of CIOMS states that interventions that do not “hold out the prospect of direct benefit for the individual subject” should generally involve no more risk than their “routine medical or psychological examination.”

In 1998, the US National Bioethics Advisory Commission (NBAC) published a report entitled Research Involving Persons with Mental Disorders That May Affect Decision-making Capacity ( National Bioethics Advisory Commission, 1998 ). As the title suggests, this report concentrates on issues related to the capacity or competence of research participants to give informed consent. Its recommendations are largely consistent with those made in the Declaration of Helsinki and CIOMS, although it is able to devote much more space to detailed policy questions (at least in the United States context). Two domains of more specific guidance are of particular interest. First, the NBAC report considers the conditions under which individuals who lack the capacity to consent may be enrolled in research posing different levels of risk and supplying different levels of expected benefits to participants. Second, it provides some analysis of who should be recognized as an appropriate proxy decision maker (or “legally authorized representative”) for participation in clinical trials.

Finally, the World Psychiatric Association's Madrid Declaration gives guidelines on the ethics of psychiatric practice. This declaration may have implications for what is permissible in psychiatric research, insofar as the duties of psychiatrists as personal physicians are also duties of psychiatrists as medical researchers. It also briefly considers the ethics of psychiatric research, although it notes only the special vulnerability of psychiatric patients as a concern distinctive of mental health research ( World Psychiatric Association, 2002 ).

The opinions expressed are the author's own. They do not reflect any position or policy of the National Institutes of Health, U.S. Public Health Service, or Department of Health and Human Services.

Disclosure : The author declares no conflict of interest.

• Basil B, Adetunji B, Mathews M, Budur K. Trial of risperidone in India—concerns. Br J Psychiatry. 2006; 188 :489–490. [ PubMed ] [ Google Scholar ]
• Beitz C. Political theory and international relations. Princeton, NJ: Princeton University Press; 1973. [ Google Scholar ]
• Carpenter WT, Appelbaum PS, Levine RJ. The Declaration of Helsinki and clinical trials: A focus on placebo-controlled trials in schizophrenia. Am J Psychiatry. 2003; 160 :356–362. [ PubMed ] [ Google Scholar ]
• Cash R, Wikler D, Saxena A, Capron A, editors. Casebook on ethical issues in international health research. Geneva, Switzerland: World Health Organization; 2009. Available at: http://whqlibdoc.who.int/publications/2009/9789241547727_eng.pdf . [ Google Scholar ]
• Collins PY, Patel V, Joestl SS, March D, Insel TR, Daar AS on behalf of the Scientific Advisory Board and the Executive Committee of the Grand Challenges on Global Mental Health. Grand challenges in global mental health. Nature. 2011; 475 :27–30. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Council for International Organizations of Medical Sciences. The international ethical guidelines for biomedical research involving human subjects. [Accessed on January 31, 2012]; 2002 Available at: http://www.cioms.ch/publications/layout_guide2002.pdf . [ PubMed ]
• DuBois JM. Ethical research in mental health. New York, NY: Oxford University Press; 2007. [ Google Scholar ]
• Emanuel EJ, Wendler D, Grady C. What makes clinical research ethical? JAMA. 2000; 283 :2701–2711. [ PubMed ] [ Google Scholar ]
• Freedman B. Equipoise and the ethics of clinical research. N Engl J Med. 1987; 317 :141–145. [ PubMed ] [ Google Scholar ]
• Herr SS. Self-determination, autonomy, and alternatives for guardianship. In: Herr SS, Gostin LO, Koh HH, editors. The human rights of persons with intellectual disabilities: Different but equal. Oxford, UK: Oxford University Press; 2003. pp. 429–450. [ Google Scholar ]
• Hoagwood K, Jensen PS, Fisher CB. Ethical issues in mental health research with children and adolescents. Mahwah, NJ: Lawrence Erlbaum Associates; 1996. [ Google Scholar ]
• Hyman S, Chisholm D, Kessler R, Patel V, Whiteford H. Chapter 31: Mental disorders. In: Jamison DT, Breman JG, Measham AR, Alleyne G, Claeson M, Evans DB, Jha P, Mills A, Musgrove P, editors. Disease control priorities in developing countries. 2nd. Washington, DC; New York, NY: Oxford University Press and the World Bank; 2006. [ Google Scholar ]
• Jacob KS, Sharan P, Mirza I, Garrido-Cumbrera M, Seedat S, Mari JJ, Sreenivas V, Saxena S. Global Mental Health 4: Mental health systems in countries: where are we now? Lancet. 2007; 370 :1061–1077. [ PubMed ] [ Google Scholar ]
• Joffe S, Truog RD. Equipoise and randomization. In: Emanuel EJ, Grady C, Crouch RA, Lie RK, Miller FG, Wendler D, editors. The oxford textbook of clinical research ethics. Oxford, UK: Oxford University Press; 2008. pp. 245–260. [ Google Scholar ]
• Khanna S, Vieta E, Lyons B, Grossman F, Eerdekens M, Kramer M. Risperidone in the treatment of acute mania: Double-blind, placebo-controlled study. Br J Psychiatry. 2005; 187 :229–234. [ PubMed ] [ Google Scholar ]
• Khanna S, Vieta E, Lyons B, Grossman F, Kramer M, Eerdekens M. Trial of risperidone in India—concerns. Authors' reply. Br J Psychiatry. 2006; 188 :491. [ Google Scholar ]
• Kim SYH, Caine ED. Utility and limits of the Mini Mental State Examination in evaluating consent capacity in Alzheimer's disease. Psychiatr Serv. 2002; 53 :1322–1324. [ PubMed ] [ Google Scholar ]
• Lancet Global Mental Health Group. Global Mental Health 6: Scale up services for mental disorders: a call for action. Lancet. 2007; 370 :1241–1252. [ PubMed ] [ Google Scholar ]
• Lavery J, Grady C, Wahl ER, Emanuel EJ, editors. Ethical issues in international biomedical research: A casebook. New York, NY: Oxford University Press; 2007. [ Google Scholar ]
• Miller FG, Rosenstein DL. Psychiatric symptom-provoking studies: An ethical appraisal. Biol Psychiatry. 1997; 42 :403–409. [ PubMed ] [ Google Scholar ]
• Miller FG. Placebo-controlled trials in psychiatric research: An ethical perspective. Biol Psychiatry. 2000; 47 :707–716. [ PubMed ] [ Google Scholar ]
• Miller FG, Brody H. A critique of clinical equipoise: Therapeutic misconception in the ethics of clinical trials. Hastings Cent Rep. 2003; 33 :19–28. [ PubMed ] [ Google Scholar ]
• Millum J. Post-trial access to antiretrovirals: Who owes what to whom? Bioethics. 2011; 25 :145–154. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Murtagh A, Murphy KC. Trial of risperidone in India—concerns. Br J Psychiatry. 2006; 188 :489. [ PubMed ] [ Google Scholar ]
• National Bioethics Advisory Commission. Research involving persons with mental disorders that may affect decision-making capacity. [Accessed January 31, 2012]; 1998 Available at: http://bioethics.georgetown.edu/nbac/capacity/toc.htm .
• National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. The Belmont Report: Ethical principles and guidelines for the protection of human subjects of research. Washington, DC: DHEW Publication; 1978. pp. 78–0012. No. (OS) [ PubMed ] [ Google Scholar ]
• Patel V, Araya R, Chatterjee S, Chisholm D, Cohen A, De Silva M, Hosman C, McGuire H, Rojas G, van Ommeren M. Global Mental Health 3: Treatment and prevention of mental disorders in low-income and middle-income countries. Lancet. 2007; 370 :991–1005. [ PubMed ] [ Google Scholar ]
• Prince M, Patel V, Saxena S, Maj M, Maselko J, Phillips MR, Rahman A. Global Mental Health 1: No health without mental health. Lancet. 2007; 370 :859–877. [ PubMed ] [ Google Scholar ]
• Rawls J. A theory of justice. Cambridge, MA: Harvard University Press; 1971. [ Google Scholar ]
• Thiers FA, Sinskey AJ, Berndt ER. Trends in the globalization of clinical trials. Nature Rev Drug Discov. 2008; 7 :13–14. [ Google Scholar ]
• Srinivasan S, Pai SA, Bhan A, Jesani A, Thomas G. Trial of risperidone in India—concerns. Br J Psychiatry. 2006; 188 :489. [ PubMed ] [ Google Scholar ]
• United Nations. Legal capacity and supported decision-making in from exclusion to equality: Realizing the rights of persons with disabilities. Geneva, Switzerland: United Nations; 2007. Available at: http://www.un.org/disabilities/default.asp?id=242 . [ Google Scholar ]
• Weiss MG, Jadhav S, Raguram R, Vounatsou P, Littlewood R. Psychiatric stigma across cultures: Local validation in Bangalore and London. Anthropol Med. 2001; 8 :71–87. [ Google Scholar ]
• Wertheimer A. Exploitation. Princeton, UK: Princeton University Press; 1996. [ Google Scholar ]
• World Health Organization. Glaring inequalities for people with mental disorders addressed in new WHO effort. [Accessed January 31, 2012]; 2005a Available at: http://www.who.int/mediacentre/news/notes/2005/np14/en/index.html .
• World Health Organization. Mental Health Atlas: 2005. [Accessed January 31, 2012]; 2005b Available at: http://www.who.int/mental_health/evidence/mhatlas05/en/index.html .
• World Medical Association. The Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects. [Accessed January 31, 2012]; 2008 Available at: http://www.wma.net/en/30publications/10policies/b3/index.html .
• World Psychiatric Association. Madrid Declaration on Ethical Standards for Psychiatric Practice. [Accessed January 31, 2012]; 2002 Available at: http://www.wpanet.org/detail.php?section_id=5&content_id=48 .
• Zipursky RB. Ethical issues in schizophrenia research. Curr Psychiatry Rep. 1999; 1 :13–19. [ PubMed ] [ Google Scholar ]

Princeton Dialogues on AI and Ethics

Princeton University

Case Studies

Princeton Dialogues on AI and Ethics Case Studies

The development of artificial intelligence (AI) systems and their deployment in society gives rise to ethical dilemmas and hard questions. By situating ethical considerations in terms of real-world scenarios, case studies facilitate in-depth and multi-faceted explorations of complex philosophical questions about what is right, good and feasible. Case studies provide a useful jumping-off point for considering the various moral and practical trade-offs inherent in the study of practical ethics.

Case Study PDFs : The Princeton Dialogues on AI and Ethics has released six long-format case studies exploring issues at the intersection of AI, ethics and society. Three additional case studies are scheduled for release in spring 2019.

Methodology : The Princeton Dialogues on AI and Ethics case studies are unique in their adherence to five guiding principles: 1) empirical foundations, 2) broad accessibility, 3) interactiveness, 4) multiple viewpoints and 5) depth over brevity.

Research and Practice of AI Ethics: A Case Study Approach Juxtaposing Academic Discourse with Organisational Reality

• Original Research/Scholarship
• Open access
• Published: 08 March 2021
• Volume 27 , article number  16 , ( 2021 )

Cite this article

You have full access to this open access article

• Mark Ryan   ORCID: orcid.org/0000-0003-4850-0111 1 ,
• Josephina Antoniou 2 ,
• Laurence Brooks 3 ,
• Tilimbe Jiya 4 ,
• Kevin Macnish 5 &
• Bernd Stahl 3  

13k Accesses

21 Citations

7 Altmetric

Explore all metrics

This study investigates the ethical use of Big Data and Artificial Intelligence (AI) technologies (BD + AI)—using an empirical approach. The paper categorises the current literature and presents a multi-case study of 'on-the-ground' ethical issues that uses qualitative tools to analyse findings from ten targeted case-studies from a range of domains. The analysis coalesces identified singular ethical issues, (from the literature), into clusters to offer a comparison with the proposed classification in the literature. The results show that despite the variety of different social domains, fields, and applications of AI, there is overlap and correlation between the organisations’ ethical concerns. This more detailed understanding of ethics in AI + BD is required to ensure that the multitude of suggested ways of addressing them can be targeted and succeed in mitigating the pertinent ethical issues that are often discussed in the literature.

Similar content being viewed by others

Artificial Intelligence and Business Value: a Literature Review

The Ethics of AI Ethics: An Evaluation of Guidelines

The Ethical Implications of Artificial Intelligence (AI) For Meaningful Work

Avoid common mistakes on your manuscript.

Introduction

Big Data and Artificial Intelligence (BD + AI) are emerging technologies that offer great potential for business, healthcare, the public sector, and development agencies alike. The increasing impact of these two technologies and their combined potential in these sectors can be highlighted for diverse organisational aspects such as for customisation of organisational processes and for automated decision making. The combination of Big Data and AI, often in the form of machine learning applications, can better exploit the granularity of data and analyse it to offer better insights into behaviours, incidents, and risk, eventually aiming at positive organisational transformation.

Big Data offers fresh and interesting insights into structural patterns, anomalies, and decision-making in a broad range of different applications (Cuquet & Fensel, 2018 ), while AI provides predictive foresight, intelligent recommendations, and sophisticated modelling. The integration and combination of AI + BD offer phenomenal potential for correlating, predicting and prescribing recommendations in insurance, human resources (HR), agriculture, and energy, as well as many other sectors. While BD + AI provides a wide range of benefits, they also pose risks to users, including but not limited to privacy infringements, threats of unemployment, discrimination, security concerns, and increasing inequalities (O’Neil, 2016 ). Footnote 1 Adequate and timely policy needs to be implemented to prevent many of these risks occurring.

One of the main limitations preventing key decision-making for ethical BD + AI use is that there are few rigorous empirical studies carried out on the ethical implications of these technologies across multiple application domains. This renders it difficult for policymakers and developers to identify when ethical issues resulting from BD + AI use are only relevant for isolated domains and applications, or whether there are repeated/universal concerns which can be seen across different sectors. While the field lacks literature evaluating ethical issues Footnote 2 ‘on the ground’, there are even fewer multi-case evaluations.

This paper provides a cohesive multi-case study analysis across ten different application domains, including domains such as government, agriculture, insurance, and the media. It reviews ethical concerns found within these case studies to establish cross-cutting thematic issues arising from the implementation and use of BD + AI. The paper collects relevant literature and proposes a simple classification of ethical issues (short term, medium term, long term), which is then juxtaposed with the ethical concerns highlighted from the multiple-case study analysis. This multiple-case study analysis of BD + AI offers an understanding of current organisational practices.

The work described in this paper makes an important contribution to the literature, based on its empirical findings. By presenting the ethical issues across an array of application areas, the paper provides much-needed rigorous empirical insight into the social and organisational reality of ethics of AI + BD. Our empirical research brings together a collection of domains that gives a broad oversight about issues that underpin the implementation of AI. Through its empirical insights the paper provides a basis for a broader discussion of how these issues can and should be addressed.

This paper is structured in six main sections: this introduction is followed by a literature review, which allows for an integrated review of ethical issues, contrasting them with those found in the cases. This provides the basis for a categorisation or classification of ethical issues in BD + AI. The third section contains a description of the interpretivist qualitative case study methodology used in this paper. The subsequent section provides an overview of the organisations participating in the cases to contrast similarities and divisions, while also comparing the diversity of their use of BD + AI. Footnote 3 The fifth section provides a detailed analysis of the ethical issues derived from using BD + AI, as identified in the cases. The concluding section analyses the differences between theoretical and empirical work and spells out implications and further work.

Literature Review

An initial challenge that any researcher faces when investigating ethical issues of AI + BD is that, due to the popularity of the topic, there is a vast and rapidly growing literature to be considered. Ethical issues of AI + BD are covered by a number of academic venues, including some specific ones such as the AAAI/ACM Conference on AI, Ethics, and Society ( https://dl.acm.org/doi/proceedings/10.1145/3306618 ), policy initiative and many publicly and privately financed research reports (Whittlestone, Nyrup, Alexandrova, Dihal, & Cave, 2019 ). Initial attempts to provide overviews of the area have been published (Jobin, 2019 ; Mittelstadt, Allo, Taddeo, Wachter, & Floridi, 2016 ), but there is no settled view on what counts as an ethical issue and why. In this paper we aim to provide a broad overview of issues found through the case studies. This paper puts forward what are commonly perceived to be ethical issues within the literature or concerns that have ethical impacts and repercussions. We explicitly do not apply a particular philosophical framework of ethics but accept as ethical issues those issues that we encounter in the literature. This review is based on an understanding of the current state of the literature by the paper's authors. It is not a structured review and does not claim comprehensive coverage but does share some interesting insights.

To be able to undertake the analysis of ethical issues in our case studies, we sought to categorise the ethical issues found in the literature. There are potentially numerous ways of doing so and our suggestion does not claim to be authoritative. Our suggestion is to order ethical issues in terms of their temporal horizon, i.e., the amount of time it is likely to take to be able to address them. Time is a continuous variable, but we suggest that it is possible to sort the issues into three clusters: short term, medium term, and long term (see Fig.  1 ).

Temporal horizon for addressing ethical issues

As suggested by Baum ( 2017 ), it is best to acknowledge that there will be ethical issues and related mitigating activities that cannot exclusively fit in as short, medium or long term.

ather than seeing it as an authoritative classification, we see this as a heuristic that reflects aspects of the current discussion. One reason why this categorisation is useful is that the temporal horizon of ethical issues is a potentially useful variable, with companies often being accused of favouring short-term gains over long-term benefits. Similarly, short-term issues must be able to be addressed on the local level for short-term fixes to work.

Short-term issues

These are issues for which there is a reasonable assumption that they are capable of being addressed in the short term. We do not wish to quantify what exactly counts as short term, as any definition put forward will be contentious when analysing the boundaries and transition periods. A better definition of short term might therefore be that such issues can be expected to be successfully addressed in technical systems that are currently in operation or development. Many of the issues we discuss under the heading of short-term issues are directly linked to some of the key technologies driving the current AI debate, notably machine learning and some of its enabling techniques and approaches such as neural networks and reinforcement learning.

Many of the advantages promised by BD + AI involve the use of personal data, data which can be used to identify individuals. This includes health data; customer data; ANPR data (Automated Number Plate Recognition); bank data; and even includes data about farmers’ land, livestock, and harvests. Issues surrounding privacy and control of data are widely discussed and recognized as major ethical concerns that need to be addressed (Boyd & Crawford, 2012 ; Tene & Polonetsky, 2012 , 2013 ; Mittelstadt, Allo, Taddeo, Wachter, & Floridi, 2016 ; Jain, Gyanchandani, & Khare, 2016 ; Mai, 2016 ; Macnish, 2018 ). The concern surrounding privacy can be put down to a combination of a general level of awareness of privacy issues and the recently-introduced General Data Protection Regulation (GDPR). Closely aligned with privacy issues are those relating to transparency of processes dealing with data, which can often be classified as internal, external, and deliberate opaqueness (Burrell, 2016 ; Lepri, Staiano, Sangokoya, Letouzé, & Oliver, 2017 ; Mittelstadt, Allo, Taddeo, Wachter, & Floridi, 2016 ).

The Guidelines for Trustworthy AI Footnote 4 were released in 2018 by the High-Level Expert Group on Artificial Intelligence (AI HLEG Footnote 5 ), and address the need for technical robustness and safety, including accuracy, reproducibility, and reliability. Reliability is further linked to the requirements of diversity, fairness, and social impact because it addresses freedom from bias from a technical point of view. The concept of reliability, when it comes to BD + AI, refers to the capability to verify the stability or consistency of a set of results (Bush, 2012 ; Ferraggine, Doorn, & Rivera, 2009 ; Meeker and Hong, 2014 ).

If a technology is unreliable, error-prone, and unfit-for-purpose, adverse ethical issues may result from decisions made by the technology. The accuracy of recommendations made by BD + AI is a direct consequence of the degree of reliability of the technology (Barolli, Takizawa, Xhafa, & Enokido, 2019 ). Bias and discrimination in algorithms may be introduced consciously or unconsciously by those employing the BD + AI or because of algorithms reflecting pre-existing biases (Baroccas and Selbst, 2016 ). Examples of bias have been documented often reflecting “an imbalance in socio-economic or other ‘class’ categories—ie, a certain group or groups are not sampled as much as others or at all” (Panch et al., 2019 ). have the potential to affect levels of inequality and discrimination, and if biases are not corrected these systems can reproduce existing patterns of discrimination and inherit the prejudices of prior decision makers (Barocas & Selbst, 2016 , p. 674). An example of inherited prejudices is documented in the United States, where African-American citizens, more often than not, have been given longer prison sentences than Caucasians for the same crime.

Medium-term issues

Medium-term issues are not clearly linked to a particular technology but typically arise from the integration of AI techniques including machine learning into larger socio-technical systems and contexts. They are thus related to the way life in modern societies is affected by new technologies. These can be based on the specific issues listed above but have their main impact on the societal level. The use of BD + AI may allow individuals’ behaviour to be put under scrutiny and surveillance , leading to infringements on privacy, freedom, autonomy, and self-determination (Wolf, 2015 ). There is also the possibility that the increased use of algorithmic methods for societal decision-making may create a type of technocratic governance (Couldry & Powell, 2014 ; Janssen & Kuk, 2016 ), which could infringe on people’s decision-making processes (Kuriakose & Iyer, 2018 ). For example, because of the high levels of public data retrieval, BD + AI may harm people’s freedom of expression, association, and movement, through fear of surveillance and chilling effects (Latonero, 2018 ).

