What is scientific integrity? 

The term 'scientific integrity' describes a set of universal principles and good practices that researchers in any discipline should follow to conduct responsible and honest science.  

These principles and practices are detailed in various codes of conduct and statements at the institutional, national or international levels. They are not laws, but guidelines that researchers should adhere to. Some important texts include:  

• The European Code of Conduct for Research Integrity, developed by ALLEA, the European Federation of Academies of Sciences and Humanities. The latest version is available here, and the previous version from 2017 was translated into Spanish and Catalan. 
• The European Charter for Researchers, which focuses on the relationships between researchers and the institutions that employ them, was signed by over 1,400 institutions, including 318 Spanish ones.  
• The Best Practices for Ensuring Scientific Integrity and Preventing Misconduct, issued by the OECD in 2007, provide a list of scientific misconducts and institutional responses to prevent or address them.  
• In Spain, the National Declaration on Scientific Integrity, signed in 2015 by CSIC, COSCE (Confederation of Spanish Scientific Societies), and CRUE (Conference of Rectors of Spanish Universities), outlines general principles.  
• CSIC's Code of Good Scientific Practices, revised in 2021. 

In European Union funding programs, the acronym RRI (Responsible Research and Innovation) is used to describe science that engages with the public and aligns with societal values. These aspects partially overlap with the concept of scientific integrity.  

Is ethics the same as integrity? 

"They are related, but very different concepts," says Jordi Camí, President of the Spanish Committee on Research Ethics in an interview published in July on the El·lipse website of the Barcelona Biomedical Research Park.  

Ethics seeks agreements in the face of moral dilemmas; it requires researchers to pay attention to the effects of their work on other people, animals, and society in general. Integrity refers to research practice itself, and to rules within the scientific community: "to be honest and not cheat," as Camí puts it.  

What specific practices are we talking about? 

The most significant types of misconduct are known internationally by the English acronym: Fabrication, Falsification, and Plagiarism (FFP). These can be considered as the 'three deadly sins' of scientific research. 

• Fabrication: Inventing data or observations that don't match the reality of experimental work or data collection. 

For example, in May 2023, McMaster University in Canada published the summary of its internal investigation into the misconduct of Jonathan Pruitt, a biologist specialised in spider behaviour. It concluded that Pruitt "engaged in fabrication and falsification with respect to whether spiders were collected for the study conducted and concerning which spiders were used and whether the assays were conducted to support the papers.”

• Falsification: Manipulating research materials or equipment, altering or omitting data or results. 

For instance, in 2017, the University of Tokyo concluded that a research group studying chromosome dynamics during cell division had falsified data and images in five articles published in prestigious journals. "The problems include image alteration and graphs with data apparently not derived from the reported experiments," Science wrote in a news article at the time.

• Plagiarism: Appropriating ideas, words and/or results from another person without giving them credit. 

Physicist Ranga Dias, whose 2020 publication on room temperature superconductivity was retracted in 2022, was also accused of plagiarising sources in several sections of his doctoral thesis, published years earlier. The suspicions of plagiarism suggest that the researcher "[doesn't] mind cutting corners; that they may not have as many original ideas as they're presenting themselves to have," Lisa Rasmussen, a research ethicist at the University of North Carolina in the US, told Science magazine.

There is another acronym often used to describe to describe less severe, and likely more frequent behaviours: Questionable Research Practices (QRPs). Their definition is not as clear-cut, but QRPs could include, for example: 

• Insufficient review of the existing scientific literature. 
• Selecting favorable results and ignoring inconvenient data (cherry-picking). 
• Self-plagiarism. 
• Fragmenting results across various scientific articles to inflate publication statistics (salami-slicing). 

On the contrary, ideal behaviour is often called "Responsible Conduct of Research" (RCR).  

The table below summarises different types of misconduct. 

Source: Best Practices for Ensuring Scientific Integrity and Preventing Misconduct - OECD, 2007

How do we find out about research misconduct?

Aside from a handful of scandals that appear in mainstream media, most cases of misconduct are discussed and litigated within the research system itself.

One valuable communication channel for journalists is the retraction of publications. The research community shares its results in peer-reviewed journals: before being published, studies are reviewed by other researchers with similar expertise to the authors, and who have not participated in the work they review. This internal system ensures, in principle, the quality and rigor of scientific publications. When —despite this filter— misconduct affecting the value of a published study is established, that publication is retracted. In general, withdrawn articles remain in the scientific literature with an editorial note indicating their status. However, these brief notes often do not explicitly state whether it is an unintentional error or a real fraud.

Retractions can be a valuable source for journalistic work, especially when they are repeated cases for an individual or research team; they alert us to suspicious results and, possibly, to misconduct behind them.

The website Retraction Watch, created by two US journalists in 2010, follows many cases of retractions and maintains a database that has already served as a basis for journalistic and academic research.

Retractions are the official and most visible part of the iceberg; before an article is withdrawn, it will have been discussed among scientists and with journals. According to data analysed by Cristina Candal, who researches research integrity issues at the University of Santiago de Compostela, the average time from publication to retraction of an article is 487 days.

In recent years, such debates have been taking place more openly, though often anonymously, on online forums such as PubPeer, where researchers critique thousands of recent or older scientific publications.

How often does misconduct occur? 

