Look at this!

My colleague was pointing at the screen. One profile in the publication report for 2025 stood out from the others: a researcher at UiA was listed as a contributor to an unusually high number of publications.

To give some perspective, here is the publication distribution within the researcher’s faculty in 2025:

• 77.3% of researchers produced 1–2 publications
• 14.2% produced 3–5 publications
• 5% produced 6–9 publications

The number of publications associated with the researcher on the screen far exceeded the norm, prompting us to take a closer look.

At first glance, everything seemed compliant. Each publication was registered in the Nasjonalt vitenarkiv (NVA), each publication channel met the required standards, and every record passed the automatic thresholds for inclusion in the NVI reporting list.

Yet when we dig a little deeper into cases like this, we realise that our current routines for quality assurance are not as watertight as they might seem.

Everyday quality control

At the University Library, we are responsible for validating the quality of the research registered in NVA, the national repository for research carried out across Norway.

Each year, publications are automatically included in the annual NVI reporting list when they meet the following two criteria:

1. The work is published in an approved Level 1 or Level 2 channel listed in the Kanalregister.
2. It belongs to one of the five eligible categories:
   • Academic article
   • Academic review article
   • Academic chapter
   • Academic monograph
   • Academic commentary

The library reviews every single entry in the reporting list to approve or reject it. Once the reporting window closes, the finalised data is transferred to the Database for Statistics for Higher Education (DBH), which tracks research output across Norway.

Most of the decisions in this process are straightforward. Since the Kanalregister already verifies peer-review practices, we rarely need to reassess whether a work meets scholarly standards. Instead, most of our effort goes into catching obvious reporting errors, such as:

Reused articles: Articles republished in formats such as Festschriften, anthologies, or collected works, where the original publication has already been counted in a previous reporting cycle.

Revised editions: Updated or republished books that do not qualify as new academic contributions.

Misclassified research: Editorials or professional articles mistakenly registered as academic articles, or standard introductory chapters labelled as academic chapters (which are only eligible if they provide new scientific knowledge and have undergone peer review).

These are the clear-cut cases. Unfortunately, the system has its blind spots, allowing some questionable research results to slip through the cracks of our quality assurance checks.

The blind spots

The case described in the introduction highlights one of the most persistent blind spots in research reporting: co-authorship and the challenge of determining genuine author contribution.

Contemporary research is often collaborative. Co-authored articles can include anything from two to thousands of authors across multiple institutions around the world. While the reporting system recognises every listed author, it does not measure the actual extent of each individual contribution.

From a registration perspective, authorship is binary: you’re either registered as an author or you’re not. In practice, however, contributions exist on a spectrum, ranging from leading the research to executing a narrowly defined task connected to the project. This tension becomes more pronounced when publications provide little or no information about each author’s specific role (and, of course, the recent trend of listing AI tools as co-authors complicates matters even further).

To regulate this complexity, many institutions (including UiA) look to the Vancouver Recommendations. These guidelines state that authorship must be based on all of the following four criteria:

• Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; AND
• Drafting the work or reviewing it critically for important intellectual content; AND
• Final approval of the version to be published; AND
• Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

The Vancouver Recommendations underscore the importance of authorship, not least to give credit where it is due and, when needed, to hold researchers accountable. Still, even with these guidelines in place, we continue to encounter issues in NVA, such as:

Authorship inflation: Large collaborative networks can generate high publication counts for individual researchers.

Ambiguous roles: Contributor descriptions are often too limited to allow meaningful verification of individual contributions, particularly in publications with large author lists.

Cross-institutional barriers: Many collaborations extend beyond UiA, limiting the information that is locally available to us.

System reliance and trust: The validation process relies heavily on researchers registering their publications accurately and responsibly.

The trouble is that none of these factors are inherently indicative of academic misconduct. As curators of UiA’s research records in NVA, the library is well positioned to identify unusual patterns and raise concerns when something appears out of the ordinary. Yet many of the underlying issues extend beyond our mandate and expertise. Authorship norms vary considerably across disciplines, and even with the Vancouver Recommendations as a common reference point, there remains uncertainty about where meaningful contribution ends and authorship begins.

This raises broader questions about what can be done in response to these grey areas. Should emphasis be placed on developing more robust regulations within the system, such as mandatory authorship statements or automated alerts for unusual publication patterns? Or is there a greater need for cross-disciplinary education and discussion around authorship and research integrity?

The library cannot resolve these challenges alone, but it has an important role to play. We can be attentive to patterns that warrant closer scrutiny, continue refining our quality assurance practices, and contribute to ongoing conversations about responsible research reporting. As awareness of these challenges grows, so too does our ability to detect weaknesses in the system and respond appropriately when they are exploited.

With this in mind, let’s return to the case that prompted this reflection.

The outcome of the case

Following a deeper investigation into the circumstances surrounding the researcher with the unusually high publication rate in the 2025 reporting cycle, a decision was reached: publications registered under this researcher in 2026 will not be eligible for approval within the NVI reporting process.

Cases like this serve as a stark reminder for our team and the university at large. Quality assurance is not simply a box-ticking exercise. It is a constant effort to protect the credibility and integrity of our institution’s research within a system that necessarily relies on trust, professional judgement, and shared standards.

When it comes to quality assurance, there will always be blind spots. Part of the library’s role is to keep looking for them.

AI was not used for any part of the writing process for this blogpost