Corporations have a responsibility to the end-user to ensure compliance, accountability, and transparency of their BD + AI (Mittelstadt, Allo, Taddeo, Wachter, & Floridi, 2016 ). However, when the source of a problem is difficult to trace, owing to issues of opacity, it becomes challenging to identify who is responsible for the decisions made by the BD + AI. It is worth noting that a large-scale survey in Australia in 2020 indicated that 57.9% of end-users are not at all confident that most companies take adequate steps to protect user data. The significance of understanding and employing responsibility is an issue targeted in many studies (Chatfield et al., 2017 ; Fothergill et al., 2019 ; Jirotka et al., 2017 ; Pellé & Reber, 2015 ). Trust and control over BD + AI as an issue is reiterated by a recent ICO report demonstrating that most UK citizens do not trust organisations with their data (ICO, 2017 ).

Justice is a central concern in BD + AI (Johnson, 2014 , 2018 ). As a starting point, justice consists in giving each person his or her due or treating people equitably (De George, p. 101). A key concern is that benefits will be reaped by powerful individuals and organisations, while the burden falls predominantly on poorer members of society (Taylor, 2017 ). BD + AI can also reflect human intentionality, deploying patterns of power and authority (Portmess & Tower, 2015 , p. 1). The knowledge offered by BD + AI is often in the hands of a few powerful corporations (Wheeler, 2016 ). Power imbalances are heightened because companies and governments can deploy BD + AI for surveillance, privacy invasions and manipulation, through personalised marketing efforts and social control strategies (Lepri, Staiano, Sangokoya, Letouzé, & Oliver, 2017 , p. 11). They play a role in the ascent of datafication, especially when specific groups (such as corporate, academic, and state institutions) have greater unrestrained access to big datasets (van Dijck, 2014 , p. 203).

Discrimination , in BD + AI use, can occur when individuals are profiled based on their online choices and behaviour, but also their gender, ethnicity and belonging to specific groups (Calders, Kamiran, & Pechenizkiy, 2009 ; Cohen et al., 2014 ; and Danna & Gandy, 2002 ). Data-driven algorithmic decision-making may lead to discrimination that is then adopted by decision-makers and those in power (Lepri, Staiano, Sangokoya, Letouzé, & Oliver, 2017 , p. 4). Biases and discrimination can contribute to inequality . Some groups that are already disadvantaged may face worse inequalities, especially if those belonging to historically marginalised groups have less access and representation (Barocas & Selbst, 2016 , p. 685; Schradie, 2017 ). Inequality-enhancing biases can be reproduced in BD + AI, such as the use of predictive policing to target neighbourhoods of largely ethnic minorities or historically marginalised groups (O’Neil, 2016 ).

BD + AI offers great potential for increasing profit, reducing physical burdens on staff, and employing innovative sustainability practices (Badri, Boudreau-Trudel, & Souissi, 2018 ). They offer the potential to bring about improvements in innovation, science, and knowledge; allowing organisations to progress, expand, and economically benefit from their development and application (Crawford et al., 2014 ). BD + AI are being heralded as monumental for the economic growth and development of a wide diversity of industries around the world (Einav & Levin, 2014 ). The economic benefits accrued from BD + AI may be the strongest driver for their use, but BD + AI also holds the potential to cause economic harm to citizens and businesses or create other adverse ethical issues (Newman, 2013 ).

However, some in the literature view the co-development of employment and automation as somewhat naïve outlook (Zuboff, 2015 ). BD + AI companies may benefit from a ‘post-labour’ automation economy, which may have a negative impact on the labour market (Bossman, 2016 ), replacing up to 47% of all US jobs within the next 20 years (Frey & Osborne, 2017 ). The professions most at risk of affecting employment correlated with three of our case studies: farming, administration support and the insurance sector (Frey & Osborne, 2017 ).

Long-term issues

Long-term issues are those pertaining to fundamental aspects of nature of reality, society, or humanity. For example, that AI will develop capabilities far exceeding human beings (Kurzweil, 2006 ). At this point, sometimes called the ‘ singularity ’ machines achieve human intelligence, are expected to be able to improve on themselves and thereby surpass human intelligence and become superintelligent (Bostrom, 2016 ). If this were to happen, then it might have dystopian consequences for humanity as often depicted in science fiction. Also, it stands to reason that the superintelligent, or even just the normally intelligent machines may acquire a moral status.

It should be clear that these expectations are not universally shared. They refer to what is often called ‘ artificial general intelligence’ (AGI), a set of technologies that emulate human reasoning capacities more broadly. Footnote 6

Furthermore, if we may acquire new capabilities, e.g. by using technical implants to enhance human nature. The resulting being might be called a transhuman , the next step of human evolution or development. Again, it is important to underline that this is a contested idea (Livingstone, 2015 ) but one that has increasing traction in public discourse and popular science accounts (Harari, 2017 ).

We chose this distinction of three groups of issues for understanding how mitigation strategies within organisations can be contextualised. We concede that this is one reading of the literature and that many others are possible. In this account of the literature we tried to make sense of the current discourse to allow us to understand our empirical findings which are introduced in the following sections.

Case Study Methodology

Despite the impressive amount of research undertaken on ethical issues of AI + BD (e.g. Mittelstadt, Allo, Taddeo, Wachter, & Floridi, 2016 ; Zwitter, 2014 ), there are few case studies exploring such issues. This paper builds upon this research and employs an interpretivist methodology to do so, focusing on how, what, and why questions relevant to the ethical use of BD + AI (Walsham, 1995a , b ). The primary research questions for the case studies were: How do organisations perceive ethical concerns related to BD + AI and in what ways do they deal with them?

We sought to elicit insights from interviews, rather than attempting to reach an objective truth about the ethical impacts of BD + AI. The interpretivist case study approach (Stake 2003) allowed the researchers ‘to understand ‘reality’ as the blending of the various (and sometimes conflicting) perspectives which coexist in social contexts, the common threads that connect the different perspectives and the value systems that give rise to the seeming contradictions and disagreements around the topics discussed. Whether one sees this reality as static (social constructivism) or dynamic (social constructionism) was also a point of consideration, as they both belong in the same “family” approach where methodological flexibility is as important a value as rigour’ (XXX).

Through extensive brainstorming within the research team, and evaluations of relevant literature, 16 social application domains were established as topics for case study analysis. Footnote 7 The project focused on ten out of these application domains in accordance with the partners’ competencies. The case studies have covered ten domains, and each had their own unique focus, specifications, and niches, which added to the richness of the evaluations (Table 1 ).

The qualitative analysis approach adopted in this study focused on these ten standalone operational case studies that were directly related to the application domains presented in Table 1 . These individual case studies provide valuable insights (Yin, 2014 , 2015 ); however, a multiple-case study approach offers a more comprehensive analysis of ethical issues related to BD + AI use (Herriott & Firestone, 1983 ). Thus, this paper adopts a multiple-case study methodology to identify what insights can be obtained from the ten cases, identifies whether any generalisable understandings can be retrieved, and evaluates how different organisations deal with issues pertaining to BD + AI development and use. The paper does not attempt to derive universal findings from this analysis, in line with the principles of interpretive research, but further attempts to gain an in-depth understanding of the implications of selected BD + AI applications.

The data collection was guided by specific research questions identified through each case, including five desk research questions (see appendix 1); 24 interview questions (see appendix 2); and a checklist of 17 potential ethical issues, developed by the project leader Footnote 8 (see appendix 3). A thematic analysis framework was used to ‘highlight, expose, explore, and record patterns within the collected data. The themes were patterns across data sets that were important to describe several ethical issues which arise through the use of BD  +  AI across different types of organisations and application domains’ (XXX).

A workshop was then held after the interviews were carried out. The workshop brought together the experts in the case study team to discuss their findings. This culminated in 26 ethical issues Footnote 9 that were inductively derived from the data collected throughout the interviews (see Fig.  2 and Table 3). Footnote 10 In order to ensure consistency and rigour in the multiple-case study approach, researchers followed a standardised case study protocol (Yin, 2014 ). Footnote 11

The Prevalence of Ethical Issues in the Case Studies

Thirteen different organisations were interviewed for 10 case studies, consisting of 22 interviews in total. Footnote 12 These ranged from 30 min to 1 ½ hours in-person or Skype interviews. The participants that were selected for interviews represented a very broad range of application domains and organisations that use BD + AI. The case study organisations were selected according to their relevance to the overall case study domains and considering their fit with the domains and likelihood of providing interesting insights. The interviewees were then selected according to their ability to explain their BD + AI and its role in their organisation. In addition to interviews, a document review provided supporting information about the organisation. Thus, websites and published material were used to provide background to the research.

Findings: Ten Case Studies

This section gives a brief overview of the cases, before analysing their similarities and differences. It also highlights the different types of BD + AI being used, and the types of data used by the BD + AI in the case study organisations, before conducting an ethical analysis of the cases. Table 2 presents an overview of the 10 cases to show the roles of the interviewees, the focus of the technologies being used, and the data retrieved by each organisation’s BD + AI. All interviews were conducted in English.

The types of organisations that were used in the case studies varied extensively. They included start-ups (CS10), niche software companies (CS1), national health insurers (Organisation X in CS6), national energy providers (CS7), chemical/agricultural multinational (CS3), and national (CS9) and international (CS8) telecommunications providers. The case studies also included public (CS2, Organisation 1 and 4 in CS4) and semi-public (Organisation 2 in CS4) organisations, as well as a large scientific research project (CS5).

The types of individuals interviewed also varied extensively. For example, CS6 and CS7 did not have anyone with a specific technical background, which limited the possibility of analysing issues related to the technology itself. Some case studies only had technology experts (such as CS1, CS8, and CS9), who mostly concentrated on technical issues, with much less of a focus on ethical concerns. Other case studies had a combination of both technical and policy-focused experts (i.e. CS3, CS4, and CS5). Footnote 13

Therefore, it must be made fundamentally clear that we are not proposing that all of the interviewees were authorities in the field, or that even collectively they represent a unified authority on the matter, but instead, that we are hoping to show what are the insights and perceived ethical issues of those currently working with AI on the ground view as ethical concerns. While the paper is presenting the ethical concerns found within an array of domains, we do not claim that any individual case study is representative of their entire industry, but instead, our intent was to capture a wide diversity of viewpoints, domains, and applications of AI, to encompass a broad amalgamation of concerns. We should also state that this is not a shortcoming of the study but that it is the normal approach that social science often takes.

The diversity of organisations and their application focus areas also varied. Some organisations focused more so on the Big Data component of their AI, while others more strictly on the AI programming and analytics. Even when organisations concentrated on a specific type of BD + AI, such as Big Data, its use varied immensely, including retrieval (CS1), analysis (CS2), predictive analytics (CS10), and transactional value (Organisation 2 in CS4). Some domains adopted BD + AI earlier and more emphatically than others (such as communications, healthcare, and insurance). Also, the size, investment, and type of organisation played a part in the level of BD + AI innovation (for example, the two large multinationals in CS3 and CS8 had well-developed BD + AI).

The maturity level of BD + AI was also determined by how it was integrated, and its importance, within an organisation. For instance, in organisations where BD + AI were fundamental for the success of the business (e.g. CS1 and CS10), they played a much more important role than in companies where there was less of a reliance (e.g. CS7). In some organisations, even when BD + AI was not central to success, the level of development was still quite advanced because of economic investment capabilities (e.g. CS3 and CS8).

These differences provided important questions to ask throughout this multi-case study analysis, such as: Do certain organisations respond to ethical issues relating to BD + AI in a certain way? Does the type of interviewee affect the ethical issues discussed—e.g. case studies without technical experts, those that only had technical experts, and those that had both? Does the type of BD + AI used impact the types of ethical issues discussed? What significance does the type of data retrieved have on ethical issues identified by the organisations? These inductive ethical questions provided a template for the qualitative analysis in the following section.

Ethical Issues in the Case Studies

Based on the interview data, the ethical issues identified in the case studies were grouped into six specific thematic sections to provide a more conducive, concise, and pragmatic methodology. Those six sections are: control of data, reliability of data, justice, economic issues, role of organisations, and individual freedoms. From the 26 ethical issues, privacy was the only ethical issue addressed in all 10 case studies, which was not surprising because it has received a great deal of attention recently because of the GDPR. Also, security, transparency, and algorithmic bias are regularly discussed in the literature, so we expected them to be significant issues across many of the cases. However, there were many issues that received less attention in the literature—such as access to BD + AI, trust, and power asymmetries—which were discussed frequently in the interviews. In contrast to this, there were ethical issues that were heavily discussed in the literature which received far less attention in the interviews, such as employment, autonomy, and criminal or malicious use of BD + AI (Fig.  2 ).

The ethical analysis was conducted using a combination of literature reviews and interviews carried out with stakeholders. The purpose of the interviews was to ensure that there were no obvious ethical issues faced by stakeholders in their day-to-day activities which had been missed in the academic literature. As such, the starting point was not an overarching normative theory, which might have meant that we looked for issues which fit well with the theory but ignored anything that fell outside of that theory. Instead the combined approach led to the identification of the 26 ethical issues, each labelled based on particular words or phrases used in the literature or by the interviewees. For example, the term "privacy" was used frequently and so became the label for references to and instances of privacy-relevant concerns. In this section we have clustered issues together based on similar problems faced (e.g. accuracy of data and accuracy of algorithms within the category of ‘reliability of data’).

In an attempt to highlight similar ethical issues and improve the overall analysis to better capture similar perspectives, the research team decided to use the method of clustering, a technique often used in data mining to efficiently group similar elements together. Through discussion in the research team, and bearing in mind that the purpose of the clustering process was to form clusters that would enhance understanding of the impact of these ethical issues, we arrived at the following six clusters: the control of data (covering privacy, security, and informed consent); the reliability of data (accuracy of data and accuracy of algorithms); justice (power asymmetries, justice, discrimination, and bias); economic issues (economic concerns, sustainability, and employment); the role of organisations (trust and responsibility); and human freedoms (autonomy, freedom, and human rights). Both the titles and the precise composition of each cluster of issues are the outcome of a reasoned agreement of the research team. However, it should be clear that we could have used different titles and different clustering. The point is not that each cluster forms a distinct group of ethical issues, independent from any other. Rather the ethical issues faced overlap and play into one another, but to present them in a manageable format we have opted to use this bottom-up clustering approach.

Human Freedoms

An interviewee from CS10 stated that they were concerned about human rights because they were an integral part of the company’s ethics framework. This was beneficial to their business because they were required to incorporate human rights to receive public funding by the Austrian government. The company ensured that they would not grant ‘full exclusivity on generated social unrest event data to any single party, unless the data is used to minimise the risk of suppression of unrest events, or to protect the violation of human rights’ (XXX). The company demonstrates that while BD + AI has been criticised for infringing upon human rights in the literature, they also offer the opportunity to identify and prevent human rights abuses. The company’s moral framework definitively stemmed from regulatory and funding requirements, which lends itself to the benefit of effective ethical top-down approaches, which is a divisive topic in the literature, with diverging views about whether top-down or bottom-up approaches are better options for improved AI ethics.

Trust & Responsibility

Responsibility was a concern in 5 of the case studies, confirming the importance it is given in the literature (see Sect.  3 ). Trust appeared in seven of the case studies. The cases focused on concerns found in the literature, such as BD + AI use in policy development, public distrust about automated decision-making and the integrity of corporations utilising datafication methods (van Dijck 2014 ).

Trust and control over BD + AI were an issue throughout the case studies. The organisation from the predictive intelligence case study (CS10) identified that their use of social media data raised trust issues. They converged with perspectives found in the literature that when people feel disempowered to use or be part of the BD + AI development process, they tend to lose trust in the BD + AI (Accenture, 2016 , 2017 ). In CS6, stakeholders (health insurers) trusted the decisions made by BD + AI when they were engaged and empowered to give feedback on how their data was used. Trust is enhanced when users can refuse the use of their data (CS7), which correlates with the literature. Companies discussed the benefits of establishing trustworthy relationships. For example, in CS9, they have “ been trying really hard to avoid the existence of fake [mobile phone] base stations, because [these raise] an issue with the trust that people put in their networks” (XXX).

Corporations need to determine the objective of the data analysis (CS3), what data is required for the BD + AI to work (CS2), and accountability for when it does not work as intended or causes undesirable outcomes (CS4). The issue here is whether the organisation takes direct responsibility for these outcomes, or, if informed consent has been given, can responsibility be shared with the granter of consent (CS3). The cases also raised the question of ‘responsible to whom’, the person whose data is being used or the proxy organisation who has provided data (CS6). For example, in the insurance case study, the company stated that they only had a responsibility towards the proxy organisation and not the sources of the data. All these issues are covered extensively in the literature in most application domains.

Control of Data

Concerns surrounding the control of data for privacy reasons can be put down to a general awareness of privacy issues in the press, reinforced by the recently-introduced GDPR. This was supported in the cases, where interviewees expressed the opinion that the GDPR had raised general awareness of privacy issues (CS1, CS9) or that it had lent weight to arguments concerning the importance of privacy (CS8).

The discussion of privacy ranged from stressing that it was not an issue for some interviewees, because there was no personal information in the data they used (CS4), to its being an issue for others, but one which was being dealt with (CS2 and CS8). One interviewee (CS5) expressed apprehension that privacy concerns conflicted with scientific innovation, introducing hitherto unforeseen costs. This view is not uncommon in scientific and medical innovation, where harms arising from the use of anonymised medical data are often seen as minimal and the potential benefits significant (Manson & O’Neill, 2007 ). In other cases (CS1), there was a confusion between anonymisation (data which cannot be traced back to the originating source) and pseudonymisation (where data can be traced back, albeit with difficulty) of users’ data. A common response from the cases was that providing informed consent for the use of personal data waived some of the rights to privacy of the user.

Consent may come in the form of a company contract Footnote 14 or an individual agreement. Footnote 15 In the former, the company often has the advantage of legal support prior to entering a contract and so should be fully aware of the information provided. In individual agreements, though, the individual is less likely to be legally supported, and so may be at risk of exploitation through not reading the information sufficiently (CS3), or of responding without adequate understanding (CS9). In one case (CS5), referring to anonymised data, consent was implied rather than given: the interviewee suggested that those involved in the project may have contributed data without giving clear informed consent. The interviewee also noted that some data may have been shared without the permission, or indeed knowledge, of those contributing individuals. This was acknowledged by the interviewee as a potential issue.

In one case (CS6), data was used without informed consent for fraud detection purposes. The interviewees noted that their organisation was working within the parameters of national and EU legislation, which allows for non-consensual use of data for these ends. One interviewee in this case stated that informed consent was sought for every novel use of the data they held. However, this was sought from the perceived owner of the data (an insurance company) rather than from the originating individuals. This case demonstrates how people may expect their data to be used without having a full understanding of the legal framework under which the data are collected. For example, data relating to individuals may legally be accessed for fraud detection without notifying the individual and without relying on the individual’s consent.

This use of personal data for fraud detection in CS6 also led to concerns regarding opacity. In both CS6 and CS10 there was transparency within the organisations (a shared understanding among staff as to the various uses of the data) but that did not extend to the public outside those organisations. In some cases (CS5) the internal transparency/external opacity meant that those responsible for developing BD + AI were often hard to meet. Of those who were interviewed in CS5, many did not know the providence of the data or the algorithms they were using. Equally, some organisations saw external opacity as integral to the business environment in which they were operating (CS9, CS10) for reasons of commercial advantage. The interviewee in CS9 cautioned that this approach, coupled with a lack of public education and the speed of transformation within the industry, would challenge any meaningful level of public accountability. This would render processes effectively opaque to the public, despite their being transparent to experts.

Reliability of Data

There can be multiple sources of unreliability in BD + AI. Unreliability originating from faults in the technology can lead to algorithmic bias, which can cause ethical issues such as unfairness, discrimination, and general negative social impact (CS3 and CS6). Considering algorithmic bias as a key input to data reliability, there exist two types of issues that may need to be addressed. Primarily, bias may stem from the input data, referred to as training data, if such data excludes adequate representation of the world, e.g. gender-biased datasets (CS6). Secondly, an inadequate representation of the world may be the result of lack of data, e.g. a correctly designed algorithm to learn from and predict a rare disease, may not have sufficient representative data to achieve correct predictions (CS5). In either case the input data are biased and may result in inaccurate decision-making and recommendations.

The issues of reliability of data stemming from data accuracy and/or algorithmic bias, may escalate depending on their use, as for example in predictive or risk-assessment algorithms (CS10). Consider the risks of unreliable data in employee monitoring situations (CS1), detecting pests and diseases in agriculture (CS3), in human brain research (CS5) or cybersecurity applications (CS8). Such issues are not singular in nature but closely linked to other ethical issues such as information asymmetries, trust, and discrimination. Consequently, the umbrella issue of reliability of data must be approached from different perspectives to ensure the validity of the decision-making processes of the BD + AI.

Data may over-represent some people or social groups who are likely to be already privileged or under-represent disadvantaged and vulnerable groups (CS3). Furthermore, people who are better positioned to gain access to data and have the expertise to interpret them may have an unfair advantage over people devoid of such competencies. In addition, BD + AI can work as a tool of disciplinary power, used to evaluate people’s conformity to norms representing the standards of disciplinary systems (CS5). We focus on the following aspects of justice in our case study analysis: power asymmetries, discrimination, inequality, and access.

The fact that issues of power can arise in public as well as private organisations was discussed in our case studies. The smart city case (CS4) showed that the public organisations were aware of potential problems arising from companies using public data and were trying to put legal safeguards in place to avoid such misuse. As a result of misuse, there is the potential that cities, or the companies with which they contract, may use data in harmful or discriminatory ways. Our case study on the use of BD + AI in scientific research showed that the interviewees were acutely aware of the potential of discrimination (CS10). They stated that biases in the data may not be easy to identify, and may lead to misclassification or misinterpretation of findings, which may in turn skew results. Discrimination refers to the recognition of difference, but it may also refer to unjust treatment of different categories of people based on their gender, sex, religion, race, class, or disability. BD + AI are often employed to distinguish between different cases, e.g. between normal and abnormal behaviour in cybersecurity. Determining whether such classification entails discrimination in the latter sense can be difficult, due to the nature of the data and algorithms involved.

