Getty Images For Unsplash+
Who Decides What’s Ethical in Science? It Depends on Which Lab You’re Standing In.
In A Nutshell
- A survey of more than 11,000 researchers in Sweden found that attitudes toward questionable research practices vary consistently across academic disciplines, with medical researchers holding the strictest views overall.
- Fraud and plagiarism are universally condemned across all fields, but grayer behaviors, such as not sharing data, skipping pre-registration, or using deception in studies, produce significant disagreement.
- Ethics review board members hold stricter views than the researchers they oversee, with the biggest gaps around data sharing, pre-registration, and deception.
- Many research ethics rules are rooted in medical traditions and may not reflect the different ethical realities faced by social scientists, humanists, and others outside medicine.
Ask a medical researcher and a social scientist whether a given research habit crosses an ethical line, and you may get very different answers. A new survey of more than 11,000 researchers in Sweden confirms that what counts as ethically questionable behavior can vary in consistent ways depending on which corner of academia someone calls home.
Medical researchers tend to draw harder lines around dubious research habits than their peers in the natural sciences, social sciences, or humanities. Ethics board reviewers hold even stricter views than the researchers they oversee. And the biggest disagreements aren’t about obvious violations like faking data. They’re about grayer practices that don’t technically break any rules but can quietly erode the trustworthiness of science.
Published in Research Policy, the study raises uncomfortable questions about whether ethics rules, largely rooted in medical traditions, suit every field.
When ‘Questionable’ Depends on Who You Ask
Researchers at Linköping University in Sweden designed a 52-item survey covering a wide range of research behaviors. Some were clear-cut violations: falsifying data, plagiarism. Others fell into a category experts call “questionable research practices,” or QRPs, behaviors that aren’t classified as outright misconduct but can still damage the reliability of published findings. Examples include cherry-picking which results to report, tossing out inconvenient data points, or stopping an experiment early because the numbers already look favorable.
Every researcher and doctoral student at publicly funded Swedish universities received the survey, a pool of 33,290 people. About a third responded, producing a sample of 11,050 researchers, along with a separate group of 144 ethics review board members. Each practice was rated on a seven-point scale, from “completely unacceptable” to “completely acceptable.”
Falsifying data and plagiarism were universally condemned, with nearly every respondent placing them at the extreme unacceptable end of the scale regardless of field. Once the survey moved into murkier territory, the consensus started to fracture.
Research Ethics Look Different Across Disciplines
Some of the largest gaps between fields appeared on questions that might seem straightforward. Whether it’s acceptable to conduct a study without ever publishing the results split the room: natural scientists landed on the acceptable side of the scale’s midpoint on average, while medical researchers landed solidly on the unacceptable side.
Other fault lines emerged around open science practices, the push to make research more transparent by sharing data, code, and study plans publicly. Not sharing data when a study is published, not pre-registering a study’s design before data collection, and not storing data in a way others could use all drew sharper disapproval from medical researchers, while social scientists and humanities scholars tended to be more tolerant.
Across all six categories, medical researchers consistently gave the lowest acceptability ratings, even for the practices most researchers were willing to forgive.
Research Ethics Boards Hold Stricter Views Than Scientists
Beyond discipline, gender and career stage also mattered. Female researchers held stricter attitudes than male researchers. Full professors were less accepting of questionable practices than people at earlier career stages, though the research team noted that finding was entangled with age, since older researchers tended to be stricter and full professors tend to be older.
Ethics review board members stood out as a particularly strict group, with ratings generally lower than those of the researchers they evaluate. The biggest gaps appeared around pre-registration, data sharing, and the use of deception in studies. One exception: ethics reviewers were somewhat more accepting than researchers of making edits to satisfy a journal reviewer’s suggestions, even when the author disagrees.
Why the Gaps Matter for Science Oversight
Modern research ethics codes trace their origins to scandals in biomedical experimentation, cases where individual participants were harmed. Foundational guidelines like the Nuremberg Code and the Declaration of Helsinki were built to prevent that kind of harm. But researchers in the social sciences and humanities often face different ethical questions: risks to communities rather than individuals, the potential misuse of knowledge, or the power dynamics between researchers and the people they study.
Many ethical review systems used across academia are built on a foundation shaped heavily by medical research. Survey results suggest that researchers outside the medical sciences may not fully share the values baked into those systems, not because they’re less ethical, but because the dilemmas they face look different.
Take deception in research, meaning not telling participants the true purpose of a study. Medical researchers rated it close to completely unacceptable, while social scientists were notably more tolerant. One likely explanation: deception is a recognized tool in fields like psychology, where revealing a study’s true purpose can change how participants behave.
With more than 11,000 researcher responses, the study offers a rare large-scale snapshot of how researchers across disciplines think about the ethics of their own profession. The gap between how ethics reviewers judge these practices and how researchers across disciplines view them is worth watching, especially if the goal is a research culture where integrity is genuinely embraced rather than performed on a form.
Disclaimer: This article is based on a peer-reviewed survey study and reflects the findings and interpretations of the researchers involved. The results are observational in nature and measure attitudes, not actual researcher behavior. Findings from Swedish academia may not be fully generalizable to other countries or institutional systems.
Paper Notes
Limitations
The study captured what researchers believe is acceptable rather than what they actually do in practice, an important distinction the authors acknowledge. Exploratory analyses also found that most variation in attitudes arose at the individual level rather than the field level, meaning personal values and experiences likely play a larger role than shared disciplinary norms. The response rate was 33.2 percent, and while the sample closely mirrored the overall population of Swedish researchers, doctoral students were slightly underrepresented. Calibration weights were used to adjust for this. The ethics reviewer sample was relatively small at 144 respondents, limiting the precision of comparisons within that group. Responses on ethically sensitive topics may also have been influenced by social desirability bias, and the authors note this effect could be more pronounced in medicine, where a strongly rule-based tradition may make tolerant attitudes harder to admit. The study was conducted entirely within Sweden, where a centralized national ethics review authority differs structurally from institution-based systems in other countries, which may limit generalizability. The survey was also conducted before the widespread adoption of AI tools in research, which may have since shifted how certain practices, such as text reuse or automated analysis, are perceived.
Funding and Disclosures
The authors declare no known competing financial interests or personal relationships that could have influenced the work. No external funding source is specified in the provided paper materials.
Publication Details
Title: “Is research ethics discipline-specific? A survey of researchers’ and ethics reviewers’ views on research misconduct and questionable practices” | Authors: Amanda M. Lindkvist, Lina Koppel, David Andersson, Daniel Västfjäll, Gustav Tinghög | Affiliations: Linköping University (Division of Economics, Department of Management and Engineering; Division of Psychology, Department of Behavioral Sciences and Learning; Swedish National Centre for Priorities in Health, Department of Health, Medicine, and Caring Sciences); Decision Research, Eugene, OR, USA | Journal: Research Policy, Volume 55 (2026), Article 105435 | DOI: https://doi.org/10.1016/j.respol.2026.105435 | Published online: February 20, 2026. Open access under CC BY license.
