What is research integrity? Why does it matter?
Research integrity refers to the quality of the research process. Ideally, research should be conducted honestly and reliably. We should perform any scientific experiment, debate, or peer review without any conflicting interest, and in good faith.
Peer review and research integrity
Without research integrity, the science that is added to the existing body of knowledge will be compromised. False science can lead people to drawing the wrong conclusions. This can be disastrous for science and for humanity.
Peer review is tightly bound with research integrity. That is why publishers – including Bentham Science – have created systems to ensure robust peer review.
Here are eight common research integrity issues that all peer reviewers and authors need to keep in mind.
1. Plagiarism and Self-Plagiarism
Plagiarism means using someone else’s ideas, words, or work as your own.
In scholarly research, authors sometimes fail to properly credit their original source of information. This can mislead readers and compromise the integrity of authors’ work. For this reason, plagiarism is a breach of research ethics.
Self-plagiarism occurs when an author reuses their own previously published material without proper citation or disclosure.
Unintentional plagiarism
Unintentional plagiarism can occur when we are careless or we just don’t know the rules of citation. Even when unintentional, plagiarism has significant consequences. Authors may damage their scholarly reputation. They could be banned from institutions and journals.
The impact of plagiarism
In the context of peer review, plagiarism raises alarm bells about the credibility and originality of a manuscript. It undermines the trust that peer reviewers and readers place in the research community. To safeguard research integrity, authors must diligently cite sources and follow ethical guidelines, ensuring that their work contributes to the advancement of knowledge without copying text in prior research.
2. Data and Image Manipulation
Data and image manipulation in research refer to the deliberate alteration or fabrication of research data, figures, or images to misrepresent results, either by enhancing or distorting them.
What are the implications of data manipulation?
Researchers may manipulate data or images to achieve desired results, secure funding, or gain recognition. Such actions distort the scientific process, and allow the propagation of false conclusions.Fake science can have profound consequences for scientists and society at large.
Examples of data and image manipulation
Peer reviewers depend on the accuracy and authenticity of data to evaluate the validity and significance of a study. When data is manipulated, reviewers are misled. This can lead to the acceptance of flawed or fraudulent research.
Data manipulation can be hard to direct. It requires a good eye, especially in the case of image manipulation. Fortunately, there are tools to detect image manipulation.
Catching manipulated data like statistics can be harder. Often, such a discovery is made after a paper is published. Detecting data manipulation as a peer reviewer is a broad topic that we will explore in another article.
Numerous high-profile cases of data and image manipulation have shaken the scientific community, for example:
- Dr. Andrew Wakefield’s infamous research on the MMR vaccine. Wakefield had falsely linked MMR to autism, based on manipulated data.
- Hwang Woo-suk’s stem cell scandal, where fabricated data led to the retraction of multiple papers. This case damaged the credibility of the stem cell research field.
These cases underscore the critical importance of rigorous data and image integrity in research.
3. Questionable Citations
Questionable citations present a critical challenge in the peer review process. Authors may, intentionally or unintentionally, misrepresent the sources they cite. This can distort the credibility and validity of the entire research paper.
Peer reviewers face the difficult task of finding and fixing these issues to maintain the integrity of scholarly research. They must ensure that scientific literature accurately reflects the contributions of the researchers in the peer group.
How do questionable citations challenge peer reviewers?
Detecting such inaccuracies demands meticulous cross-referencing and a deep familiarity with the cited literature.
Peer reviewers need:
- a broad understanding of the relevant literature.
- to assess the accuracy of citations in submitted manuscripts.
- to effectively identify questionable citations
- the diligence to cross-reference cited works for scientific correctness
For editors and publishers, matching the right experts for peer review is critical.
4. Paper Mills
Paper mills are publishing enterprises that churn out large volumes of low-quality, plagiarized, (or even entirely fabricated) research papers. These fake papers are then sold to researchers and students looking for an easy path to publication or academic success.
There is great pressure within academia to “publish or perish”. This can press authors to ignore research integrity and ethics. Paper mills exploit this pressure.
Challenges for publishers and peer reviewers
Paper mills flood the research landscape with subpar or counterfeit work. This deluge of fake research poses a substantial challenge for journals. Peer reviewers must sift through a growing list of questionable articles and try to identify genuine contributions.
Furthermore, paper mills sometimes find ways to get around plagiarism detection software. This makes it harder for reviewers to spot and reject illegitimate submissions.
As a result, peer reviewers must remain vigilant. They must use a combination of skills, experience, and available tools to ensure the integrity of scholarly publishing.
5. Problematic Peer Review
In some cases, the peer review selection process is flawed. This can distort the evaluation of research manuscripts. It may undermine the objectivity and fairness of the peer review system.