Examples of potential inequality based on BD + AI could be seen in several case studies. The agricultural case (CS3) highlighted the power differential between farmers and companies with potential implications for inequality, but also the global inequality between farmers, linked to farming practices in different countries (CS3). Subsistence farmers in developing countries, for example, might find it more difficult to benefit from these technologies than large agro-businesses. The diverging levels of access to BD + AI entail different levels of ability to benefit from them and counteract possible disadvantages (CS3). Some companies restrict access to their data entirely, and others sell access at a fee, while others offer small datasets to university-based researchers (Boyd & Crawford, 2012 , p. 674).

Economic Issues

One economic impact of BD + AI outlined in the agriculture case study (CS3) focused on whether this technology, and their ethical implementation, were economically affordable. If BD + AI could not improve economic efficiency, they would be rejected by the end-user, whether they were more productive, sustainable, and ethical options. This is striking, as it raises a serious challenge for the AI ethics literature and industry. It establishes that no matter how well intentioned and principled AI ethics guidelines and charters are, unless their implementation can be done in an economically viable way, their implementation will be challenged and resisted by those footing the bill.

The telecommunications case study (CS9) focused on how GDPR legislation may economically impact businesses using BD + AI by creating disparities in competitiveness between EU and non-EU companies developing BD + AI. Owing to the larger data pools of the latter, their BD + AI may prove to be more effective than European-manufactured alternatives, which cannot bypass the ethical boundaries of European law in the same way (CS8). This is something that is also being addressed in the literature and is a very serious concern for the future profitability and development of AI in Europe (Wallace & Castro, 2018 ). The literature notes additional issues in this area that were not covered in the cases. There is the potential that the GDPR will increase costs of European AI companies by having to manually review algorithmic decision-making; the right to explanation could reduce AI accuracy; and the right to erasure could damage AI systems (Wallace & Castro, 2018 , p. 2).

One interviewee stated that public–private BD + AI projects should be conducted in a collaborative manner, rather than a sale-of-service (CS4). However, this harmonious partnership is often not possible. Another interviewee discussed the tension between public and private interests on their project—while the municipality tried to focus on citizen value, the ICT company focused on the project’s economic success. The interviewee stated that the project would have terminated earlier if it were the company’s decision, because it was unprofitable (CS4). This is a huge concern in the literature, whereby private interests will cloud, influence, and damage public decision-making within the city because of their sometimes-incompatible goals (citizen value vs. economic growth) (Sadowski & Pasquale, 2015 ). One interviewee said that the municipality officials were aware of the problems of corporate influence and thus are attempting to implement the approach of ‘data sovereignty’ (CS2).

During our interviews, some viewed BD + AI as complementary to human employment (CS3), collaborative with such employment (CS4), or as a replacement to employment (CS6). The interviewees from the agriculture case study (CS3) stated that their BD + AI were not sufficiently advanced to replace humans and were meant to complement the agronomist, rather than replace them. However, they did not indicate what would happen when the technology is advanced enough, and it becomes profitable to replace the agronomist. The insurance company interviewee (CS6) stated that they use BD + AI to reduce flaws in personal judgment. The literature also supports this viewpoint, where BD + AI is seen to offer the potential to evaluate cases impartially, which is beneficial to the insurance industry (Belliveau, Gray, & Wilson, 2019 ). Footnote 16 The interviewee reiterated this and also stated that BD + AI would reduce the number of people required to work on fraud cases. The interviewee stated that BD + AI are designed to replace these individuals, but did not indicate whether their jobs were secure or whether they would be retrained for different positions, highlighting a concern found in the literature about the replacement and unemployment of workers by AI (Bossman, 2016 ). In contrast to this, a municipality interviewee from CS4 stated that their chat-bots are used in a collaborative way to assist customer service agents, allowing them to concentrate on higher-level tasks, and that there are clear policies set in place to protect their jobs.

Sustainability was only explicitly discussed in two interviews (CS3 and CS4). The agriculture interviewees stated that they wanted to be the ‘first’ to incorporate sustainability metrics into agricultural BD + AI, indicating a competitive and innovative rationale for their company (CS3). Whereas the interviewee from the sustainable development case study (CS4) stated that their goal of using BD + AI was to reduce Co2 emissions and improve energy and air quality. He stated that there are often tensions between ecological and economic goals and that this tension tends to slow down the efforts of BD + AI public–private projects—an observation also supported by the literature (Keeso, 2014 ). This tension between public and private interests in BD + AI projects was a recurring issue throughout the cases, which will be the focus of the next section on the role of organisations.

Discussion and Conclusion

The motivation behind this paper is to come to a better understanding of ethical issues related to BD + AI based on a rich empirical basis across different application domains. The exploratory and interpretive approach chosen for this study means that we cannot generalise from our research to all possible examples of BD + AI, but it does allow us to generalise to theory and rich insights (Walsham, 1995a , b , 2006 ). These theoretical insights can then provide the basis for further empirical research, possibly using other methods to allow an even wider set of inputs to move beyond some of the limitations of the current study.

Organisational Practice and the Literature

The first point worth stating is that there is a high level of consistency both among the case studies and between cases and literature. Many of the ethical issues identified cut across the cases and are interpreted in similar ways by different stakeholders. The frequency distribution of ethical issues indicates that very few, if any, issues are relevant to all cases but many, such as privacy, have a high level of prevalence. Despite appearing in all case studies, privacy was not seen as overly problematic and could be dealt with in the context of current regulatory principles (GDPR). Most of the issues that we found in the literature (see Sect.  2 ) were also present in the case studies. In addition to privacy and data protection, this included accuracy, reliability, economic and power imbalances, justice, employment, discrimination and bias, autonomy and human rights and freedoms.

Beyond the general confirmation of the relevance of topics discussed in the literature, though, the case studies provide some further interesting insights. From the perspective of an individual case some societal factors are taken for granted and outside of the control of individual actors. For example, intellectual property regimes have significant and well-recognised consequences for justice, as demonstrated in the literature. However, there is often little that individuals or organisations can do about them. Even in cases where individuals may be able to make a difference and the problem is clear, it is not always obvious how to do this. Some well-publicised discrimination cases may be easy to recognise, for example where an HR system discriminates against women or where a facial recognition system discriminates against black people. But in many cases, it may be exceedingly difficult to recognise discrimination where it is not clear how a person is discriminated against. If, for example, an image-based medical diagnostic system leads to disadvantages for people with genetic profiles, this may not be easy to identify.

With regards to the classification of the literature suggested in Sect.  2 along the temporal dimension, we can see that the attention of the case study respondents seems to be correlated to the temporal horizon of the issues. The issues we see as short-term figures most prominently, whereas the medium-term issues, while still relevant and recognisable, appear to be less pronounced. The long-term questions are least visible in the cases. This is not very surprising, as the short-term issues are those that are at least potentially capable of being addressed relatively quickly and thus must be accessible on the local level. Organisations deploying or using AI therefore are likely to have a responsibility to address these issues and our case studies have shown that they are aware of this and putting measures in place. This is clearly true for data protection or security issues. The medium-term issues that are less likely to find local resolutions still figure prominently, even though an individual organisation has less influence on how they can be addressed. Examples of this would be questions of unemployment, justice, or fairness. There was little reference to what we call long-term issues, which can partly be explained by the fact that the type of AI user organisations we investigated have very limited influence on how they are perceived and how they may be addressed.

Interpretative Differences on Ethical Issues

Despite general agreement on the terminology used to describe ethical issues, there are often important differences in interpretation and understanding. In the first ethics theme, control of data, the perceptions of privacy ranged from ‘not an issue’ to an issue that was being dealt with. Some of this arose from the question of informed consent and the GDPR. However, a reliance on legislation, such as GDPR, without full knowledge of the intricacies of its details (i.e. that informed consent is only one of several legal bases of lawful data processing), may give rise to a false sense of security over people’s perceived privacy. This was also linked to the issue of transparency (of processes dealing with data), which may be external to the organisation (do people outside understand how an organisation holds and processes their data), or internal (how well does the organisation understand the algorithms developed internally) and sometimes involve deliberate opacity (used in specific contexts where it is perceived as necessary, such as in monitoring political unrest and its possible consequences). Therefore, a clearer and more nuanced understanding of privacy and other ethical terms raised here might well be useful, albeit tricky to derive in a public setting (for an example of complications in defining privacy, see Macnish, 2018 ).

Some issues from the literature were not mentioned in the cases, such as warfare. This can easily be explained by our choice of case studies, none of which drew on work done in this area. It indicates that even a set of 10 case studies falls short of covering all issues.

A further empirical insight is in the category we called ‘role of organisations’, which covers trust and responsibility. Trust is a key term in the discussion of the ethics of AI, prominently highlighted by the focus on trustworthy AI by the EU’s High-Level Expert Group, among others. We put this into the ‘role of organisations’ category because our interaction with the case study respondents suggested that they felt it was part of the role of their organisations to foster trust and establish responsibilities. But we are open to the suggestion that these are concepts on a slightly different level that may provide the link between specific issues in applications and broader societal debate.

Next Steps: Addressing the Ethics of AI and Big Data

This paper is predominantly descriptive, and it aims to provide a theoretically sound and empirically rich account of ethical concerns in AI + BD. While we hope that it proves to be insightful it is only a first step in the broader journey towards addressing and resolving these issues. The categorisation suggested here gives an initial indication of which type of actor may be called upon to address which type of issue. The distinction between micro-, meso- and macro perspectives suggested by Haenlein and Kaplan ( 2019 ) resonates to some degree with our categorisation of issues.

This points to the question what can be done to address these ethical issues and by whom should it be done? We have not touched on this question in the theoretical or empirical part of the paper, but the question of mitigation is the motivating force behind much of the AI + BD ethics research. The purpose of understanding these ethical questions is to find ways of addressing them.

This calls for a more detailed investigation of the ethical nature of the issues described here. As indicated earlier, we did not begin with a specific ethical theoretical framework imposed onto the case studies, but did have some derived ethics concepts which we explored within the context of the cases and allowed others to emerge over the course of the interviews. One issue is the philosophical question whether the different ethical issues discussed here are of a similar or comparable nature and what characterises them as ethical issues. This is not only a philosophical question but also a practical one for policymakers and decision makers. We have alluded to the idea that privacy and data protection are ethical issues, but they also have strong legal implications and can also be human rights issues. It would therefore be beneficial to undertake a further analysis to investigate which of these ethical issues are already regulated and to what degree current regulation covers BD + AI, and how this varies across the various EU nations and beyond.

Another step could be to expand an investigation like the one presented here to cover the ethics of AI + BD debate with a focus on suggested resolutions and policies. This could be achieved by adopting the categorisation and structure presented here and extending it to the currently discussed option for addressing the ethical issues. These include individual and collective activities ranging from technical measures to measure bias in data or individual professional guidance to standardisation, legislation, the creation of a specific regulator and many more. It will be important to understand how these measures are conceptualised as well as which ones are already used to which effect. Any such future work, however, will need to be based on a sound understanding of the issues themselves, which this paper contributes to. The key contribution of the paper, namely the presentation of empirical findings from 10 case studies show in more detail how ethical issues play out in practice. While this work can and should be expanded by including an even broader variety of cases and could be supplemented by other empirical research methods, it marks an important step in the development of our understanding of these ethical issues. This should form a part of the broader societal debate about what these new technologies can and should be used for and how we can ensure that their consequences are beneficial for individuals and society.

Throughout the paper, XXX will be used to anonymise relevant text that may identify the authors, either through the project and/or publications resulting from the individual case studies. All case studies have been published individually. Several the XXX references in the findings refer to these individual publications which provide more detail on the cases than can be provided in this cross-case analysis.

The ethical issues that we discussed throughout the case studies refers to issues broadly construed as ethical issues, or issues that have ethical significance. While some issues may not be directly obvious how they are ethical issues, they may give rise to significant harm relevant to ethics. For example, accuracy of data may not explicitly be an ethical issue, if inaccurate data is used in algorithms, it may lead to discrimination, unfair bias, or harms to individuals.

Such as chat-bots, natural language processing AI, IoT data retrieval, predictive risk analysis, cybersecurity machine-learning, and large dataset exchanges.

https://ec.europa.eu/futurium/en/ai-alliance-consultation/guidelines/1 .

https://ec.europa.eu/digital-single-market/en/high-level-expert-group-artificial-intelligence .

The type of AI currently in vogue, as outlined earlier, is based on machine learning, typically employing artificial neural networks for big data analysis. This is typically seen as ‘narrow AI’ and it is not clear whether there is a way from narrow to general AI, even if one were to accept that achieving general AI is fundamentally possible.

The 16 social domains were: Banking and securities; Healthcare; Insurance; Retail and wholesale trade; Science; Education; Energy and utilities; Manufacturing and natural resources; Agriculture; Communications, media and entertainment; Transportation; Employee monitoring and administration; Government; Law enforcement and justice; Sustainable development; and Defence and national security.

This increased to 26 ethical issues following a group brainstorming session at the case study workshop.

The nine additional ethical issues from the initial 17 drafted by the project leader were: human rights, transparency, responsibility, ownership of data, algorithmic bias, integrity, human rights, human contact, and accuracy of data.

The additional ethical issues were access to BD + AI, accuracy of data, accuracy of recommendations, algorithmic bias, economic, human contact, human rights, integrity, ownership of data, responsibility, and transparency. Two of the initial ethical concerns were removed (inclusion of stakeholders and environmental impact). The issues raised concerning inclusion of stakeholders were deemed to be sufficiently included in access to BD + AI, and those relating to environmental impact were felt to be sufficiently covered by sustainability.

The three appendices attached in this paper comprise much of this case study protocol.

CS4 evaluated four organisations, but one of these organisations was also part of CS2 – Organisation 1. CS6 analysed two insurance organisations.

Starting out, we aimed to have both policy/ethics-focused experts within the organisation and individuals that could also speak with us about the technical aspects of the organisation’s BD + AI. However, this was often not possible, due to availability, organisations’ inability to free up resources (e.g. employee’s time) for interviews, or lack of designated experts in those areas.

For example, in CS1, CS6, and CS8.

For example, in CS2, CS3, CS4, CS5, CS6, and CS9.

As is discussed elsewhere in this paper, algorithms also hold the possibility of reinforcing our prejudices and biases or creating new ones entirely.

Accenture. (2016). Building digital trust: The role of data ethics in the digital age. Retrieved December 1, 2020 from https://www.accenture.com/t20160613T024441__w__/us-en/_acnmedia/PDF-22/Accenture-Data-Ethics-POV-WEB.pdf .

Accenture. (2017). Embracing artificial intelligence. Enabling strong and inclusive AI driven growth. Retrieved December 1, 2020 from https://www.accenture.com/t20170614T130615Z__w__/us-en/_acnmedia/Accenture/next-gen-5/event-g20-yea-summit/pdfs/Accenture-Intelligent-Economy.pdf .

Antoniou, J., & Andreou, A. (2019). Case study: The Internet of Things and Ethics. The Orbit Journal, 2 (2), 67.

Google Scholar  

Badri, A., Boudreau-Trudel, B., & Souissi, A. S. (2018). Occupational health and safety in the industry 4.0 era: A cause for major concern? Safety Science, 109, 403–411. https://doi.org/10.1016/j.ssci.2018.06.012

Article   Google Scholar  

Barolli, L., Takizawa, M., Xhafa, F., & Enokido, T. (ed.) (2019). Web, artificial intelligence and network applications. In Proceedings of the workshops of the 33rd international conference on advanced information networking and applications , Springer.

Barocas, S., & Selbst, A. D. (2016). Big data’s disparate impact. California Law Review, 104 (671), 671–732. https://doi.org/10.15779/Z38BG31

Baum, S. D. (2017). Reconciliation between factions focused on near-term and long-term artificial intelligence. AI Society, 2018 (33), 565–572.

Belliveau, K. M., Gray, L. E., & Wilson, R. J. (2019). Busting the Black Box: Big Data Employment and Privacy | IADC LAW. https://www.iadclaw.org/publications-news/defensecounseljournal/busting-the-black-box-big-data-employment-and-privacy/ . Accessed 10 May 2019.

Bossman, J. (2016). Top 9 ethical issues in artificial intelligence. World Economic Forum . https://www.weforum.org/agenda/2016/10/top-10-ethical-issues-in-artificial-intelligence/ . Accessed 10 May 2019.

Bostrom, N. (2016). Superintelligence: Paths . OUP Oxford.

Boyd, D., & Crawford, K. (2012). Critical questions for big data. Information, Communication and Society, 15 (5), 662–679. https://doi.org/10.1080/1369118X.2012.678878

Burrell, J. (2016). How the machine ‘thinks’: Understanding opacity in machine learning algorithms. Big Data and Society, 3 (1), 2053951715622512.

Bush, T., (2012). Authenticity in Research: Reliability, Validity and Triangulation. Chapter 6 in edited “Research Methods in Educational Leadership and Management”, SAGE Publications.

Calders, T., Kamiran, F., & Pechenizkiy, M. (2009). Building classifiers with independency constraints. In IEEE international conference data mining workshops , ICDMW’09, Miami, USA.

Chatfield, K., Iatridis, K., Stahl, B. C., & Paspallis, N. (2017). Innovating responsibly in ICT for ageing: Drivers, obstacles and implementation. Sustainability, 9 (6), 971. https://doi.org/10.3390/su9060971 .

Cohen, I. G., Amarasingham, R., Shah, A., et al. (2014). The legal and ethical concerns that arise from using complex predictive analytics in health care. Health Affairs, 33 (7), 1139–1147.

Couldry, N., & Powell, A. (2014). Big Data from the bottom up. Big Data and Society, 1 (2), 205395171453927. https://doi.org/10.1177/2053951714539277

Crawford, K., Gray, M. L., & Miltner, K. (2014). Big data| critiquing big data: Politics, ethics, epistemology | special section introduction. International Journal of Communication, 8, 10.

Cuquet, M., & Fensel, A. (2018). The societal impact of big data: A research roadmap for Europe. Technology in Society, 54, 74–86.

Danna, A., & Gandy, O. H., Jr. (2002). All that glitters is not gold: Digging beneath the surface of data mining. Journal of Business Ethics, 40 (4), 373–438.

European Convention for the Protection of HUman Rights and Fundamental Freedoms, pmbl., Nov. 4, 1950, 213 UNTS 221.

Herriott, E. R., & Firestone, W. (1983). Multisite qualitative policy research: Optimizing description and generalizability. Educational Researcher, 12, 14–19. https://doi.org/10.3102/0013189X012002014

Einav, L., & Levin, J. (2014). Economics in the age of big data. Science, 346 (6210), 1243089. https://doi.org/10.1126/science.1243089

Ferraggine, V. E., Doorn, J. H., & Rivera, L. C. (2009). Handbook of research on innovations in database technologies and applications: Current and future trends (pp. 1–1124). IGI Global.

Fothergill, B. T., Knight, W., Stahl, B. C., & Ulnicane, I. (2019). Responsible data governance of neuroscience big data. Frontiers in Neuroinformatics, 13 . https://doi.org/10.3389/fninf.2019.00028

Frey, C. B., & Osborne, M. A. (2017). The future of employment: How susceptible are jobs to computerisation? Technological Forecasting and Social Change, 114, 254–280. https://doi.org/10.1016/j.techfore.2016.08.019

Haenlein, M., & Kaplan, A. (2019). A brief history of artificial intelligence: On the past, present, and future of artificial intelligence. California Management Review, 61 (4), 5–14.

Harari, Y. N. (2017). Homo deus: A brief history of tomorrow (1st ed.). Vintage.

Book   Google Scholar  

ICO. (2017). Big data, artificial intelligence, machine learning and data protection. Retrieved December 1, 2020 from Information Commissioner’s Office website: https://iconewsblog.wordpress.com/2017/03/03/ai-machine-learning-and-personal-data/ .

Ioannidis, J. P. (2013). Informed consent, big data, and the oxymoron of research that is not research. The American Journal of Bioethics., 2, 15.

Jain, P., Gyanchandani, M., & Khare, N. (2016). Big data privacy: A technological perspective and review. Journal of Big Data, 3 (1), 25.

Janssen, M., & Kuk, G. (2016). The challenges and limits of big data algorithms in technocratic governance. Government Information Quarterly, 33 (3), 371–377. https://doi.org/10.1016/j.giq.2016.08.011

Jirotka, M., Grimpe, B., Stahl, B., Hartswood, M., & Eden, G. (2017). Responsible research and innovation in the digital age. Communications of the ACM, 60 (5), 62–68. https://doi.org/10.1145/3064940

Jiya, T. (2019). Ethical Implications Of Predictive Risk Intelligence. ORBIT Journal, 2 (2), 51.

Jiya, T. (2019). Ethical reflections of human brain research and smart information systems. The ORBIT Journal, 2 (2), 1–24.

Jobin, A., Ienca, M., & Vayena, E. (2019). The global landscape of AI ethics guidelines. Nature Machine Intelligence, 1 (9), 389–399. https://doi.org/10.1038/s42256-019-0088-2

Johnson, J. A. (2014). From open data to information justice. Ethics and Information Technology, 4 (16), 263–274.

Johnson, J. A. (2018). Open data, big data, and just data. In J. A. Johnson (Ed.), Toward information justice (pp. 23–49). Berlin: Springer.

Chapter   Google Scholar  

Kancevičienė, N. (2019). Insurance, smart information systems and ethics: a case study. The ORBIT Journal, 2 (2), 1–27.