In a recent study on the experiences of researchers in biomedicine in Spain, 7 out of 10 respondents admitted to knowing cases of scientific misconduct firsthand, and 4 out of 10 acknowledged intentionally engaging in some form of malpractice. The most frequent type of misconduct was false authorship of scientific articles: 35% of the 403 respondents said they had participated in it, according to the study published in the journal Accountability in Research. 10% reported a lack of informed consent, and 3.6% admitted to having participated at least once in data falsification or manipulation.   

Candal, a co-author of this article, also analysed in another study the 374 retractions of publications with an author affiliated with a Spanish institution until 26 January 2023. The results show increase in retractions, from less than 1 case per 10,000 publications in the year 2000 to 5 retractions per 10,000 publications in 2023. These data were presented at the Annual Meeting of the Spanish Society of Epidemiology in Porto in September of 2023. 77% of these retractions occurred in biomedical research. In this field, duplication —publishing the same results in two different publications— was the leading cause of retraction (35% of cases). In other disciplines, plagiarism is the most common type of misconduct (45.8% of retractions). Understanding the causes can help to design prevention strategies adapted to the national context, Candal said at the conference. 

What leads researchers to engage in misconduct?

There are two main explanations, one systemic and the other individual.

The first explanation recognises that the research system pushes scientists to publish a large number of articles, which is crucial to advance their careers and secure funding for their projects. This pressure, known as publish or perish, sucks researcher into a machinery and "there is always someone who takes a step they shouldn't take” to meet the expectations of their profession, Eduard Aibar, professor of Science and Technology Studies at the Universitat Oberta de Catalunya (UOC), tells the SMC Spain.

The second explanation is known as the ‘bad apples’ theory, to describe some individuals with a compromised personal sense of integrity who commit reprehensible acts. “Like in all professions, there are people who cheat,” says Cristina Candal to the SMC Spain. “It's a multifactorial problem, not just systemic. There are also [individual] factors such as ego and greed.”

Manzanas Podridas (Bad Apples in English, 2019) is also the title of a Spanish book written by Ángel Abril-Ruiz and available for free online. The book provides details to understand the systemic context of scientific integrity and offers a review of various high-profile cases of fraud and misconduct, such as that of Spanish biologist Susana González in 2017, German physicist Jan-Hendrik Schön in 2002, or Dutch psychologist Diederik Stapel around 2012.

How are cases of misconduct investigated and punished? 

Allegations of misconduct are usually investigated through internal procedures organised by universities or research centres themselves. In cases of proven fraud or misconduct, sanctions vary depending on the severity of the case and the decision of the institutions involved. A researcher might be excluded from certain activities or projects for a given period; or may lose their job entirely; or a funding agency might rescind its grants.  

There is no central organisation that regulates researchers' right to work, as there is in medicine. There are cases of medical researchers who lost their medical license due to research integrity violations. For example, Andrew Wakefield was struck off the medical register in the UK in 2010 after his publication that wrongly linked the MMR vaccine to cases of childhood autism was retracted.  

The most serious cases of fraud can go to trial and have penal consequences for researchers who are found guilty. In 2009, the South Korean justice system sentenced biologist Hwang Woo-suk, a specialist in cloning embryonic stem cells, to a suspended two-year prison sentence for fabricating data and misappropriating public funds. 

Aibar, who studies institutional responses to misconduct, deplores that investigation processes are usually opaque, as institutions often protect their reputation. “Absolutely nothing is published” about internal investigations, he tells the SMC Spain. He adds that people who report possible breaches of integrity (known as whistleblowers) do not receive the necessary protection and may suffer negative consequences in their personal and professional lives. 

In recent years, national offices for scientific integrity have been established, such as the French Office of Scientific Integrity, founded in 2017e. The following year, the Committee for Research Integrity in Catalonia (CIR-CAT) was created. At the national level, in 2023, the Spanish Committee on Research Ethics was launched. However, unlike the Office of Research Integrity, established in the United States in 1989 in the field of public health, the vast majority of these organisations have a mere advisory role: they issue recommendations and have no power to investigate or sanction misconduct.  

Many researchers advocate for the creation of an independent office with the authority to investigate and punish misconduct, and that could also act as a deterrent. “This would be especially important in the case of research funded with public funds, as in that case, research fraud would amount to the misappropriation of public money”, says an article published in the Gaceta Sanitaria in November 2022.  

How does scientific misconduct affect the public? 

Research misconduct leads to incorrect results and unfounded interpretations. In the case of biomedical research, misconduct can also have direct consequences on human or animal health. All forms of malpractice, even the mildest ones, contribute to a science that is not robust and cannot be replicated by other research teams. They impact public trust in science, misuse money and other resources, and may lead—either directly or indirectly—to unfounded decisions, for example, in the design of public policies. 

Why cover topics of scientific integrity in the media? Wouldn't it be better to focus on interesting discoveries and positive news? 

Some people believe that the role of science journalism is to promote science and support researchers who want to communicate their work; this position is known as cheerleading in English. However, one of the functions of science journalism is to critically examine science, with its bright and dark sides, just like journalism in any other field. For example, a sports journalist not only echoes the results of the matches, but also can and should report critically on sports organisations, as well as on the context and impact of sports at the social, economic, and political levels. In this resource, we explain some of the differences between science communication and journalism. 

In his book Manzanas podridas, Abril-Ruiz writes: “The journalist's job - investigating, scrutinising, and bringing to light the dark intricacies of organisations - appears publicly as a necessary mechanism to keep the dishonest behaviours of organisations and individuals in check." He adds: “(Mediocre) institutions only seem to take action when a case of misconduct becomes public knowledge; if it doesn't become public, they will most likely to try to cover it up."