Known peer review problems
- Reviewer bias may lead peer reviewers to unfairly favor or disfavor a manuscript based on personal opinions or preconceptions, rather than evaluating it objectively.
- Conflicts of interest put commercial interests of reviewers before objectivity. It is important to check for corporate or academic affiliations of potential candidates before assigning them an article.
- Breaches of confidentiality can introduce bias and other ethics issues into the peer review process
Peer review rings
Sometimes groups of researchers try to covertly manipulate the peer review process. Authors arrange to have their own work reviewed favorably by fellow members. This type of scam is called a “peer review ring”, “peer review cartel”, or “buddy system”.
Members may use peer review rings to pass questionable or dangerous content into scholarly journals. Often, the goal is to give credit to incompetent researchers. Or, they may try to push viewpoints that can be used to justify harmful actions. This happens at the expense of honest scientific reporting.
There have been many cases of peer review rings – including cases where the reviewer conducted their own peer review.
However, publishers have been taking measures to stop these scams:
- requiring verified email addresses for access to publisher interfaces
- requesting reviewers with diverse affiliations (different countries and institutions)
- double-blind peer review (where both authors and reviewers are anonymous)
- clear guidelines for ethical peer review
- the use of digital tools to detect potential conflicts of interest and unusual online activity.
Publishers also encourage researchers to anonymously report suspicious peer review practices to journal editors and institutions.
6. Salami Slicing of Research Results
Salami slicing refers to the practice of dividing a single comprehensive study into smaller, often insignificant, parts and publishing them as separate research papers. The intention is to artificially increase the number of publications, exaggerating the apparent productivity of the researcher.
Salami slicing means dividing one comprehensive study into smaller parts, then publishing each part as a separate paper. This artificially increases the number of publications and makes the researcher seem more productive than they really are.
Ethical concerns of salami slicing, and implications for peer review
Salami slicing distorts the true value and contribution of the research. It floods academic journals with potentially low-quality or repetitive work. This dilutes the quality of the literature.
Salami slicing can cause peer reviewers to make biased judgments. Reviewers may have fragmented insights and an incomplete understanding of the overall research.
Strategies for avoiding salami slicing
Researchers should aim to publish complete, substantial studies that represent a significant contribution to the field.
Journals can discourage salami slicing by:
- promoting quality over quantity
- considering the completeness and significance of the research when evaluating submissions
- rejecting papers that lack substantial content or meaningful contributions.
Academia should promote collaborative research and encourage comprehensive reporting. This can help to prevent salami slicing.
7. Fake Authorship
Another known issue in research integrity is fake authorship. Two common types are “gift” and “ghost” authorship.
Gift authorship
In many fields, researchers work in multidisciplinary teams to complete a project, which is reported as a journal article. Such studies have multiple contributors that are credited as authors.
The problem here is that researchers sometimes add senior colleagues or friends to the list of authors – despite them doing almost none of the work.
Some authors are only included to boost their publication lists. As a result, you might see their name appear in a list of articles that belong to a discipline outside their expertise – bonus points to them if the paper is cited.
Ghost authorship
Other times, articles are ghostwritten by research assistants, who never receive any credit for their work.
Making authorship more transparent
Publishers are increasingly incorporating guidelines to improve transparency in reporting and attributing contribution to work. The Committee on Publications Ethics (COPE) has a specific set of authorship guidelines, which many journals are implementing.
8. Using AI in Article Writing
In recent years, artificial intelligence tools have been increasingly used to create content, including text-based content and images. The launch of ChatGPT in 2022 opens the floodgates of AI-derived articles, and research is no exception. Academics are short on time, and AI tools help in quickly preparing and summarizing paragraphs on information albeit with some text plagiarized from sources in the training datasets and the internet. The same is true for image content — although the use of AI-upscaling is beneficial for improving image quality.
The pros and cons of AI
Many AI technologies are based on Open Source code. Anyone who uses the tools should attribute them correctly. But it has become acceptable to use AI for routine tasks.
AI tools like ChaptGPT need to be considered seriously when reviewing articles. Researchers are expected to submit their own original content, yet AI tools are antithetical to this.
Academics have debated whether AI should be considered as an author (researchers have shown their disapproval). Some consider it plagiarism, while others feel it is malpractice.
Ensuring the ethical use of AI
Journals — including Bentham Open journals — have released updated authorship guidelines for the use of AI tools that are intended to assist reviewers and implement ethical peer review practices.
Conclusion
Research integrity is vitally important. It guards the very foundations of scholarly progress and knowledge dissemination. Many challenges remain, and AI has added another dimension of complexity to the issue.
Peer review as protection
Peer review plays a critical role in protecting research integrity.
At Bentham Science, our team works hard to strengthen our peer review systems and to encourage ethical research.
All stakeholders – researchers, associations, institutions, and funders – must work together to ensure that science is conducted and published reliably and honestly.