Keeso, A. (2014). Big data and environmental sustainability: A conversation starter . https://www.google.com/search?rlz=1C1CHBF_nlNL796NL796&ei=YF3VXN3qCMLCwAKp4qjYBQ&q=Keeso+Big+Data+and+Environmental+Sustainability%3A+A+Conversation+Starter&oq=Keeso+Big+Data+and+Environmental+Sustainability%3A+A+Conversation+Starter&gs_l=psy-ab.3...15460.16163..16528...0.0..0.76.371.6......0....1..gws-wiz.......0i71j35i304i39j0i13i30.M_8nNbaL2E8 . Accessed 10 May 2019.

Kuriakose, F., & Iyer, D. (2018). Human Rights in the Big Data World (SSRN Scholarly Paper No. ID 3246969). Rochester, NY: Social Science Research Network. https://papers.ssrn.com/abstract=3246969 . Accessed 13 May 2019.

Kurzweil, R. (2006). The singularity is near . Gerald Duckworth & Co Ltd.

Latonero, M. (2018). Big data analytics and human rights. New Technologies for Human Rights Law and Practice. https://doi.org/10.1017/9781316838952.007

Lepri, B., Staiano, J., Sangokoya, D., Letouzé, E., & Oliver, N. (2017). The tyranny of data? the bright and dark sides of data-driven decision-making for social good. In Transparent data mining for big and small data (pp. 3–24). Springer.

Livingstone, D. (2015). Transhumanism: The history of a dangerous idea . CreateSpace Independent Publishing Platform.

Macnish, K. (2018). Government surveillance and why defining privacy matters in a post-snowden world. Journal of Applied Philosophy, 35 (2), 417–432.

Macnish, K., & Inguanzo, A. (2019). Case study-customer relation management, smart information systems and ethics. The ORBIT Journal, 2 (2), 1–24.

Macnish, K., Inguanzo, A. F., & Kirichenko, A. (2019). Smart information systems in cybersecurity. ORBIT Journal, 2 (2), 15.

Mai, J. E. (2016). Big data privacy: The datafication of personal information. The Information Society, 32 (3), 192–199.

Manson, N. C., & O’Neill, O. (2007). Rethinking informed consent in bioethics . Cambridge University Press.

Mittelstadt, B. D., Allo, P., Taddeo, M., Wachter, S., & Floridi, L. (2016). The ethics of algorithms: Mapping the debate. Big Data and Society, 3 (2), 2053951716679679.

Meeker, Q. W., & , Hong, Y. . (2014). Reliability Meets big data: Opportunities and challenges. Quality Engineering, 26 (1), 102–116.

Newman, N. (2013). The costs of lost privacy: Consumer harm and rising economic inequality in the age of google (SSRN Scholarly Paper No. ID 2310146). Rochester: Social Science Research Network. https://papers.ssrn.com/abstract=2310146 . Accessed 10 May 2019.

O’Neil, C. (2016). Weapons of math destruction: How big data increases inequality and threatens democracy . Crown Publishers.

Panch, T., Mattie, H., & Atun, R. (2019). Artificial intelligence and algorithmic bias: implications for health systems. Journal of global health, 9 (2).

Pellé, S., & Reber, B. (2015). Responsible innovation in the light of moral responsibility. Journal on Chain and Network Science, 15 (2), 107–117. https://doi.org/10.3920/JCNS2014.x017

Portmess, L., & Tower, S. (2015). Data barns, ambient intelligence and cloud computing: The tacit epistemology and linguistic representation of Big Data. Ethics and Information Technology, 17 (1), 1–9. https://doi.org/10.1007/s10676-014-9357-2

Ryan, M. (2019). Ethics of public use of AI and big data. ORBIT Journal, 2 (2), 15.

Ryan, M. (2019). Ethics of using AI and big data in agriculture: The case of a large agriculture multinational. The ORBIT Journal, 2 (2), 1–27.

Ryan, M., & Gregory, A. (2019). Ethics of using smart city AI and big data: The case of four large European cities. The ORBIT Journal, 2 (2), 1–36.

Sadowski, J., & Pasquale, F. A. (2015). The spectrum of control: A social theory of the smart city. First Monday, 20 (7), 16.

Schradie, J. (2017). Big data is too small: Research implications of class inequality for online data collection. In D. June & P. Andrea (Eds.), Media and class: TV, film and digital culture . Abingdon: Taylor and Francis.

Taylor, L. (2017). ‘What is data justice? The case for connecting digital rights and freedoms globally’ In Big data and society (pp. 1–14). https://doi.org/10.1177/2053951717736335 .

Tene, O., & Polonetsky, J. (2012). Big data for all: Privacy and user control in the age of analytics. The Northwestern Journal of Technology and Intellectual Property, 11, 10.

Tene, O., & Polonetsky, J. (2013). A theory of creepy: technology, privacy and shifting social norms. Yale JL and Technology, 16, 59.

Van Dijck, J., & Poell, T. (2013). Understanding social media logic. Media and Communication, 1 (1), 2–14.

Voinea, C., & Uszkai, R. (n.d.). An assessement of algorithmic accountability methods .

Walsham, G. (1995). Interpretive case studies in IS research: nature and method. European Journal of Information Systems, 4 (2), 74–81.

Wallace, N., & Castro, D. (2018) The Impact of the EU’s New Data Protection Regulation on AI, Centre for Data Innovation .

Walsham, G. (1995). Interpretive case-studies in IS research-nature and method. European Journal of Information Systems, 4 (2), 74–81.

Walsham, G. (2006). Doing interpretive research. European Journal of Information Systems, 15 (3), 320–330.

Wheeler, G. (2016). Machine epistemology and big data. In L. McIntyre & A. Rosenburg (Eds.), Routledge Companion to Philosophy of Social Science . Routledge.

Whittlestone, J., Nyrup, R., Alexandrova, A., Dihal, K., & Cave, S. (2019). Ethical and societal implications of algorithms, data, and artificial intelligence: A roadmap for research. https://www.nuffieldfoundation.org/sites/default/files/files/Ethical-and-Societal-Implications-of-Data-and-AI-report-Nuffield-Foundat.pdf .

Wolf, B. (2015). Burkhardt Wolf: Big data, small freedom? / Radical Philosophy. Radical Philosophy . https://www.radicalphilosophy.com/commentary/big-data-small-freedom . Accessed 13 May 2019.

Yin, R. K. (2014). Case study research: Design and methods (5th ed.). SAGE.

Yin, R. K. (2015). Qualitative research from start to finish . Guilford Publications.

Zwitter, A. (2014). Big data ethics. Big Data and Society, 1 (2), 51.

Zuboff, S. (2015). Big other: Surveillance capitalism and the prospects of an information civilization (April 4, 2015). Journal of Information Technology, 2015 (30), 75–89. https://doi.org/10.1057/jit.2015.5

Download references

Acknowledgements

This SHERPA Project has received funding from the European Union’s Horizon 2020 Framework Programme for Research and Innovation under the Specific Grant Agreement No. 786641. The author(s) acknowledge the contribution of the consortium to the development and design of the case study approach.

Author information

Authors and affiliations.

Wageningen Economic Research, Wageningen University and Research, Wageningen, The Netherlands

UCLan Cyprus, Larnaka, Cyprus

Josephina Antoniou

De Montford University, Leicester, UK

Laurence Brooks & Bernd Stahl

Northampton University, Northampton, UK

Tilimbe Jiya

The University of Twente, Enschede, The Netherlands

Kevin Macnish

You can also search for this author in PubMed   Google Scholar

Corresponding author

Correspondence to Mark Ryan .

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix 1: Desk Research Questions

Number Research Question.

In which sector is the organisation located (e.g. industry, government, NGO, etc.)?

What is the name of the organisation?

What is the geographic scope of the organisation?

What is the name of the interviewee?

What is the interviewee’s role within the organisation?

Appendix 2: Interview Research Questions

No Research Question.

What involvement has the interviewee had with BD + AI within the organisation?

What type of BD + AI is the organisation using? (e.g. IBM Watson, Google Deepmind)

What is the field of application of the BD + AI (e.g. administration, healthcare, retail)

Does the BD + AI work as intended or are there problems with its operation?

What are the innovative elements introduced by the BD + AI (e.g. what has the technology enabled within the organisation?)

What is the level of maturity of the BD + AI ? (i.e. has the technology been used for long at the organisation? Is it a recent development or an established approach?)

How does the BD + AI interact with other technologies within the organisation?

What are the parameters/inputs used to inform the BD + AI ? (e.g. which sorts of data are input, how is the data understood within the algorithm?). Does the BD + AI collect and/or use data which identifies or can be used to identify a living person (personal data)?. Does the BD + AI collect personal data without the consent of the person to whom those data relate?

What are the principles informing the algorithm used in the BD + AI (e.g. does the algorithm assume that people walk in similar ways, does it assume that loitering involves not moving outside a particular radius in a particular time frame?). Does the BD + AI classify people into groups? If so, how are these groups determined? Does the BD + AI identify abnormal behaviour? If so, what is abnormal behaviour to the BD + AI ?

Are there policies in place governing the use of the BD + AI ?

How transparent is the technology to administrators within the organisation, to users within the organisation?

Who are the stakeholders in the organisation?

What has been the impact of the BD + AI on stakeholders?

How transparent is the technology to people outside the organisation?

Are those stakeholders engaged with the BD + AI ? (e.g. are those affected aware of the BD + AI, do they have any say in its operation?). If so, what is the nature of this engagement? (focus groups, feedback, etc.)

In what way are stakeholders impacted by the BD + AI ? (e.g. what is the societal impact: are there issues of inequality, fairness, safety, filter bubbles, etc.?)

What are the costs of using the BD + AI to stakeholders? (e.g. potential loss of privacy, loss of potential to sell information, potential loss of reputation)

What is the expected longevity of this impact? (e.g. is this expected to be temporary or long-term?)

Are those stakeholders engaged with the BD + AI ? (e.g. are those affected aware of the BD + AI, do they have any say in its operation?)

If so, what is the nature of this engagement? (focus groups, feedback, etc.)

Appendix 3: Checklist of Ethical Issues

Rights and permissions.

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .

Reprints and permissions

About this article

Ryan, M., Antoniou, J., Brooks, L. et al. Research and Practice of AI Ethics: A Case Study Approach Juxtaposing Academic Discourse with Organisational Reality. Sci Eng Ethics 27 , 16 (2021). https://doi.org/10.1007/s11948-021-00293-x

Download citation

Received : 26 August 2019

Accepted : 10 February 2021

Published : 08 March 2021

DOI : https://doi.org/10.1007/s11948-021-00293-x

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Smart information systems
• Big data analytics
• Artificial intelligence ethics
• Multiple-case study analysis
• Philosophy of technology
• Find a journal
• Publish with us
• Track your research

Publications

• Analysis & Opinions
• News & Announcements
• Newsletters
• Policy Briefs & Testimonies
• Presentations & Speeches
• Reports & Papers
• Quarterly Journal: International Security
• Artificial Intelligence
• Conflict & Conflict Resolution
• Coronavirus
• Economics & Global Affairs
• Environment & Climate Change
• International Relations
• International Security & Defense
• Nuclear Issues
• Science & Technology
• Student Publications
• War in Ukraine
• Asia & the Pacific
• Middle East & North Africa
• North America
• South America
• Infographics & Charts

US-Russian Contention in Cyberspace

The overarching question imparting urgency to this exploration is: Can U.S.-Russian contention in cyberspace cause the two nuclear superpowers to stumble into war? In considering this question we were constantly reminded of recent comments by a prominent U.S. arms control expert: At least as dangerous as the risk of an actual cyberattack, he observed, is cyber operations’ “blurring of the line between peace and war.” Or, as Nye wrote, “in the cyber realm, the difference between a weapon and a non-weapon may come down to a single line of code, or simply the intent of a computer program’s user.”

The Geopolitics of Renewable Hydrogen

Renewables are widely perceived as an opportunity to shatter the hegemony of fossil fuel-rich states and democratize the energy landscape. Virtually all countries have access to some renewable energy resources (especially solar and wind power) and could thus substitute foreign supply with local resources. Our research shows, however, that the role countries are likely to assume in decarbonized energy systems will be based not only on their resource endowment but also on their policy choices.

What Comes After the Forever Wars

As the United States emerges from the era of so-called forever wars, it should abandon the regime change business for good. Then, Washington must understand why it failed, writes Stephen Walt.

Telling Black Stories: What We All Can Do

Full event video and after-event thoughts from the panelists.

• Defense, Emerging Technology, and Strategy
• Diplomacy and International Politics
• Environment and Natural Resources
• International Security
• Science, Technology, and Public Policy
• Africa Futures Project
• Applied History Project
• Arctic Initiative
• Asia-Pacific Initiative
• Cyber Project
• Defending Digital Democracy
• Defense Project
• Economic Diplomacy Initiative
• Future of Diplomacy Project
• Geopolitics of Energy Project
• Harvard Project on Climate Agreements
• Homeland Security Project
• Intelligence Project
• Korea Project
• Managing the Atom
• Middle East Initiative
• Project on Europe and the Transatlantic Relationship
• Security and Global Health
• Technology and Public Purpose
• US-Russia Initiative to Prevent Nuclear Terrorism

Special Initiatives

• American Secretaries of State
• An Economic View of the Environment  
• Cuban Missile Crisis  
• Russia Matters
• Thucydides's Trap

Analysis & Opinions - O'Reilly Media

• Mike Loukidos
• Hilary Mason

These studies provide a foundation for discussing ethical issues so we can better integrate data ethics in real life.

To help us think seriously about data ethics, we need case studies that we can discuss, argue about, and come to terms with as we engage with the real world. Good case studies give us the opportunity to think through problems before facing them in real life. And case studies show us that ethical problems aren't simple. They are multi-faceted, and frequently there's no single right answer. And they help us to recognize there are few situations that don't raise ethical questions.

Princeton's  Center for Information Technology Policy  and  Center for Human Values  have created four anonymized  case studies  to promote the discussion of ethics. The first of these studies,  Automated Healthcare App , discusses a smartphone app designed to help adult onset diabetes patients. It raises issues like paternalism, consent, and even language choices. Is it OK to “nudge” patients toward more healthy behaviors? What about automatically moderating the users’ discussion groups to emphasize scientifically accurate information? And how do you deal with minorities who don’t respond to treatment as well? Could the problem be the language itself that is used to discuss treatment?

The next case study,  Dynamic Sound Identification , covers an application that can identify voices, raising issues about privacy, language, and even gender. How far should developers go in identifying potential harm that can be caused by an application? What are acceptable error rates for an application that can potentially do harm? How can a voice application handle people with different accents or dialects? And what responsibility do developers have when a small experimental tool is bought by a large corporation that wants to commercialize it?

The  Optimizing Schools  case study deals with the problem of finding at-risk children in school systems. Privacy and language are again an issue; it also raises the issue of how decisions to use data are made. Who makes those decisions, and who needs to be informed about them? What are the consequences when people find out how their data has been used? And how do you interpret the results of an experiment? Under what conditions can you say that a data experiment has really yielded improved educational results?

The final case study,  Law Enforcement Chatbots , raises issues about the tradeoff between liberty and security, entrapment, openness and accountability, and compliance with international law.

None of these issues are simple, and there are few (if any) "right answers." For example, it’s easy to react against perceived paternalism in a medical application, but the purpose of such an application is to encourage patients to comply with their treatment program. It’s easy to object to monitoring students in a public school, but students are minors, and schools by nature handle a lot of private personal data. Where is the boundary between what is, and isn’t, acceptable? What's important isn’t getting to the correct answer on any issue, but to make sure the issue is discussed and understood, and that we know what tradeoffs we are making. What is important is that we get practice in discussing ethical issues and put that practice to work in our jobs. That’s what these case studies give us.

Want to Read More?

The authors.

• Senior Fellow, Technology and Public Purpose Project
• Former Senior Fellow, Cyber Project
• Former U.S Chief Data Scientist
• Former CTO, Devoted Health
• Bio/Profile
• More by this author

Recommended

In the spotlight, most viewed.

Journal Article - Issues in Science and Technology

Nuclear Power Needs Leadership, but Not from the Military

• Michael J Ford
• Ahmed Abdulla
• M. Granger Morgan

Analysis & Opinions - The Washington Post

Don't Fear the TSA Cutting Airport Security. Be Glad That They're Talking about It.

• Bruce Schneier

Data's Day of Reckoning

AI Agents in Diplomacy – Promoting Stability or Chaos?

• Gerald M. McMahon

The Day-After Peace in Gaza Will be Fragile. Here’s How to Make it Work.

• Shady ElGhazaly Harb
• Sharan Grewal

Russian Wargame Practicing Tactical Nukes Use Is Warning to West

• Simon Saradzhyan

Analysis & Opinions - Belfer Center for Science and International Affairs, Harvard Kennedy School

Seeing Like a Data Structure

• Barath Raghavan

Policy Brief - Belfer Center for Science and International Affairs, Harvard Kennedy School

How Multimodal AI Could Retool Global Crisis Response

• Ben Ellencweig
• Jessica Lamb
• Mihir Mysore
• Jesse Salazar
• Juliette Kayyem
• Nate Bruggeman

Analysis & Opinions - New Straits Times

Gorbachev and the End of the Cold War

• Joseph S. Nye

Belfer Center Email Updates

Belfer center of science and international affairs.

79 John F. Kennedy Street, Cambridge, MA 02138 (617) 495-1400

• Clinical Ethics Services
• Ethical AI Services
• Custom Workshops
• Medical Student Education
• In the News
• Impact Videos and Stories
• Frequently Asked Questions
• Policies, Disclosures and Reports

Explore Our Resources

Search for case studies, audio interviews, videos, and more.

Search for:

Clear Filters

Filter by Type

• All Categories
• Case Studies
• Medical Ethics and Policy Guidance
• Shared Decision Making and Advance Care Planning

Filter by Category

• Ethical Theory and Principles (23)
• Informed Consent (6)
• Medical Ethics (37)
• Resource Allocation (10)
• Patient/Physician Relationship (43)
• Pediatrics (3)
• Public and Population Health (71)
• Religion and Morality (14)
• Advance Care Planning (42)
• End of Life Ethics (81)
• Biotechnologies and Genomics (1)
• Procreation and Reproduction (6)
• Research Ethics (17)

Case Study – Moral Culpability for Respecting Patients’ Autonomy

Case Study – Decisional Capacity of the Patient’s Surrogate

Case Study – A Multidisciplinary Healthcare Team Disagrees

Case Study – Whose Decision?

Case Study –Too little, too late… almost

Case Study – Trying to Honor Johnny’s Wishes

Case Study – “God will restore his leg. The doctors will see.” Patient Nonadherence.

Case Study – Moral Dimensions of Medical Negligence

Case Study – Moral Distress of Hospital Workers

Doha Declaration

Education for justice.

• Agenda Day 1
• Agenda Day 2
• Agenda Day 3
• Agenda Day 4
• Registration
• Breakout Sessions for Primary and Secondary Level
• Breakout Sessions for Tertiary Level
• E4J Youth Competition
• India - Lockdown Learners
• Chuka, Break the Silence
• The Online Zoo
• I would like a community where ...
• Staying safe online
• Let's be respectful online
• We can all be heroes
• Respect for all
• We all have rights
• A mosaic of differences
• The right thing to do
• Solving ethical dilemmas
• UNODC-UNESCO Guide for Policymakers
• UNODC-UNESCO Handbooks for Teachers
• Justice Accelerators
• Introduction
• Organized Crime
• Trafficking in Persons & Smuggling of Migrants
• Crime Prevention & Criminal Justice Reform
• Crime Prevention, Criminal Justice & SDGs
• UN Congress on Crime Prevention & Criminal Justice
• Commission on Crime Prevention & Criminal Justice
• Conference of the Parties to UNTOC
• Conference of the States Parties to UNCAC
• Rules for Simulating Crime Prevention & Criminal Justice Bodies
• Crime Prevention & Criminal Justice
• Engage with Us
• Contact Us about MUN
• Conferences Supporting E4J
• Cyberstrike
• Play for Integrity
• Running out of Time
• Zorbs Reloaded
• Developing a Rationale for Using the Video
• Previewing the Anti-Corruption Video
• Viewing the Video with a Purpose
• Post-viewing Activities
• Previewing the Firearms Video
• Rationale for Using the Video
• Previewing the Human Trafficking Video
• Previewing the Organized Crime Video
• Previewing the Video
• Criminal Justice & Crime Prevention
• Corruption & Integrity
• Human Trafficking & Migrant Smuggling
• Firearms Trafficking
• Terrorism & Violent Extremism
• Introduction & Learning Outcomes
• Corruption - Baseline Definition
• Effects of Corruption
• Deeper Meanings of Corruption
• Measuring Corruption
• Possible Class Structure
• Core Reading
• Advanced Reading
• Student Assessment
• Additional Teaching Tools
• Guidelines for Stand-Alone Course
• Appendix: How Corruption Affects the SDGs
• What is Governance?
• What is Good Governance?
• Corruption and Bad Governance
• Governance Reforms and Anti-Corruption
• Guidelines for Stand-alone Course
• Corruption and Democracy
• Corruption and Authoritarian Systems
• Hybrid Systems and Syndromes of Corruption
• The Deep Democratization Approach
• Political Parties and Political Finance
• Political Institution-building as a Means to Counter Corruption
• Manifestations and Consequences of Public Sector Corruption
• Causes of Public Sector Corruption
• Theories that Explain Corruption
• Corruption in Public Procurement
• Corruption in State-Owned Enterprises
• Responses to Public Sector Corruption
• Preventing Public Sector Corruption
• Forms & Manifestations of Private Sector Corruption
• Consequences of Private Sector Corruption
• Causes of Private Sector Corruption
• Responses to Private Sector Corruption
• Preventing Private Sector Corruption
• Collective Action & Public-Private Partnerships against Corruption
• Transparency as a Precondition
• Detection Mechanisms - Auditing and Reporting
• Whistle-blowing Systems and Protections
• Investigation of Corruption
• Introduction and Learning Outcomes
• Brief background on the human rights system
• Overview of the corruption-human rights nexus
• Impact of corruption on specific human rights
• Approaches to assessing the corruption-human rights nexus
• Human-rights based approach
• Defining sex, gender and gender mainstreaming
• Gender differences in corruption
• Theories explaining the gender–corruption nexus
• Gendered impacts of corruption
• Anti-corruption and gender mainstreaming
• Manifestations of corruption in education
• Costs of corruption in education
• Causes of corruption in education
• Fighting corruption in education
• Core terms and concepts
• The role of citizens in fighting corruption
• The role, risks and challenges of CSOs fighting corruption
• The role of the media in fighting corruption
• Access to information: a condition for citizen participation
• ICT as a tool for citizen participation in anti-corruption efforts
• Government obligations to ensure citizen participation in anti-corruption efforts
• Teaching Guide
• Brief History of Terrorism
• 19th Century Terrorism
• League of Nations & Terrorism
• United Nations & Terrorism
• Terrorist Victimization
• Exercises & Case Studies
• Radicalization & Violent Extremism
• Preventing & Countering Violent Extremism
• Drivers of Violent Extremism
• International Approaches to PVE &CVE
• Regional & Multilateral Approaches
• Defining Rule of Law
• UN Global Counter-Terrorism Strategy
• International Cooperation & UN CT Strategy
• Legal Sources & UN CT Strategy
• Regional & National Approaches
• International Legal Frameworks
• International Human Rights Law
• International Humanitarian Law
• International Refugee Law
• Current Challenges to International Legal Framework
• Defining Terrorism
• Criminal Justice Responses
• Treaty-based Crimes of Terrorism
• Core International Crimes
• International Courts and Tribunals
• African Region
• Inter-American Region
• Asian Region
• European Region
• Middle East & Gulf Regions
• Core Principles of IHL
• Categorization of Armed Conflict
• Classification of Persons
• IHL, Terrorism & Counter-Terrorism
• Relationship between IHL & intern. human rights law
• Limitations Permitted by Human Rights Law
• Derogation during Public Emergency
• Examples of States of Emergency & Derogations
• International Human Rights Instruments
• Regional Human Rights Instruments
• Extra-territorial Application of Right to Life
• Arbitrary Deprivation of Life
• Death Penalty
• Enforced Disappearances
• Armed Conflict Context
• International Covenant on Civil and Political Rights
• Convention against Torture et al.
• International Legal Framework
• Key Contemporary Issues
• Investigative Phase
• Trial & Sentencing Phase
• Armed Conflict
• Case Studies
• Special Investigative Techniques
• Surveillance & Interception of Communications
• Privacy & Intelligence Gathering in Armed Conflict
• Accountability & Oversight of Intelligence Gathering
• Principle of Non-Discrimination
• Freedom of Religion
• Freedom of Expression
• Freedom of Assembly
• Freedom of Association
• Fundamental Freedoms
• Definition of 'Victim'
• Effects of Terrorism
• Access to Justice
• Recognition of the Victim
• Human Rights Instruments
• Criminal Justice Mechanisms
• Instruments for Victims of Terrorism
• National Approaches
• Key Challenges in Securing Reparation
• Topic 1. Contemporary issues relating to conditions conducive both to the spread of terrorism and the rule of law
• Topic 2. Contemporary issues relating to the right to life
• Topic 3. Contemporary issues relating to foreign terrorist fighters
• Topic 4. Contemporary issues relating to non-discrimination and fundamental freedoms
• Module 16: Linkages between Organized Crime and Terrorism
• Thematic Areas
• Content Breakdown
• Module Adaptation & Design Guidelines
• Teaching Methods
• Acknowledgements
• 1. Introducing United Nations Standards & Norms on CPCJ vis-à-vis International Law
• 2. Scope of United Nations Standards & Norms on CPCJ
• 3. United Nations Standards & Norms on CPCJ in Operation
• 1. Definition of Crime Prevention
• 2. Key Crime Prevention Typologies
• 2. (cont.) Tonry & Farrington’s Typology
• 3. Crime Problem-Solving Approaches
• 4. What Works
• United Nations Entities
• Regional Crime Prevention Councils/Institutions
• Key Clearinghouses
• Systematic Reviews
• 1. Introduction to International Standards & Norms
• 2. Identifying the Need for Legal Aid
• 3. Key Components of the Right of Access to Legal Aid
• 4. Access to Legal Aid for Those with Specific Needs
• 5. Models for Governing, Administering and Funding Legal Aid
• 6. Models for Delivering Legal Aid Services
• 7. Roles and Responsibilities of Legal Aid Providers
• 8. Quality Assurance and Legal Aid Services
• 1. Context for Use of Force by Law Enforcement Officials
• 2. Legal Framework
• 3. General Principles of Use of Force in Law Enforcement
• 4. Use of Firearms
• 5. Use of “Less-Lethal” Weapons
• 6. Protection of Especially Vulnerable Groups
• 7. Use of Force during Assemblies
• 1. Policing in democracies & need for accountability, integrity, oversight
• 2. Key mechanisms & actors in police accountability, oversight
• 3. Crosscutting & contemporary issues in police accountability
• 1. Introducing Aims of Punishment, Imprisonment & Prison Reform
• 2. Current Trends, Challenges & Human Rights
• 3. Towards Humane Prisons & Alternative Sanctions
• 1. Aims and Significance of Alternatives to Imprisonment
• 2. Justifying Punishment in the Community
• 3. Pretrial Alternatives
• 4. Post Trial Alternatives
• 5. Evaluating Alternatives
• 1. Concept, Values and Origin of Restorative Justice
• 2. Overview of Restorative Justice Processes
• 3. How Cost Effective is Restorative Justice?
• 4. Issues in Implementing Restorative Justice
• 1. Gender-Based Discrimination & Women in Conflict with the Law
• 2. Vulnerabilities of Girls in Conflict with the Law
• 3. Discrimination and Violence against LGBTI Individuals
• 4. Gender Diversity in Criminal Justice Workforce
• 1. Ending Violence against Women
• 2. Human Rights Approaches to Violence against Women
• 3. Who Has Rights in this Situation?
• 4. What about the Men?
• 5. Local, Regional & Global Solutions to Violence against Women & Girls
• 1. Understanding the Concept of Victims of Crime
• 2. Impact of Crime, including Trauma
• 3. Right of Victims to Adequate Response to their Needs
• 4. Collecting Victim Data
• 5. Victims and their Participation in Criminal Justice Process
• 6. Victim Services: Institutional and Non-Governmental Organizations
• 7. Outlook on Current Developments Regarding Victims
• 8. Victims of Crime and International Law
• 1. The Many Forms of Violence against Children
• 2. The Impact of Violence on Children
• 3. States' Obligations to Prevent VAC and Protect Child Victims
• 4. Improving the Prevention of Violence against Children
• 5. Improving the Criminal Justice Response to VAC
• 6. Addressing Violence against Children within the Justice System
• 1. The Role of the Justice System
• 2. Convention on the Rights of the Child & International Legal Framework on Children's Rights
• 3. Justice for Children
• 4. Justice for Children in Conflict with the Law
• 5. Realizing Justice for Children
• 1a. Judicial Independence as Fundamental Value of Rule of Law & of Constitutionalism
• 1b. Main Factors Aimed at Securing Judicial Independence
• 2a. Public Prosecutors as ‘Gate Keepers’ of Criminal Justice
• 2b. Institutional and Functional Role of Prosecutors
• 2c. Other Factors Affecting the Role of Prosecutors
• Basics of Computing
• Global Connectivity and Technology Usage Trends
• Cybercrime in Brief
• Cybercrime Trends
• Cybercrime Prevention
• Offences against computer data and systems
• Computer-related offences
• Content-related offences
• The Role of Cybercrime Law
• Harmonization of Laws
• International and Regional Instruments
• International Human Rights and Cybercrime Law
• Digital Evidence
• Digital Forensics
• Standards and Best Practices for Digital Forensics
• Reporting Cybercrime
• Who Conducts Cybercrime Investigations?
• Obstacles to Cybercrime Investigations
• Knowledge Management
• Legal and Ethical Obligations
• Handling of Digital Evidence
• Digital Evidence Admissibility
• Sovereignty and Jurisdiction
• Formal International Cooperation Mechanisms
• Informal International Cooperation Mechanisms
• Data Retention, Preservation and Access
• Challenges Relating to Extraterritorial Evidence
• National Capacity and International Cooperation
• Internet Governance
• Cybersecurity Strategies: Basic Features
• National Cybersecurity Strategies
• International Cooperation on Cybersecurity Matters
• Cybersecurity Posture
• Assets, Vulnerabilities and Threats
• Vulnerability Disclosure
• Cybersecurity Measures and Usability
• Situational Crime Prevention
• Incident Detection, Response, Recovery & Preparedness
• Privacy: What it is and Why it is Important
• Privacy and Security
• Cybercrime that Compromises Privacy
• Data Protection Legislation
• Data Breach Notification Laws
• Enforcement of Privacy and Data Protection Laws
• Intellectual Property: What it is
• Types of Intellectual Property
• Causes for Cyber-Enabled Copyright & Trademark Offences
• Protection & Prevention Efforts
• Online Child Sexual Exploitation and Abuse
• Cyberstalking and Cyberharassment
• Cyberbullying
• Gender-Based Interpersonal Cybercrime
• Interpersonal Cybercrime Prevention
• Cyber Organized Crime: What is it?
• Conceptualizing Organized Crime & Defining Actors Involved
• Criminal Groups Engaging in Cyber Organized Crime
• Cyber Organized Crime Activities
• Preventing & Countering Cyber Organized Crime
• Cyberespionage
• Cyberterrorism
• Cyberwarfare
• Information Warfare, Disinformation & Electoral Fraud
• Responses to Cyberinterventions
• Framing the Issue of Firearms
• Direct Impact of Firearms
• Indirect Impacts of Firearms on States or Communities
• International and National Responses
• Typology and Classification of Firearms
• Common Firearms Types
• 'Other' Types of Firearms
• Parts and Components
• History of the Legitimate Arms Market
• Need for a Legitimate Market
• Key Actors in the Legitimate Market
• Authorized & Unauthorized Arms Transfers
• Illegal Firearms in Social, Cultural & Political Context
• Supply, Demand & Criminal Motivations
• Larger Scale Firearms Trafficking Activities
• Smaller Scale Trafficking Activities
• Sources of Illicit Firearms
• Consequences of Illicit Markets
• International Public Law & Transnational Law
• International Instruments with Global Outreach
• Commonalities, Differences & Complementarity between Global Instruments
• Tools to Support Implementation of Global Instruments
• Other United Nations Processes
• The Sustainable Development Goals
• Multilateral & Regional Instruments
• Scope of National Firearms Regulations
• National Firearms Strategies & Action Plans
• Harmonization of National Legislation with International Firearms Instruments
• Assistance for Development of National Firearms Legislation
• Firearms Trafficking as a Cross-Cutting Element
• Organized Crime and Organized Criminal Groups
• Criminal Gangs
• Terrorist Groups
• Interconnections between Organized Criminal Groups & Terrorist Groups
• Gangs - Organized Crime & Terrorism: An Evolving Continuum
• International Response
• International and National Legal Framework
• Firearms Related Offences
• Role of Law Enforcement
• Firearms as Evidence
• Use of Special Investigative Techniques
• International Cooperation and Information Exchange
• Prosecution and Adjudication of Firearms Trafficking
• Teaching Methods & Principles
• Ethical Learning Environments
• Overview of Modules
• Module Adaption & Design Guidelines
• Table of Exercises
• Basic Terms
• Forms of Gender Discrimination
• Ethics of Care
• Case Studies for Professional Ethics
• Case Studies for Role Morality
• Additional Exercises
• Defining Organized Crime
• Definition in Convention
• Similarities & Differences
• Activities, Organization, Composition
• Thinking Critically Through Fiction
• Excerpts of Legislation
• Research & Independent Study Questions
• Legal Definitions of Organized Crimes
• Criminal Association
• Definitions in the Organized Crime Convention
• Criminal Organizations and Enterprise Laws
• Enabling Offence: Obstruction of Justice
• Drug Trafficking
• Wildlife & Forest Crime
• Counterfeit Products Trafficking
• Falsified Medical Products
• Trafficking in Cultural Property
• Trafficking in Persons
• Case Studies & Exercises
• Extortion Racketeering
• Loansharking
• Links to Corruption
• Bribery versus Extortion
• Money-Laundering
• Liability of Legal Persons
• How much Organized Crime is there?
• Alternative Ways for Measuring
• Measuring Product Markets
• Risk Assessment
• Key Concepts of Risk Assessment
• Risk Assessment of Organized Crime Groups
• Risk Assessment of Product Markets
• Risk Assessment in Practice
• Positivism: Environmental Influences
• Classical: Pain-Pleasure Decisions
• Structural Factors
• Ethical Perspective
• Crime Causes & Facilitating Factors
• Models and Structure
• Hierarchical Model
• Local, Cultural Model
• Enterprise or Business Model
• Groups vs Activities
• Networked Structure
• Jurisdiction
• Investigators of Organized Crime
• Controlled Deliveries
• Physical & Electronic Surveillance
• Undercover Operations
• Financial Analysis
• Use of Informants
• Rights of Victims & Witnesses
• Role of Prosecutors
• Adversarial vs Inquisitorial Legal Systems
• Mitigating Punishment
• Granting Immunity from Prosecution
• Witness Protection
• Aggravating & Mitigating Factors
• Sentencing Options
• Alternatives to Imprisonment
• Death Penalty & Organized Crime
• Backgrounds of Convicted Offenders
• Confiscation
• Confiscation in Practice
• Mutual Legal Assistance (MLA)
• Extradition
• Transfer of Criminal Proceedings
• Transfer of Sentenced Persons
• Module 12: Prevention of Organized Crime
• Adoption of Organized Crime Convention
• Historical Context
• Features of the Convention
• Related international instruments
• Conference of the Parties
• Roles of Participants
• Structure and Flow
• Recommended Topics
• Background Materials
• What is Sex / Gender / Intersectionality?
• Knowledge about Gender in Organized Crime
• Gender and Organized Crime
• Gender and Different Types of Organized Crime
• Definitions and Terminology
• Organized crime and Terrorism - International Legal Framework
• International Terrorism-related Conventions
• UNSC Resolutions on Terrorism
• Organized Crime Convention and its Protocols
• Theoretical Frameworks on Linkages between Organized Crime and Terrorism
• Typologies of Criminal Behaviour Associated with Terrorism
• Terrorism and Drug Trafficking
• Terrorism and Trafficking in Weapons
• Terrorism, Crime and Trafficking in Cultural Property
• Trafficking in Persons and Terrorism
• Intellectual Property Crime and Terrorism
• Kidnapping for Ransom and Terrorism
• Exploitation of Natural Resources and Terrorism
• Review and Assessment Questions
• Research and Independent Study Questions
• Criminalization of Smuggling of Migrants
• UNTOC & the Protocol against Smuggling of Migrants
• Offences under the Protocol
• Financial & Other Material Benefits
• Aggravating Circumstances
• Criminal Liability
• Non-Criminalization of Smuggled Migrants
• Scope of the Protocol
• Humanitarian Exemption
• Migrant Smuggling v. Irregular Migration
• Migrant Smuggling vis-a-vis Other Crime Types
• Other Resources
• Assistance and Protection in the Protocol
• International Human Rights and Refugee Law
• Vulnerable groups
• Positive and Negative Obligations of the State
• Identification of Smuggled Migrants
• Participation in Legal Proceedings
• Role of Non-Governmental Organizations
• Smuggled Migrants & Other Categories of Migrants
• Short-, Mid- and Long-Term Measures
• Criminal Justice Reponse: Scope
• Investigative & Prosecutorial Approaches
• Different Relevant Actors & Their Roles
• Testimonial Evidence
• Financial Investigations
• Non-Governmental Organizations
• ‘Outside the Box’ Methodologies
• Intra- and Inter-Agency Coordination
• Admissibility of Evidence
• International Cooperation
• Exchange of Information
• Non-Criminal Law Relevant to Smuggling of Migrants
• Administrative Approach
• Complementary Activities & Role of Non-criminal Justice Actors
• Macro-Perspective in Addressing Smuggling of Migrants
• Human Security
• International Aid and Cooperation
• Migration & Migrant Smuggling
• Mixed Migration Flows
• Social Politics of Migrant Smuggling
• Vulnerability
• Profile of Smugglers
• Role of Organized Criminal Groups
• Humanitarianism, Security and Migrant Smuggling
• Crime of Trafficking in Persons
• The Issue of Consent
• The Purpose of Exploitation
• The abuse of a position of vulnerability
• Indicators of Trafficking in Persons
• Distinction between Trafficking in Persons and Other Crimes
• Misconceptions Regarding Trafficking in Persons
• Root Causes
• Supply Side Prevention Strategies
• Demand Side Prevention Strategies
• Role of the Media
• Safe Migration Channels
• Crime Prevention Strategies
• Monitoring, Evaluating & Reporting on Effectiveness of Prevention
• Trafficked Persons as Victims
• Protection under the Protocol against Trafficking in Persons
• Broader International Framework
• State Responsibility for Trafficking in Persons
• Identification of Victims
• Principle of Non-Criminalization of Victims
• Criminal Justice Duties Imposed on States
• Role of the Criminal Justice System
• Current Low Levels of Prosecutions and Convictions
• Challenges to an Effective Criminal Justice Response
• Rights of Victims to Justice and Protection
• Potential Strategies to “Turn the Tide”
• State Cooperation with Civil Society
• Civil Society Actors
• The Private Sector
• Comparing SOM and TIP
• Differences and Commonalities
• Vulnerability and Continuum between SOM & TIP
• Labour Exploitation
• Forced Marriage
• Other Examples
• Children on the Move
• Protecting Smuggled and Trafficked Children
• Protection in Practice
• Children Alleged as Having Committed Smuggling or Trafficking Offences
• Basic Terms - Gender and Gender Stereotypes
• International Legal Frameworks and Definitions of TIP and SOM
• Global Overview on TIP and SOM
• Gender and Migration
• Key Debates in the Scholarship on TIP and SOM
• Gender and TIP and SOM Offenders
• Responses to TIP and SOM
• Use of Technology to Facilitate TIP and SOM
• Technology Facilitating Trafficking in Persons
• Technology in Smuggling of Migrants
• Using Technology to Prevent and Combat TIP and SOM
• Privacy and Data Concerns
• Emerging Trends
• Demand and Consumption
• Supply and Demand
• Implications of Wildlife Trafficking
• Legal and Illegal Markets
• Perpetrators and their Networks
• Locations and Activities relating to Wildlife Trafficking
• Environmental Protection & Conservation
• CITES & the International Trade in Endangered Species
• Organized Crime & Corruption
• Animal Welfare
• Criminal Justice Actors and Agencies
• Criminalization of Wildlife Trafficking
• Challenges for Law Enforcement
• Investigation Measures and Detection Methods
• Prosecution and Judiciary
• Wild Flora as the Target of Illegal Trafficking
• Purposes for which Wild Flora is Illegally Targeted
• How is it Done and Who is Involved?
• Consequences of Harms to Wild Flora
• Terminology
• Background: Communities and conservation: A history of disenfranchisement
• Incentives for communities to get involved in illegal wildlife trafficking: the cost of conservation
• Incentives to participate in illegal wildlife, logging and fishing economies
• International and regional responses that fight wildlife trafficking while supporting IPLCs
• Mechanisms for incentivizing community conservation and reducing wildlife trafficking
• Critiques of community engagement
• Other challenges posed by wildlife trafficking that affect local populations
• Global Podcast Series
• Apr. 2021: Call for Expressions of Interest: Online training for academics from francophone Africa
• Feb. 2021: Series of Seminars for Universities of Central Asia
• Dec. 2020: UNODC and TISS Conference on Access to Justice to End Violence
• Nov. 2020: Expert Workshop for University Lecturers and Trainers from the Commonwealth of Independent States
• Oct. 2020: E4J Webinar Series: Youth Empowerment through Education for Justice
• Interview: How to use E4J's tool in teaching on TIP and SOM
• E4J-Open University Online Training-of-Trainers Course
• Teaching Integrity and Ethics Modules: Survey Results
• Grants Programmes
• E4J MUN Resource Guide
• Library of Resources
• Integrity & Ethics
• Module 12: Integrity, Ethics & Law
• {{item.name}} ({{item.items.length}}) items
• Add new list

University Module Series: Integrity & Ethics

Module 12: integrity, ethics and law.

  This module is a resource for lecturers  

Case studies.

Choose one or more of the following case studies and lead a discussion which allows students to address and debate issues of integrity, ethics and law. If time allows, let the students vote on which case studies they want to discuss.

For lecturers teaching large classes, case studies with multiple parts and different methods of solution lend themselves well to the group size and energy in such an environment. Lecturers can begin by having students vote on which case study they prefer. Lecturers could break down analysis of the chosen case study into steps which appear to students in sequential order, thereby ensuring that larger groups stay on track. Lecturers may instruct students to discuss questions in a small group without moving from their seat, and nominate one person to speak for the group if called upon. There is no need to provide excessive amounts of time for group discussion, as ideas can be developed further with the class as a whole. Lecturers can vary the group they call upon to encourage responsive participation.

Back to top

Supported by the state of qatar, 60 years crime congress.

• Skip to Content
• Skip to Main Navigation
• Skip to Search

Indiana University Bloomington Indiana University Bloomington IU Bloomington

• “Ad”mission of guilt
• “Do I stop him?”
• Newspaper joins war against drugs
• Have I got a deal for you!
• Identifying what’s right
• Is “Enough!” too much?
• Issues of bench and bar
• Knowing when to say “when!”
• Stop! This is a warning…
• Strange bedfellows
• Gambling with being first
• Making the right ethical choice can mean winning by losing
• Playing into a hoaxster’s hands
• “They said it first”
• Is it news, ad or informercial?
• Letter to the editor
• Games publishers play
• An offer you can refuse
• An oily gift horse
• Public service . . . or “news-mercials”
• As life passes by
• Bringing death close
• A careless step, a rash of calls
• Distortion of reality?
• Of life and death
• Naked came the rider
• “A photo that had to be used”
• A picture of controversy
• Freedom of political expression
• Brother, can you spare some time?
• Columnist’s crusade OK with Seattle
• Kiss and tell
• The making of a govenor
• Past but not over
• Of publishers and politics
• To tell the truth
• Truth & Consequences
• “Truth boxes”
• When journalists become flacks
• A book for all journalists who believe
• The Billboard Bandit
• Food for thought
• Grand jury probe
• Judgement on journalists
• Lessons from an ancient spirit
• Lying for the story . . .
• Newspaper nabs Atlanta’s Dahmer
• One way to a good end
• Over the fence
• “Psst! Pass it on!”
• Rules aren’t neat on Crack Street
• “Someone had to be her advocate”
• Trial by Fire
• Trial by proximity
• Using deceit to get the truth
• When advocacy is okay
• Witness to an execution
• Are we our brother’s keeper? . . . You bet we are!
• Betraying a trust
• Broken promise
• “But I thought you were . . . ”
• “Can I take it back?”
• Competitive disadvantage
• Getting it on tape
• The great quote question
• How to handle suicide threats
• Let’s make a deal!
• A phone-y issue?
• The source wanted out
• The story that died in a lie
• Thou shalt not break thy promise
• Thou shalt not concoct thy quote
• Thou shalt not trick thy source
• Too good to be true
• Vulnerable sources and journalistic responsibility
• The way things used to be . . .
• When a story just isn’t worth it
• When a story source threatens suicide
• When public should remain private
• The ethics of “outing”
• “For personal reasons”
• Intruding on grief
• Intruding on private pain
• Privacy case settled against TV station
• Seeing both sides
• Two views on “outing”
• Unwanted spotlight
• Whose right is it anyway?
• Other views on the Christine Busalacchi case
• The death of a soldier
• Firing at Round Rock
• A kinder, gentler news media
• Operation: Buy yourself a parade
• Rallying ’round the flag
• “Salute to military” ads canceled
• Tell the truth, stay alive
• The windbags of war
• Absent with no malice
• Anonymity for rape victims . . .
• An exception to the rule
• The boy with a broken heart
• Civilly suitable
• Creating a victim
• “Everyone already knew”
• An exceptional case
• Innocent victims
• Minor infraction
• Names make news
• Naming a victim
• Naming “johns”
• Profile of controversy
• What the media all missed
• Punishing plagiarizers
• Sounding an alarm on AIDS
• Suffer the children
• Anchor’s away
• The day the earth stood still
• Doing your own ethics audit
• Good guys, bad guys and TV news
• Is it just me, or . . . ?
• The Post’s exam answer story
• TV station “teases” suicide
• Yanking Doonesbury
• The year in review
• Colorado media’s option play
• Deadly lesson
• Deciding which critically ill person gets coverage
• When journalists play God . . .
• A delicate balance
• The Fallen Servant
• Handle with care
• It’s the principle, really
• Killing news
• Maybe what seems so right is wrong
• On the line
• Protest and apology after Daily Beacon story
• Red flag for badgering
• Sharing the community’s grief
• The “super-crip” stereotype
• “And then he said *&%*!!!”
• When big is not better
• When the KKK comes calling
• Not the straight story
• Agreeing to disagree
• All in the family
• Family feud
• Author! Author!
• The Bee that roared
• Brewing controversy
• Building barriers
• Other views from librarians
• The ethics of information selling
• Close to home
• Family ties
• How now, sacred cow?
• The ties that bind
• “Like any other story”
• When your newspaper is the news
• Not friendly fire
• Overdraft on credibility?
• The problem is the writing
• Written rules can be hazardous
• Project censored, sins of omission and the hardest “W” of all – “why”
• Risking the newsroom’s image
• The Media School

Ethics Case Studies

Ethics cases online.

This set of cases has been created for teachers, researchers, professional journalists and consumers of news to help them explore ethical issues in journalism. The cases raise a variety of ethical problems faced by journalists, including such issues as privacy, conflict of interest, reporter- source relationships, and the role of journalists in their communities.

The initial core of this database comes from a series of cases developed by Barry Bingham, Jr., and published in his newsletter, FineLine. The school is grateful to Bingham for his permission to make these cases available to a wider audience.

You may download cases for classes, research or personal use. Permission is granted for academic use of these cases, including inclusion in course readers for specific college courses. This permission does not extend to the republication of the cases in books, journals or electronic form.

Note: We are indebted to Professor Emeritus David Boeyink, who developed this project several years ago.

Aiding law enforcement

• “Ad”mission of guilt: Court-ordered ads raise ethical questions
• “Do I stop him?”: Reporter’s arresting question is news
• Fairness: A casualty of the anti-drug crusade
• Newspaper joins war against drugs: Standard-Times publishes photos of all suspected drug offenders
• Have I got a deal for you!: The line between cooperation and collusion
• Identifying what’s right: Photographer’s ID used in hostage release
• Is “Enough!” too much?: Editors split on anti-drug coupons
• Issues of bench and bar: In this case, a TV reporter is the judge
• Knowing when to say “when!”: Drawing the line at cooperating with authorities
• Stop! This is a warning . . . : Suppressing news at police request
• Strange Bedfellows: Federal agents in a TV newsroom

Being first

• Gambling with being first: The media drive to score on the Isiah Thomas story
• Playing into a hoaxster’s hands: How the Virginia media got suckered
• “They said it first”: Is that reason for going for the story?

Bottom-line decisions

• Is it news, ad or infomercial?: The line between news and advertising is going, going . . .
• Games publishers play: Allowing an advertiser to call the shots
• An offer you can refuse: The selling of Cybill to the Enquirer
• An oily gift horse: saying “No!” to Exxon
• Public service. . .or “news-mercials”: The blending of television news and advertising

Controversial photos

• As life passes by: A journalist’s role: watch and wait
• Bringing death close: Publishing photographs of human tragedy
• A careless step, a rash of calls: “Unusual” photo of AIDS walkathon raises hackles” 
• Distortion of reality?: “Punk for Peace” photograph draws fire
• Of life and death: Photos capture woman’s last moments
• “A photo that had to be used”: Anatomy of a newspaper’s decision
• A picture of controversy: Pulitzer photos show diverse editorial standards

Covering politics

• Freedom of political expression: Do journalists forfeit their right?
• Brother, can you spare some time?: TV stations give candidates air time
• Columnist’s crusade OK with Seattle Times
• Kiss and tell: Publishing details of a mayor’s personal life
• The making of a governor: How media fantasy swayed an election
• Past but not over: When history collides with the Present
• Of publishers and politics: Byline protest threatened at Star Tribune
• To tell the truth: Why I didn’t; why I regret it
• Truth & Consequences: The public’s right to know . . . at what cost?
• “Truth boxes”: Media monitoring of TV campaign ads
• When journalists become flacks: Two views on what to do and when to do it

Getting the story

• A book for all journalists who believe: Accuracy is our highest ethical debate
• The Billboard Bandit: Did the newspaper get graffiti on its reputation
• Food for thought: You are what you eat . . . and do
• Grand jury probe: TV journalists indicted for illegal dogfight
• Judgment on journalists: Do they defiantly put themselves “above the law?”
• Lessons from an ancient spirit: Why I participated in a peyote ritual
• Lying for the story . . . :Or things they don’t teach in journalism school
• Newspaper nabs Atlanta’s Dahmer: Another predator who should’ve been stopped: Was it homophobia?
• One way to a good end: Reporter cuts corners to test capital drug program
• Over the fence: A case of crossing the line for a story
• “Psst! Pass it on!”: Why are journalists spreading rumors?
• Rules aren’t neat on Crack Street: Journalists know the rules; they also know that the rules don’t always apply when confronted with life-threatening situations
• “Someone had to be her advocate”: A newspaper’s crusade to keep a child’s death from being forgotten
• Trial by Fire: Boy “hero” story tests media
• Trial by proximity: How close is too close for a jury and a reporter?
• Using deceit to get the truth: When there’s just no other way
• When advocacy is okay: Access is an acceptable journalist’s cause
• White lies: Bending the truth to expose injustice
• Witness to an execution: KQED sues to videotape capital punishment

Handling sources

• Are we our brother’s keeper? . . . You bet we are!
• Betraying a trust: Our story wronged a naive subject
• Broken Promise: Breaching a reporter-source confidence
• “But I thought you were . . .”: When a source doesn’t know you are a reporter
• “Can I take it back?”: Why we told our source ‘yes’
• Competitive disadvantage: Business blindsided by unnamed sources
• Getting it on tape: What if you don’t tell them?
• The great quote question: How much tampering with quotations can journalists ethically do?
• Let’s make a deal!: The dangers of trading with sources
• A phone-y issue?: Caller ID raises confidentiality questions
• The source wanted out: Why our decision was ‘no’
• The story that died in a lie: Questions about truthfulness kill publication
• Thou shalt not break thy promise: Supreme Court rules on betraying sources’ anonymity 
• Thou shalt not concoct thy quote: Supreme Court decides on the rules of the quotation game
• Thou shalt not trick thy source: Many a slip twixt the promise and the page
• Too good to be true: Blowing the whistle on a lying source
• Vulnerable sources and journalistic responsibility: Are we our brother’s keeper?
• The way things used to be . . . : Who says this new “objectivity” is better?
• When a story just isn’t worth it: Holding information to protect a good source
• When a story source threatens suicide: “I’m going to kill myself!”

Invading privacy

• The ethics of “outing”: Breaking the silence code on homosexuality
• “For personal reasons”: Balancing privacy with the right to know
• Intruding on grief: Does the public really have a “need to know?”
• Intruding on private pain: Emotional TV segment offers hard choice
• Seeing both sides: A personal and professional dilemma
• Two views on “outing”: When the media do it for you
• Two views on “outing”: When you do it yourself
• Unwanted Spotlight: When private people become part of a public story
• Whose right is it anyway?: Videotape of accident victim raises questions about rights to privacy

Military Issues

• The death of a soldier: Hometown decision for hometown hero
• Firing at Round Rock: Editor says “unpatriotic” story led to dismissal  
• A kinder, gentler news media?: Post-war coverage shows sensitivity to families
• Operation: Buy yourself a parade: New York papers pitch in for hoopla celebrating hide-and-seek war
• Rallying ’round the flag: The press as U.S. propagandists
• “Salute to military” ads canceled
• Tell the truth, stay alive: In covering a civil war, honesty is the only policy
• The windbags of war: Television’s gung-ho coverage of the Persian Gulf situation

Naming newsmakers

• Absent with no malice: Omitting part of the story for a reason
• Anonymity for rape victims . . . : should the rules change?
• An exception to the rule: a decision to change names
• The boy with a broken heart: Special problems when juveniles are newsmakers
• Civilly suitable: If law requires less, should media reveal more?
• Creating a victim: Plot for a fair story may not be foolproof
• “Everyone already knew”: A weak excuse for abandoning standards
• An exceptional case: Hartford Courant names rape victim
• Innocent victims: Naming the guilty . . . but guiltless
• Minor infraction: A newspaper’s case for breaking the law
• Names make news: One newspaper debates when and why
• Naming a victim: When do you break your own rule?
• Naming “johns”: Suicide raises ethical questions about policy
• Profile of controversy: New York Times reporter defends story on Kennedy rape claimant 
• What the media all missed: Times reporter finally sets record straight on Palm Beach rape profile
• Punishing plagiarizers: Does public exposure fit the sin?
• Sounding an alarm on AIDS: Spreading the word about someone who’s spreading the disease
• Suffer the Children: Journalists are guilty of child misuse

Other topics

• Anchor’s away: Where in the world is she? Or does it matter?
• The day the earth stood still: How the media covered the “earthquake”
• Good guys, bad guys and TV news: How television and other media promote police violence
• The Post’s exam answer story
• TV station “teases” suicide
• The year in review: 1990’s biggest ethical headaches and journalistic bloopers

Sensitive news topics

• Colorado media’s option play: Most passed; did they also fumble?
• Deadly lesson: Warning about sexual asphyxiation
• A delicate balance: Mental breakdowns & news coverage
• The Fallen Servant: When a hero is not a hero
• Handle with care: Priest murder story required extra sensitivity
• It’s the principle, really: Timing and people’s money matter, too
• Killing news: Responsible coverage of suicides
• Maybe what seems so right is wrong: A medical condition media-generated money can’t cure
• On the line: A reporter’s job vs. human decency
• Red flag for badgering: Ombudsman takes sportswriter to task
• Sharing the community’s grief: Little Rock news coverage of three teen-age suicides
• Suffer the children: Was story on molestation worth the human cost?
• The “super-crip” stereotype: Press victimization of disabled people
• “And then he said *&%*!!!”: When sexist and vulgar remarks are new
• When big is not better: Playing down a story for the community good
• When the KKK comes calling: What’s the story?
• Not the straight story: Can misleading readers ever be justified?

Workplace issues

• Agreeing to disagree: How one newspaper handles off-hour activities
• All in the family: When a journalist’s spouse creates a conflict of interests
• Family feud: Handling conflicts between journalists and partners
• Author! Author!: Ethical dilemmas when reporters turn author
• The Bee that roared: Taking a stand for editorial independence
• Brewing controversy: The commercialization of Linda Ellerbee
• Building barriers: The case against financial involvement
• Other views from librarians: When interests of client and newsroom conflict
• The ethics of information selling: Problems for library reference services
• Close to home: When your newsroom is part of the story
• Family Ties: When are relationships relationships relevant?
• How now, sacred cow?: United Way’s favored treatment by the media
• The ties that bind: Publisher’s link to United Way raises questions
• “Like any other story”: Can it be when it’s your union vs. your paper?
• When your newspaper is the news: Editors discuss their experiences
• Not friendly fire: News director at odds with CBS over story
• Overdraft on credibility?: Reporter faces conflict-of-interest charges
• Written rules can be hazardous: A lawyer views ethics codes
• Project censored, sins of omission and the hardest “W” of all – “why”
• Risking the newsroom’s image: How editors, in a good cause, can strain independence

Ethics Case Studies resources and social media channels

• SOCIETY OF PROFESSIONAL JOURNALISTS

Visit SPJ's board meeting page to download the meeting materials, get livestream/call-in information and access related resources.

Home > Ethics > Ethics Case Studies

Ethics Ethics Case Studies

The SPJ Code of Ethics is voluntarily embraced by thousands of journalists, regardless of place or platform, and is widely used in newsrooms and classrooms as a guide for ethical behavior. The code is intended not as a set of "rules" but as a resource for ethical decision-making. It is not — nor can it be under the First Amendment — legally enforceable. For an expanded explanation, please follow this link .

For journalism instructors and others interested in presenting ethical dilemmas for debate and discussion, SPJ has a useful resource. We've been collecting a number of case studies for use in workshops. The Ethics AdviceLine operated by the Chicago Headline Club and Loyola University also has provided a number of examples. There seems to be no shortage of ethical issues in journalism these days. Please feel free to use these examples in your classes, speeches, columns, workshops or other modes of communication.

Kobe Bryant’s Past: A Tweet Too Soon? On January 26, 2020, Kobe Bryant died at the age of 41 in a helicopter crash in the Los Angeles area. While the majority of social media praised Bryant after his death, within a few hours after the story broke, Felicia Sonmez, a reporter for The Washington Post , tweeted a link to an article from 2003 about the allegations of sexual assault against Bryant. The question: Is there a limit to truth-telling? How long (if at all) should a journalist wait after a person’s death before resurfacing sensitive information about their past?

A controversial apology After photographs of a speech and protests at Northwestern University appeared on the university's newspaper's website, some of the participants contacted the newspaper to complain. It became a “firestorm,” — first from students who felt victimized, and then, after the newspaper apologized, from journalists and others who accused the newspaper of apologizing for simply doing its job. The question: Is an apology the appropriate response? Is there something else the student journalists should have done?

Using the ‘Holocaust’ Metaphor People for the Ethical Treatment of Animals, or PETA, is a nonprofit animal rights organization known for its controversial approach to communications and public relations. In 2003, PETA launched a new campaign, named “Holocaust on Your Plate,” that compares the slaughter of animals for human use to the murder of 6 million Jews in WWII. The question: Is “Holocaust on Your Plate” ethically wrong or a truthful comparison?

Aaargh! Pirates! (and the Press) As collections of songs, studio recordings from an upcoming album or merely unreleased demos, are leaked online, these outlets cover the leak with a breaking story or a blog post. But they don’t stop there. Rolling Stone and Billboard often also will include a link within the story to listen to the songs that were leaked. The question: If Billboard and Rolling Stone are essentially pointing readers in the right direction, to the leaked music, are they not aiding in helping the Internet community find the material and consume it?

Reigning on the Parade Frank Whelan, a features writer who also wrote a history column for the Allentown, Pennsylvania, Morning Call , took part in a gay rights parade in June 2006 and stirred up a classic ethical dilemma. The situation raises any number of questions about what is and isn’t a conflict of interest. The question: What should the “consequences” be for Frank Whelan?

Controversy over a Concert Three former members of the Eagles rock band came to Denver during the 2004 election campaign to raise money for a U.S. Senate candidate, Democrat Ken Salazar. John Temple, editor and publisher of the Rocky Mountain News, advised his reporters not to go to the fundraising concerts. The question: Is it fair to ask newspaper staffers — or employees at other news media, for that matter — not to attend events that may have a political purpose? Are the rules different for different jobs at the news outlet?

Deep Throat, and His Motive The Watergate story is considered perhaps American journalism’s defining accomplishment. Two intrepid young reporters for The Washington Post , carefully verifying and expanding upon information given to them by sources they went to great lengths to protect, revealed brutally damaging information about one of the most powerful figures on Earth, the American president. The question: Is protecting a source more important than revealing all the relevant information about a news story?

When Sources Won’t Talk The SPJ Code of Ethics offers guidance on at least three aspects of this dilemma. “Test the accuracy of information from all sources and exercise care to avoid inadvertent error.” One source was not sufficient in revealing this information. The question: How could the editors maintain credibility and remain fair to both sides yet find solid sources for a news tip with inflammatory allegations?

A Suspect “Confession” John Mark Karr, 41, was arrested in mid-August in Bangkok, Thailand, at the request of Colorado and U.S. officials. During questioning, he confessed to the murder of JonBenet Ramsey. Karr was arrested after Michael Tracey, a journalism professor at the University of Colorado, alerted authorities to information he had drawn from e-mails Karr had sent him over the past four years. The question: Do you break a confidence with your source if you think it can solve a murder — or protect children half a world away?

Who’s the “Predator”? “To Catch a Predator,” the ratings-grabbing series on NBC’s Dateline, appeared to catch on with the public. But it also raised serious ethical questions for journalists. The question: If your newspaper or television station were approached by Perverted Justice to participate in a “sting” designed to identify real and potential perverts, should you go along, or say, “No thanks”? Was NBC reporting the news or creating it?

The Media’s Foul Ball The Chicago Cubs in 2003 were five outs from advancing to the World Series for the first time since 1945 when a 26-year-old fan tried to grab a foul ball, preventing outfielder Moises Alou from catching it. The hapless fan's identity was unknown. But he became recognizable through televised replays as the young baby-faced man in glasses, a Cubs baseball cap and earphones who bobbled the ball and was blamed for costing the Cubs a trip to the World Series. The question: Given the potential danger to the man, should he be identified by the media?

Publishing Drunk Drivers’ Photos When readers of The Anderson News picked up the Dec. 31, 1997, issue of the newspaper, stripped across the top of the front page was a New Year’s greeting and a warning. “HAVE A HAPPY NEW YEAR,” the banner read. “But please don’t drink and drive and risk having your picture published.” Readers were referred to the editorial page where White explained that starting in January 1998 the newspaper would publish photographs of all persons convicted of drunken driving in Anderson County. The question: Is this an appropriate policy for a newspaper?

Naming Victims of Sex Crimes On January 8, 2007, 13-year-old Ben Ownby disappeared while walking home from school in Beaufort, Missouri. A tip from a school friend led police on a frantic four-day search that ended unusually happily: the police discovered not only Ben, but another boy as well—15-year-old Shawn Hornbeck, who, four years earlier, had disappeared while riding his bike at the age of 11. Media scrutiny on Shawn’s years of captivity became intense. The question: Question: Should children who are thought to be the victims of sexual abuse ever be named in the media? What should be done about the continued use of names of kidnap victims who are later found to be sexual assault victims? Should use of their names be discontinued at that point?

A Self-Serving Leak San Francisco Chronicle reporters Mark Fainaru-Wada and Lance Williams were widely praised for their stories about sports figures involved with steroids. They turned their investigation into a very successful book, Game of Shadows . And they won the admiration of fellow journalists because they were willing to go to prison to protect the source who had leaked testimony to them from the grand jury investigating the BALCO sports-and-steroids. Their source, however, was not quite so noble. The question: Should the two reporters have continued to protect this key source even after he admitted to lying? Should they have promised confidentiality in the first place?

The Times and Jayson Blair Jayson Blair advanced quickly during his tenure at The New York Times , where he was hired as a full-time staff writer after his internship there and others at The Boston Globe and The Washington Post . Even accusations of inaccuracy and a series of corrections to his reports on Washington, D.C.-area sniper attacks did not stop Blair from moving on to national coverage of the war in Iraq. But when suspicions arose over his reports on military families, an internal review found that he was fabricating material and communicating with editors from his Brooklyn apartment — or within the Times building — rather than from outside New York. The question: How does the Times investigate problems and correct policies that allowed the Blair scandal to happen?

Cooperating with the Government It began on Jan. 18, 2005, and ended two weeks later after the longest prison standoff in recent U.S. history. The question: Should your media outlet go along with the state’s request not to release the information?

Offensive Images Caricatures of the Prophet Muhammad didn’t cause much of a stir when they were first published in September 2005. But when they were republished in early 2006, after Muslim leaders called attention to the 12 images, it set off rioting throughout the Islamic world. Embassies were burned; people were killed. After the rioting and killing started, it was difficult to ignore the cartoons. Question: Do we publish the cartoons or not?

The Sting Perverted-Justice.com is a Web site that can be very convenient for a reporter looking for a good story. But the tactic raises some ethical questions. The Web site scans Internet chat rooms looking for men who can be lured into sexually explicit conversations with invented underage correspondents. Perverted-Justice posts the men’s pictures on its Web site. Is it ethically defensible to employ such a sting tactic? Should you buy into the agenda of an advocacy group — even if it’s an agenda as worthy as this one?

A Media-Savvy Killer Since his first murder in 1974, the “BTK” killer — his own acronym, for “bind, torture, kill” — has sent the Wichita Eagle four letters and one poem. How should a newspaper, or other media outlet, handle communications from someone who says he’s guilty of multiple sensational crimes? And how much should it cooperate with law enforcement authorities?

A Congressman’s Past The (Portland) Oregonian learned that a Democratic member of the U.S. Congress, up for re-election to his fourth term, had been accused by an ex-girlfriend of a sexual assault some 28 years previously. But criminal charges never were filed, and neither the congressman, David Wu, nor his accuser wanted to discuss the case now, only weeks before the 2004 election. Question: Should The Oregonian publish this story?

Using this Process to Craft a Policy It used to be that a reporter would absolutely NEVER let a source check out a story before it appeared. But there has been growing acceptance of the idea that it’s more important to be accurate than to be independent. Do we let sources see what we’re planning to write? And if we do, when?

SPJ News –  SPJ urges swift action by Senate panel on federal shield law to protect journalists –  SPJ calls for action by journalists against gag rules after key legal win –  SPJ Foundation awards over $45,000 in grants to seven initiatives and journalism programs

• Open access
• Published: 05 June 2024

Current status and ongoing needs for the teaching and assessment of clinical reasoning – an international mixed-methods study from the students` and teachers` perspective

• F. L Wagner 1 ,
• M. Sudacka 2 ,
• A. A Kononowicz 3 ,
• M. Elvén 4 , 5 ,
• S. J Durning 6 ,
• I. Hege 7 &
• S. Huwendiek 1  

BMC Medical Education volume  24 , Article number:  622 ( 2024 ) Cite this article

46 Accesses

Metrics details

Clinical reasoning (CR) is a crucial ability that can prevent errors in patient care. Despite its important role, CR is often not taught explicitly and, even when it is taught, typically not all aspects of this ability are addressed in health professions education. Recent research has shown the need for explicit teaching of CR for both students and teachers. To further develop the teaching and learning of CR we need to improve the understanding of students' and teachers' needs regarding content as well as teaching and assessment methods for a student and trainer CR curriculum.

Parallel mixed-methods design that used web-surveys and semi-structured interviews to gather data from both students (n survey  = 100; n interviews  = 13) and teachers (n survey  = 112; n interviews  = 28). The interviews and surveys contained similar questions to allow for triangulation of the results. This study was conducted as part of the EU-funded project DID-ACT ( https://did-act.eu ).

Both the surveys and interview data emphasized the need for content in a clinical reasoning (CR) curriculum such as “gathering, interpreting and synthesizing patient information”, “generating differential diagnoses”, “developing a diagnostic and a treatment plan” and “collaborative and interprofessional aspects of CR”. There was high agreement that case-based learning and simulations are most useful for teaching CR. Clinical and oral examinations were favored for the assessment of CR. The preferred format for a train-the-trainer (TTT)-course was blended learning. There was also some agreement between the survey and interview participants regarding contents of a TTT-course (e.g. teaching and assessment methods for CR). The interviewees placed special importance on interprofessional aspects also for the TTT-course.

Conclusions

We found some consensus on needed content, teaching and assessment methods for a student and TTT-course in CR. Future research could investigate the effects of CR curricula on desired outcomes, such as patient care.

Peer Review reports

Introduction

Clinical reasoning (CR) is a universal ability that mobilizes integration of necessary fundamental knowledge while delivering high-quality patient care in a variety of contexts in a timely and effective way [ 1 , 2 ]. Daniel et al. [ 3 ] define it as a “skill, process or outcome wherein clinicians observe, collect, and interpret data to diagnose and treat patients”. CR encompasses health professionals thinking and acting in patient assessment, diagnostic, and management processes in clinical situations, taking into account the patient ‘s specific circumstances and preferences [ 4 ]. How CR is defined can vary between health professions, but there are also similarities [ 5 ]. Poor CR is associated with low-quality patient care and increases the risk of medical errors [ 6 ]. Berner and Graber [ 7 ] suggested that the rate of diagnostic error is around 15%, underlining the threat that insufficient CR ability poses to patient safety as well as increasing healthcare costs [ 8 ]. Despite the importance of CR, it appears to be rarely taught or assessed explicitly, often only parts of the CR process are covered in existing curricula, and there seems to be a lack of progression throughout curricula (e.g. [ 9 , 10 , 11 , 12 , 13 , 14 ].). Moreover, teachers are often not trained to explicitly teach CR, including explaining their own reasoning to others [ 10 , 11 , 12 ] although this appears to be an important factor in the implementation of a CR curriculum [ 15 ]. Some teachers even question whether CR can be explicitly taught [ 16 ]. Considering these findings, efforts should be made to incorporate explicit teaching of CR into health care professions curricula and training for teachers should be established based on best evidence. However, to date, little is known about what a longitudinal CR curriculum should incorporate to meet the needs of teachers and students.

Insights regarding teaching CR were provided from a global survey by Kononowicz et al. [ 10 ], who reported a need for a longitudinal CR curriculum. However, the participants in their study were mainly health professions educators, leaving the needs of students for a CR curriculum largely unknown. As students are future participants of a CR curriculum, their needs should also be investigated. Kononowicz et al. [ 10 ] also identified a lack of qualified faculty to teach CR. A train-the-trainer course for CR could help reduce this barrier to teaching CR. To the best of our knowledge, in addition to the work by Kononowicz et al. [ 10 ], no research exists yet that addresses the needs of teachers for such a course, and Kononowicz et al. [ 10 ] did not investigate their needs beyond course content. Recently, Gupta et al. [ 12 ] and Gold et al. [ 13 ] conducted needs analyses regarding clinical reasoning instruction from the perspective of course directors at United States medical schools, yet a European perspective is missing. Thus, our research questions were the following:

What aspects of clinical reasoning are currently taught and how important are they in a clinical reasoning curriculum according to teachers and students?

What methods are currently used to teach and assess clinical reasoning and which methods would be ideal according to teachers and students?

In what study year does the teaching of clinical reasoning currently begin and when should it ideally begin according to teachers and students?

How should a train-the-trainer course for teachers of clinical reasoning be constructed regarding content and format?

In this study, we used a convergent parallel mixed-methods design [ 17 ] within a pragmatic constructivist case study approach [ 18 ]. We simultaneously collected data from students and educators using online questionnaires and semi-structured interviews to gain deeper insight into their needs on one particular situation [ 19 ]– the development of a clinical reasoning curriculum—to address our research questions. To help ensure that the results of the survey and the interviews could be compared and integrated, we constructed the questions for the survey and the interviews similarly with the exception that in the interviews, the questions were first asked openly. The design was parallel both in that we collected data simultaneously and also constructed the survey and interviews to cover similar topics. We chose this approach to obtain comprehensive answers to the research questions and to facilitate later triangulation [ 17 ] of the results.

Context of this study

We conducted this study within the EU-funded (Erasmus + program) project DID-ACT (“Developing, implementing, and disseminating an adaptive clinical reasoning curriculum for healthcare students and educators”; https://did-act.eu ). Institutions from six European countries (Augsburg University, Germany; Jagiellonian University in Kraków, Poland; Maribor University, Slovenia; Örebro University, Sweden; University of Bern, Switzerland; EDU, a higher medical education institution based in Malta, Instruct GmbH, Munich, Germany) with the support of associate partners (e.g., Prof. Steven Durning, Uniformed Services University of the Health Sciences, USA; Mälardalen University, Sweden.) were part of this project. For further information, see https://did-act.eu/team-overview/team/ . In this project, we developed an interprofessional longitudinal clinical reasoning curriculum for students in healthcare education and a train-the-trainer course for health profession educators. The current curriculum (for a description of the curriculum, see Hege et al. [ 20 ]) was also informed by this study. This study was part of the Erasmus + Knowledge Alliance DID-ACT (612,454-EPP-1–2019-1-DE-EPPKA2-KA).

Target groups

We identified two relevant target groups for this study, teachers and students, which are potential future users and participants of a train—the—trainer (TTT-) course and a clinical reasoning curriculum, respectively. The teacher group also included individuals who were considered knowledgeable regarding the current status of clinical reasoning teaching and assessment at their institutions (e.g. curriculum managers). These specific participants were individually selected by the DID-ACT project team to help ensure that they had the desired level of expertise. The target groups included different health professions from a large number of countries (see Table  1 ), as we wanted to gather insights that are not restricted to one profession.

Development of data collection instruments

Development of questions.

The questions in this study addressed the current status and needs regarding content, teaching, and assessment of clinical reasoning (CR). They were based on the questions used by Kononowicz et al. [ 10 ] and were expanded to obtain more detailed information. Specifically, regarding CR content, we added additional aspects (see Table 8 in the Appendix for details). The contents covered in this part of the study also align with the five domains of CR education (clinical reasoning concepts, history and physical examination, choosing and interpreting diagnostic tests, problem identification and management and shared decision-making) that were reported by Cooper et al. [ 14 ]. It has been shown that there are similarities between professions regarding the definition of CR (e.g. history taking or an emphasis on clinical skills), while nurses placed greater importance on a patient-centered approach [ 5 ]. We aimed to cover as many aspects of CR in the contents as possible to represent these findings. We expanded the questions on CR teaching formats to cover a broader range of formats. Furthermore, two additional assessment methods were added to the respective questions. Finally, one aspect was added to the content questions for a train-the-trainer course (see Table 8 in the Appendix ). As a lack of qualified faculty to teach CR was identified in the study by Kononowicz et al. [ 10 ], we added additional questions on the specific needs for the design of a CR train-the-trainer course beyond content. Table 8 in the Appendix shows the adaptations that we made in detail.

We discussed the questions within the interprofessional DID-ACT project team and adapted them in several iterative cycles until the final versions of the survey questionnaire and the interview guide were obtained and agreed upon. We tested the pre-final versions with think-alouds [ 21 ] to ensure that the questions were understandable and interpreted as intended, which led to a few changes. The survey questionnaires and interview-guides can be found at https://did-act.eu/results/ and accessed via links in table sections D1.1a (survey questions) and D1.1b (interview guides), respectively. Of these questions, we included only those relevant to the research questions addressed in this study. The questions included in this study can be found in the Appendix in Table8.

Teachers were asked questions about all content areas, but only the expert subgroup was asked to answer questions on the current situation regarding the teaching and assessment of clinical reasoning at their institutions, as they were considered the best informed group on the matter. Furthermore, students were not asked questions on the train-the-trainer course. Using the abovementioned procedures, we also hoped to improve the response rate as longer surveys were found to be associated with lower response rates [ 22 ].

We created two different versions of the interview guide, one for teachers and one for students. The student interview guide did not contain questions on the current status of clinical reasoning teaching and assessment or questions about the train-the-trainer course. The interview guides were prepared with detailed instructions to ensure that the interviews were conducted in a comparable manner at all locations. By using interviews, we intended to obtain a broad picture of existing needs. Individual interviews further allowed participants to speak their own languages and thus to express themselves naturally and as precisely as possible.

Reflexivity statement

Seven researchers representing different perspectives and professions form the study team. MS has been a PhD candidate representing the junior researcher perspective, while also experienced researchers with a broad background in clinical reasoning and qualitative as well as quantitative research are part of the team (SD, SH, AK, IH, ME, FW). ME represents the physiotherapist perspective, SD, SH, and MS represent the medical perspective. We discussed all steps of the study in the team and made joint decisions.

Data collection and analysis

The survey was created using LimeSurvey software (LimeSurvey GmbH). The survey links were distributed via e-mail (individual invitations, posts to institutional mailing lists, newsletters) by the DID-ACT project team and associate partners (the target groups received specific links to the online-survey). The e-mail contained information on the project and its goals. By individually contacting persons in the local language, we hoped to increase the likelihood of participation. The survey was anonymous. The data were collected from March to July 2020.

Potential interview participants were contacted personally by the DID-ACT project team members in their respective countries. We used a convenience sampling approach by personally contacting potential interview partners in the local language to motivate as many participants as possible. With this approach we also hoped to increase the likelihood of participation. The interviews were conducted in the local languages also to avoid language barriers and were audio-recorded to help with the analysis and for documentation purposes. Most interviews were conducted using online meeting services (e.g. Skype or Zoom) because of restrictions due to the ongoing coronavirus pandemic that occurred with the start of data collection at the beginning of the DID-ACT project. The data were collected from March to July 2020. All interview partners provided informed consent.

Ethics approval and consent to participate

We asked the Bern Ethics Committee to approve this multi-institutional study. This type of study was regarded as exempt from formal ethical approval according to the regulations of the Bern Ethics Committee (‘Kantonale Ethikkommission Bern’, decision Req-2020–00074). All participants voluntarily participated and provided informed consent before taking part in this study.

Data analysis

Descriptive analyses were performed using SPSS statistics software (version 28, 2021). Independent samples t-tests were computed for comparisons between teachers and students. When the variances of the two groups were unequal, Welch’s test was used. Bonferroni correction of significance levels was used to counteract alpha error accumulation in repeated tests. The answers to the free text questions were screened for recurring themes. There were very few free-text comments, typically repeating aspects from the closed questions, hence, no meaningful analysis was possible. For this reason, the survey comments are mentioned only where they made a unique contribution to the results.

The interviews were translated into English by the partners. An overarching summarizing qualitative content analysis [ 23 ] of the data was conducted. A summarizing content analysis is particularly useful when the content level of the material is of interest. Its goal is to reduce the material to manageable short texts in a way that retains the essential meaning [ 23 ]. The analysis was conducted first by two of the authors of the study (FW, SH) and then discussed by the entire author team. The analysis was carried out as an iterative process until a complete consensus was reached within the author team.

The results from the surveys and interviews were compared and are presented together in the results section. The qualitative data are reported in accordance with the standards for reporting qualitative research (SRQR, O’Brien et al. [ 24 ]).

Table 1 shows the professional background and country of the interviewees and survey samples. The survey was opened by 857 persons, 212 (25%) of whom answered the questions included in this study. The expert sub-group of teachers who answered the questions on the current status of clinical reasoning teaching and assessment encompassed 45 individuals.

Content of a clinical reasoning curriculum for students

The survey results show that “Gathering, interpreting, and synthesizing patient information”, is currently most extensively taught, while “Theories of clinical reasoning” are rarely taught (see Table  2 ). In accordance with these findings, “Gathering, interpreting, and synthesizing patient information” received the highest mean importance rating for a clinical reasoning curriculum while “Theories of clinical reasoning” received the lowest importance rating. Full results can be found in Table 9 in the Appendix .

Teachers and students differed significantly in their importance ratings of two content areas, “Gathering, interpreting, and synthesizing patient information” ( t (148.32) = 4.294, p  < 0.001, d  = 0.609) and “Developing a problem formulation/hypothesis” ( t (202) = 4.006, p  < 0.001, d  = 0.561), with teachers assigning greater importance to both of these content areas.

The results from the interviews are in line with those from the survey. Details can be found in Table 12 in the Appendix .

Clinical reasoning teaching methods

The survey participants reported that, most often, case-based learning is currently applied in the teaching of clinical reasoning (CR). This format was also rated as most important for teaching CR (see Table  3 ). Full results can be found in Table 10 in the Appendix .

Teachers and students differed significantly in their importance ratings of Team-based learning ( t (202) = 3.079, p  = 0.002, d  = 0.431), with teachers assigning greater importance to this teaching format.

Overall, the interviewees provided very similar judgements to the survey participants. Next to the teaching formats shown in Table  3 , some of them would employ blended learning, and clinical teaching formats such as bedside teaching and internships were also mentioned. Details can be found in the Appendix in Table 13. In addition to the importance of each individual teaching format, it was also argued that all of the formats can be useful because they all are meant to reach different objectives and that there is not one single best format for teaching CR.

Start of clinical reasoning teaching in curricula

Most teachers (52.5%) reported that currently, the teaching of clinical reasoning (CR) starts in the first year of study. Most often (46.4%) the participants also chose the first study year as the optimal year for starting the teaching CR. In accordance with the survey results, the interviewees also advocated for an early start of the teaching of CR. Some interview participants who advocated for a later start of CR teaching suggested that the students first need a solid knowledge base and that once the clinical/practical education starts, explicit teaching of CR should begin.

Assessment of clinical reasoning

The survey results suggest that currently written tests or clinical examinations are most often used, while Virtual Patients are used least often (see Table  4 ). Despite written tests being the most common current assessment format, they received the lowest importance rating for a future longitudinal CR curriculum. Full results can be found in Table 11 in the Appendix .

Teachers and students differed significantly in their importance ratings of clinical examinations ( t (161.81) = 2.854, p  = 0.005, d  = 0.413) and workplace-based assessments ( t (185) = 2.640, p = 0.009, d  = 0.386) with teachers assigning greater importance to both of these assessment formats.

The interviewees also placed importance on all assessment methods but found it difficult to assess CR with written assessment methods. The students seemed to associate clinical examinations more with practical skills than with CR. Details can be found in the Appendix in Table 14. Two of the interview participants mentioned that CR is currently not assessed at their institutions, and one person mentioned that students are asked to self-reflect on their interactions with patients and on potential improvements.

Train-the-trainer course

The following sections highlight the results from the needs analysis regarding a train-the-trainer (TTT-) course. The questions presented here were posed only to the teachers.

Most survey participants reported that there is currently no TTT- course on clinical reasoning at their institution but that they think such a course is necessary (see Table  5 ). The same was also true for the interviewees (no TTT- course on clinical reasoning existing but need for one).

In the interviews, 22 participants (78.6%) answered that a TTT-course is necessary for healthcare educators, two participants answered that no such course was necessary, and two other participants were undecided about its necessity. At none of the institutions represented by the interviewees, a TTT-course for teaching clinical reasoning exists.

When asked what the best format for a clinical reasoning TTT- course would be (single answer question), the majority of the survey participants favored a blended learning / flipped classroom approach, a combination of e-learning and face-to-face meetings. (see Table  6 ).

In the survey comments it was noted that blended-learning encompasses the benefits of both self-directed learning and discussion/learning from others. It would further allow teachers to gather knowledge about CR first in an online learning phase where they can take the time they need before coming to a face-to-face meeting.

The interviewees also found a blended-learning approach particularly suitable for a TTT-course. An e-learning course only was seen as more critical because teachers may lack motivation to participate in an online-only setting, while a one-time face-to-face meeting would not provide enough time. In some interviews, it was emphasized that teachers should experience themselves what they are supposed to teach to the students and also that the trainers for the teachers need to have solid education and knowledge on clinical reasoning.

Table 7 shows the importance ratings of potential content of a TTT-course generated from the survey. To elaborate on this content, comments by the interviewees were added. On average, all content was seen as (somewhat) important with teaching methods on the ward and/or clinic receiving the highest ratings. Some interviewees also mentioned the importance of interprofessional aspects and interdisciplinary understanding of CR. In the survey comments, some participants further expressed their interest in such a course.

Finally, the interviewees were asked about the ideal length of a clinical reasoning TTT-course. The answers varied greatly from 2–3 hours to a two-year educational program, with a tendency toward 1–2 days. Several interviewees commented that the time teachers are able to spend on a TTT-course is limited. This should be considered in the planning of such a course to make participation feasible for teachers.

In this study, we investigated the current status of and suggestions for teaching and assessment of clinical reasoning (CR) in a longitudinal curriculum as well as suggestions for a train-the-trainer (TTT-) course for CR. Teachers and students were invited to participate in online-surveys as well as semi-structured interviews to derive answers to our research questions. Regarding the contents of a CR curriculum for students, the results of the surveys and interviews were comparable and favoured content such as gathering, interpreting, and synthesizing patient information, generating differential diagnoses, and developing a diagnostic and a treatment plan. In the interviews, high importance was additionally placed on collaborative and interprofessional aspects of CR. Case-based learning and simulations were seen as the most useful methods for teaching CR, and clinical and oral examinations were favoured for the assessment of CR. The preferred format for a TTT-course was blended learning. In terms of course content, teaching and assessment methods for CR were emphasized. In addition to research from the North American region [ 11 ], this study provides results from predominantly European countries that support the existing findings.

Content of a clinical reasoning curriculum

Our results revealed that there are still aspects of clinical reasoning (CR), such as “Errors in the clinical reasoning process and strategies to avoid them” or “Interprofessional aspects of CR” that are rarely taught despite their high importance, corroborating the findings of Kononowicz et al. [ 10 ]. According to the interviewees, students should have basic knowledge of CR before they are taught about errors in the CR process and strategies to avoid them. The lack of teaching of errors in CR may also stem from a lack of institutional culture regarding how to manage failures in a constructive way (e.g. [ 16 , 25 ]), making it difficult to explicitly address errors and strategies to avoid them. Although highly relevant in the everyday practice of healthcare professions and underpinned by CR theoretical frameworks (e.g., distributed cognition [ 26 ]), interprofessional and collaborative aspects of CR are currently rarely considered in the teaching of CR. The interviews suggested that hierarchical distance and cultural barriers may contribute to this finding. Sudacka et al. [ 16 ] also reported cultural barriers as one reason for a lack of CR teaching. Generally, the interviewees seemed to place greater importance on interprofessional and collaborative aspects than did the survey-participants This may have been due to differences in the professions represented in the two modalities (e.g., a greater percentage of nurses among the interview participants, who tend to define CR more broadly than physicians [ 5 ]).

“Self-reflection on clinical reasoning performance and strategies for future improvement”, “Developing a problem formulation/hypothesis” and “Aspects of patient-participation in CR” were rated as important but are currently rarely taught, a finding not previously reported. The aspect “Self-reflection on clinical reasoning performance and strategies for future improvement”, received high importance ratings, but only 25% of the survey-participants answered that it is currently taught to a great extent. The interviewees agreed that self-reflection is important and added that ideally, it should be guided by specific questions. Ogdie et al. [ 27 ] found that reflective writing exercises helped students identify errors in their reasoning and biases that contributed to these errors.

“Gathering, interpreting, and synthesizing patient information” and “Developing a problem formulation/hypothesis” were rated significantly more important by teachers than by students. It appears that students may be less aware yet of the importance of gathering, interpreting, and synthesizing patient information in the clinical reasoning process. There was some indication in the interviews that the students may not have had enough experience yet with “Developing a problem formulation/hypothesis” or associate this aspect with research, possibly contributing to the observed difference.

Overall, our results on the contents of a CR curriculum suggest that all content is important and should be included in a CR curriculum, starting with basic theoretical knowledge and data gathering to more advanced aspects such as errors in CR and collaboration. Two other recent surveys conducted in the United States among pre-clerkship clinical skills course directors [ 12 ] and members of clerkship organizations [ 13 ] came to similar conclusions regarding the inclusion of clinical reasoning content at various stages of medical curricula. How to fit the content into already dense study programs, however, can still be a challenge [ 16 ].

In addition to case-based learning and clinical teaching, human simulated patients and Team-based learning also received high importance ratings for teaching clinical reasoning (CR), a finding not previously reported. Lectures, on the other hand, are seen as the least important to teach CR (see also Kononowicz et al. [ 10 ]), as they mainly deliver factual knowledge according to the interviewees. High-fidelity simulations (mannequins) and Virtual Patients (VPs) are rarely used to teach CR at the moment and are rated less important compared to other teaching formats. Some interviewees see high-fidelity simulations as more useful for teaching practical skills. The lower importance rating of VPs was surprising given that this format is case-based, provides a safe environment for learning, and is described in the literature as a well-suited tool for teaching CR [ 28 , 29 ]. Considering that VPs seemed to be used less often at the institutions involved in this study, the lack of experience with this format may have led to this result.

Teachers rated Team-based learning as significantly more important for teaching clinical reasoning than students. In the interviews, many students seemed not to be familiar with Team-based learning, possibly explaining the lower ratings the students gave this format in the survey.

Taken together, our results suggest that there is not one best format for teaching all aspects of clinical reasoning but rather that the use of all teaching formats is justified depending on the specific content to be taught and goals to be achieved. However, there was agreement that a safe learning environment where no patients can be harmed is preferred for teaching clinical reasoning, and that discussions should be possible.

There was wide agreement that clinical reasoning (CR) teaching should start in the first year of study in the curriculum. However, a few participants of this study argued that students first need to develop some general knowledge before CR is taught. Rencic et al. [ 11 ] reported that according to internal medicine clerkship directors, CR should be taught throughout all years of medical school, with a particular focus during the clinical teaching years. A similar remark was made by participants in a survey among pre-clerkship clinical skills course directors by Gupta et al. [ 12 ] where the current structure of some curricula (e.g. late introduction of the pathophysiology) was regarded as a barrier to introducing CR from the first year of study on [ 12 ].

Our results show that the most important format for assessing clinical reasoning (CR) that is also currently used to the greatest extent are clinical examinations (e.g. OSCE), consistent with Kononowicz et al. [ 10 ]. The interviewees emphasized that CR should ideally be assessed in a conversation or discussion where the learners can explain their reasoning. Given this argument, all assessment formats enabling a conversation are suitable for assessing CR. This is reflected in our survey results, where assessment formats that allow for a discussion with the learner received the most favourable importance ratings, including oral examinations. In agreement with Kononowicz et al. [ 10 ], we also found that written tests are currently used most often to assess CR but are rated as least important and suitable only for the assessment of some aspects of CR. Daniel et al. [ 3 ] argued that written exams such as MCQs, where correct answers have to be selected from a list of choices, are not the best representation of real practical CR ability. Thus, there still seems to be potential for improvement in the way CR is assessed.

Teachers rated clinical examinations and workplace-based assessments significantly higher than students. Based on the interviews, the students seemed to associate clinical examinations such as OSCEs more with a focus on practical skills than CR, potentially explaining their lower ratings of this format.

What a clinical reasoning train-the-trainer course should look like

Our results show a clear need for a clinical reasoning (CR) train-the-trainer course (see also Singh et al. [ 15 ]), which currently does not exist at most institutions represented in this study, corroborating findings by Kononowicz et al. [ 10 ]. A lack of adequately trained teachers is a common barrier to the introduction of CR content into curricula [ 12 , 16 ]. According to our results such a course should follow a blended learning/flipped classroom approach or consist of a series of face-to-face meetings. A blended-learning course would combine the benefits of both self-directed learning and the possibility for trainers to discuss with and learn from their peers, which could also increase their motivation to participate in such a course. An e-learning only course or a one-time face-to-face meeting were considered insufficient. The contents “Clinical reasoning strategies” and “Common errors in the clinical reasoning process” were given greater importance for the trainer-curriculum than for the students-curriculum, possibly reflecting higher expectations of trainers as “CR experts” compared with students. There was some agreement in the interviews that ideally, the course should not be too time-consuming, with participants tending towards an overall duration of 1–2 days, considering that most teachers usually have many duties and may not be able or willing to attend the course if it were too long. Lack of time was also identified as a barrier to attending teacher training [ 12 , 13 , 16 ].

Strengths and limitations

The strengths of this study include its international and interprofessional participants. Furthermore, we explicitly included teachers and students as target groups in the same study, which enables a comparison of different perspectives. Members of the target groups not only participated in a survey but were also interviewed to gain in-depth knowledge. A distinct strength of this study is its mixed-methods design. The two data collection methods employed in parallel provided convergent results, with responses from the web survey indicating global needs and semi-structured interviews contributing to a deeper understanding of the stakeholder groups’ nuanced expectations and perspectives on CR education.

This study is limited in that most answers came from physicians, making the results potentially less generalizable to other professions. Furthermore, there were participants from a great variety of countries, with some countries overrepresented. Because of the way the survey-invitations were distributed, the exact number of recipients is unknown, making it impossible to compute an exact response rate. Also, the response rate of the survey was rather low for individuals who opened the survey. Because the survey was anonymous, it cannot completely be ruled out that some individuals participated in both interviews and survey. Finally, there could have been some language issues in the interview analysis, as the data were translated to English at the local partner institutions before they were submitted for further analysis.

Our study provides evidence of an existing need for explicit clinical reasoning (CR) longitudinal teaching and dedicated CR teacher training. More specifically, there are aspects of CR that are rarely taught that our participants believe should be given priority, such as self-reflection on clinical reasoning performance and strategies for future improvement and aspects of patient participation in CR that have not been previously reported. Case-based learning and clinical teaching methods were again identified as the most important formats for teaching CR, while lectures were considered relevant only for certain aspects of CR. To assess CR, students should have to explain their reasoning, and assessment formats should be chosen accordingly. There was also still a clear need for a CR train-the-trainer course. In addition to existing research, our results show that such a course should ideally have a blended-learning format and should not be too time-consuming. The most important contents of the train-the-trainer course were confirmed to be teaching methods, CR strategies, and strategies to avoid errors in the CR process. Examples exist for what a longitudinal CR curriculum for students and a corresponding train-the-trainer course could look like and how these components could be integrated into existing curricula (e.g. DID-ACT curriculum [ 20 ], https://did-act.eu/integration-guide/ or the described curriculum of Singh et al. [ 15 ]). Further research should focus on whether and to what extent the intended outcomes of such a curriculum are actually reached, including the potential impact on patient care.

Availability of data and materials

All materials described in this manuscript generated during the current study are available from the corresponding author on reasonable request without breaching participant confidentiality.

Connor DM, Durning SJ, Rencic JJ. Clinical reasoning as a core competency. Acad Med. 2020;95:1166–71.

Article   Google Scholar  

Young M, Szulewski A, Anderson R, Gomez-Garibello C, Thoma B, Monteiro S. Clinical reasoning in CanMEDS 2025. Can Med Educ J. 2023;14:58–62.

Google Scholar  

Daniel M, Rencic J, Durning SJ, Holmboe E, Santen SA, Lang V, Gruppen LD. Clinical reasoning assessment methods: a scoping review and practical guidance. Acad Med. 2019;94:902–12.

Scott IA. Errors in clinical reasoning: causes and remedial strategies. BMJ. 2009. https://doi.org/10.1136/bmj.b1860 .

Huesmann L, Sudacka M, Durning SJ, Georg C, Huwendiek S, Kononowicz AA, Schlegel C, Hege I. Clinical reasoning: what do nurses, physicians, and students reason about. J Interprof Care. 2023;37:990–8.

Norman GR, Eva KW. Diagnostic error and clinical reasoning. Med Educ. 2010;44:94–100.

Berner E, Graber M. Overconfidence as a cause of diagnostic error in medicine. Am J Med. 2008;121:2–23.

Cooper N, Da Silva AL, Powell S. Teaching clinical reasoning. In: Cooper N, Frain J, editors. ABC of clinical reasoning. 1st ed. Hoboken, NJ: John Wiley & Sons Ltd; 2016. p. 44–50.

Elvén M, Welin E, Wiegleb Edström D, Petreski T, Szopa M, Durning SJ, Edelbring S. Clinical reasoning curricula in health professions education: a scoping review. J Med Educ Curric Dev. 2023. https://doi.org/10.1177/23821205231209093 .

Kononowicz AA, Hege I, Edelbring S, Sobocan M, Huwendiek S, Durning SJ. The need for longitudinal clinical reasoning teaching and assessment: results of an international survey. Med Teach. 2020;42:457–62.

Rencic J, Trowbridge RL, Fagan M, Szauter K, Durning SJ. Clinical reasoning education at US medical schools: results from a national survey of internal medicine clerkship directors. J Gen Intern Med. 2017;32:1242–6.

Gupta S, Jackson JM, Appel JL, Ovitsh RK, Oza SK, Pinto-Powell R, Chow CJ, Roussel D. Perspectives on the current state of pre-clerkship clinical reasoning instruction in United States medical schools: a survey of clinical skills course directors. Diagnosis. 2021;9:59–68.

Gold JG, Knight CL, Christner JG, Mooney CE, Manthey DE, Lang VJ. Clinical reasoning education in the clerkship years: a cross-disciplinary national needs assessment. PLoS One. 2022;17:e0273250.

Cooper N, Bartlett M, Gay S, Hammond A, Lillicrap M, Matthan J, Singh M. UK Clinical Reasoning in Medical Education (CReME) consensus statement group. Consensus statement on the content of clinical reasoning curricula in undergraduate medical education. Med Teach. 2021;43:152–9.

Singh M, Collins L, Farrington R, Jones M, Thampy H, Watson P, Grundy J. From principles to practice: embedding clinical reasoning as a longitudinal curriculum theme in a medical school programme. Diagnosis. 2021;9:184–94.

Sudacka M, Adler M, Durning SJ, Edelbring S, Frankowska A, Hartmann D, Hege I, Huwendiek S, Sobočan M, Thiessen N, Wagner FL, Kononowicz AA. Why is it so difficult to implement a longitudinal clinical reasoning curriculum? A multicenter interview study on the barriers perceived by European health professions educators. BMC Med Educ. 2021. https://doi.org/10.1186/s12909-021-02960-w .

Hingley A, Kavaliova A, Montgomery J, O’Barr G. Mixed methods designs. In: Creswell JW, editor. Educational research: planning, conducting, and evaluating quantitative and qualitative research. 4th ed. Boston: Pearson; 2012. p. 534–75.

Merriam SB. Qualitative research and case study applications in education. In: from" case study research in education.". Sansome St. Revised and Expanded. San Francisco, CA: Jossey-Bass Publishers; 1998.

Cleland J, MacLeod A, Ellaway RH. The curious case of case study research. Med Educ. 2021;55:1131–41.

Hege I, Adler M, Donath D, Durning SJ, Edelbring S, Elvén M, Wiegleb Edström D. Developing a European longitudinal and interprofessional curriculum for clinical reasoning. Diagnosis. 2023;10:218–24.

Collins D. Pretesting survey instruments: an overview of cognitive methods. Qual Life Res. 2003;12:229–38.

Liu M, Wronski L. Examining completion rates in web surveys via over 25,000 real-world surveys. Soc Sci Comput Rev. 2018;36:116–24.

Mayring P, Fenzl T. Qualitative inhaltsanalyse. In: Baur N, Blasius J, editors. Handbuch methoden der empirischen Sozialforschung. Wiesbaden: Springer VS; 2019. p. 633–48.

Chapter   Google Scholar  

O’Brien BC, Harris IB, Beckman TJ, Reed DA, Cook DA. Standards for reporting qualitative research: a synthesis of recommendations. Acad Med. 2014;89:1245–51.

Edmondson AC. Learning from failure in health care: frequent opportunities, pervasive barriers. BMJ Qual Saf. 2004;13 Suppl 2:ii3-ii9.

Merkebu J, Battistone M, McMains K, McOwen K, Witkop C, Konopasky A, Durning SJ. Situativity: a family of social cognitive theories for understanding clinical reasoning and diagnostic error. Diagnosis. 2020;7:169–76.

Ogdie AR, Reilly JB, Pang WG, Keddem S, Barg FK, Von Feldt JM, Myers JS. Seen through their eyes: residents’ reflections on the cognitive and contextual components of diagnostic errors in medicine. Acad Med. 2012;87:1361–7.

Berman NB, Durning SJ, Fischer MR, Huwendiek S, Triola MM. The role for virtual patients in the future of medical education. Acad Med. 2016;91:1217–22.

Plackett R, Kassianos AP, Mylan S, Kambouri M, Raine R, Sheringham J. The effectiveness of using virtual patient educational tools to improve medical students’ clinical reasoning skills: a systematic review. BMC Med Educ. 2022. https://doi.org/10.1186/s12909-022-03410-x .

Download references

Acknowledgements

We want to thank all participants of the interviews and survey who took their time to contribute to this study despite the ongoing pandemic in 2020. Furthermore, we thank the members of the DID-ACT project team who supported collection and analysis of survey and interview data.

The views expressed herein are those of the authors and not necessarily those of the Department of Defense, the Uniformed Services University or other Federal Agencies.

This study was partially supported by the Erasmus + Knowledge Alliance DID-ACT (612454-EPP-1–2019-1-DE-EPPKA2-KA).

Author information

Authors and affiliations.

Institute for Medical Education, Department for Assessment and Evaluation, University of Bern, Bern, Switzerland

F. L Wagner & S. Huwendiek

Center of Innovative Medical Education, Department of Medical Education, Jagiellonian University, Kraków, Poland

Faculty of Medicine, Department of Bioinformatics and Telemedicine, Jagiellonian University, Kraków, Poland

A. A Kononowicz

School of Health, Care and Social Welfare, Mälardalen University, Västerås, Sweden

Faculty of Medicine and Health, School of Health Sciences, Örebro University, Örebro, Sweden

Uniformed Services University of the Health Sciences, Bethesda, MD, USA

S. J Durning

Institute of Medical Education, University Hospital, LMU Munich, Munich, Germany

You can also search for this author in PubMed   Google Scholar

Contributions

FW and SH wrote the first draft of the manuscript. All authors critically revised the manu-script in several rounds and approved the final manuscript.

Corresponding author

Correspondence to F. L Wagner .

Ethics declarations

This type of study was regarded as exempt from formal ethical approval according to the regulations of the Bern Ethics Committee (‘Kantonale Ethikkommission Bern’, decision Req-2020–00074). All participants voluntarily participated and provided informed consent before taking part in this study.

Consent for publication

All authors consent to publication of this manuscript.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Wagner, F., Sudacka, M., Kononowicz, A. et al. Current status and ongoing needs for the teaching and assessment of clinical reasoning – an international mixed-methods study from the students` and teachers` perspective. BMC Med Educ 24 , 622 (2024). https://doi.org/10.1186/s12909-024-05518-8

Download citation

Received : 16 January 2024

Accepted : 06 May 2024

Published : 05 June 2024

DOI : https://doi.org/10.1186/s12909-024-05518-8

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Clinical reasoning

BMC Medical Education

ISSN: 1472-6920

• Submission enquiries: [email protected]
• General enquiries: [email protected]

• Engineering Ethics Cases
• Markkula Center for Applied Ethics
• Focus Areas
• More Focus Areas
• Engineering Ethics

The following series of engineering ethics cases were created by interviewing numerous engineers from Silicon Valley and beyond.

The cases have been written, anonymized, and honed to highlight the ethical content from each interview. While these cases are meant for engineering students and professionals for their professional development, nearly all of the cases occur in the context of business, and therefore are also relevant for those seeking business ethics cases.

These cases are suitable as homework and/or for classroom discussion. The goal of this project is to acquaint engineering students and professionals with the variety of ethical experiences of engineering as practiced “in the field.” By becoming familiar with problems faced by other engineers we hope to thereby prepare those reading these cases if they too encounter difficult ethical dilemmas in their work.

Cases range from the mundane to the deadly. While we do not reveal how each particular case turned out, in general they turned out well – the people involved made the right decisions. But this is not to say that all of these right decisions came without personal cost. A few of the engineers did face negative repercussions and a very few even needed to find new employment. However, overall the interviewees were satisfied with how events turned out, even if they faced negative repercussions for their good decisions. They understood that doing the right thing is good in itself, regardless of the personal consequences they may have faced.

The engineering ethics cases can be sorted into the following categories:

• Academic Ethics
• Bioengineering
• Business Ethics
• Civil Engineering
• Computer/Software Engineering
• Electrical Engineering
• International
• Mechanical Engineering
• Science/Research Ethics

A quality assurance engineer must decide whether or not to ship products that might be defective.

An intern at a power electronics startup faces unkind comments from a fellow engineer. She suspects that her colleague is prejudice toward female engineers.

A chemical engineering professor discovers that a colleague has taken credit for his research.

A bioengineering researcher discovers an error in protocol and feels pressured not to report it to her supervisor.

A graduate student suspects her research adviser has earned tenure under false pretenses.

A computer startup company risks violating copyright laws if it reuses a code that is the intellectual property of another company.

A recently promoted manager at an industrial engineering company discovers that factory workers are asked to work more than eight hours a day without getting paid overtime.

Full transparency might prevent a project leader from closing a deal with a valuable client. Should he still clarify the situation to his client?

A manager at a consumer electronics company struggles over whether or not he should disclose confidential information to a valued customer.

A medical researcher is asked to trim data before presenting it to the scientific advisory board.

A technical sales engineer feels pressure to falsify a sales report in order to prevent the delay of her company's IPO.

When a computer filled with personal data gets stolen, a data company must decide how to manage the breach in security.

Employees of a computer hardware company are angered by a manager that demonstrates favoritism.

A project engineer believes his company is providing the wrong form of technology to an in-need community in East Africa.

A computer engineer is asked to divulge private medical data for marketing purposes.

Environmental engineers face pressure to come up with data that favors their employers.

In this ethics case, a woman is displeased with her work role at a computer hardware company.

A systems engineering company employee quits after getting pressured to falsify product testing paperwork.

A manager at a nonprofit mechanical engineering firm questions how responsible her company should be for ongoing maintenance on past projects.

An engineer for an environmental consulting firm must decide whether or not he should encourage his client to go with a more environmentally sustainable construction plan.

A genetic engineer feels a responsibility to educate colleagues on the truth behind stem cell research.

An engineering manager gets pressured to bribe a foreign official in order to secure a business venture in East Africa.

An African-American electronics design lead wonders whether his colleague's contentious behavior is motivated by racism.

A medical company asks blood sample suppliers to sign an ethically questionable consent form.

A quality assurance tester gets pressured to falsify data about a new product from a major cell phone company.

Should a production engineer prioritize a customer's desires over safety?

A female intern at a construction company faces disrespectful treatment because of her gender.

A new hire at an electronics startup struggles to decide between telling the truth and maximizing the company's profit.

A fellow for a global services program faces an ethical dilemma when a colleague asks him to falsify receipts.

A researcher of regenerative medicine meets a man who is eager sign up for potentially dangerous human testing.

A bioengineer's research leads to the discovery that a patient might have prostate cancer.

Two support engineers at a South Bay audio visual electronics startup question the fairness of a supervisor's decision.

An employee overseeing data analysis on a clinical drug trial has concerns about the safety of a client's drug.

The engineering ethics cases in this series were written by Santa Clara University School of Engineering students Clare Bartlett, Nabilah Deen, and Jocelyn Tan, who worked as Hackworth Engineering Ethics Fellows at the Markkula Center for Applied Ethics over the course of the 2014-2015 academic year. In order to write these cases, the fellows interviewed numerous engineers and collected nearly 40 engineering ethics cases from Silicon Valley and beyond.  The Hackworth Fellowships are made possible by a generous gift from Joan and the late Michael Hackworth.

• Engineering Business