A study published in the journal Accountability in Research in 2026, titled “Faculty-student differences in authorship perceptions before and after authorship ethics interventions” (DOI: 10.1080/08989621.2026.2651757), adds new survey evidence to a long-standing concern in authorship ethics: that faculty and students frequently do not agree on what counts as an authorship-worthy contribution — and it goes a step further than most prior work by testing whether a sustained, campus-level ethics-education effort actually narrows that gap.
What the study set out to measure
Per the published abstract and journal listing, the research surveyed faculty and students on prevailing authorship norms, the frequency of authorship disagreements, and how perceptions of appropriate authorship credit vary by role (faculty vs. student), gender, and disciplinary context. This design puts it in the same broad tradition as earlier faculty-student authorship research — including the U.S. Office of Research Integrity’s long-running guidance on authorship in faculty-student collaborations — which has repeatedly flagged the power asymmetry between a supervising faculty member and a student contributor as a structural risk factor for authorship disputes, independent of anyone’s bad intent.
What distinguishes this study is its second phase: rather than only documenting a gap, the researchers report on a follow-up survey conducted after a three-year, campus-based effort to improve authorship ethics. That effort combined two interventions that are common individually across research institutions but less often evaluated together over a multi-year window: structured training on ethical authorship practices, and the adoption of a formal, written institutional authorship policy.
What the findings suggest — and what remains unconfirmed
According to the published abstract, the baseline survey found systematic differences between faculty and students in how authorship distribution is perceived and how different types of researcher contributions (for example, data collection versus manuscript writing versus study design) are weighted toward earning authorship credit. The follow-up survey, administered after the training-plus-policy intervention, reportedly showed measurable shifts in participants’ awareness, expectations, and attitudes toward authorship ethics and responsibilities.
A note on precision: the full study sits behind a publisher paywall at the time of this writing, so this page relies on the publicly available abstract and journal listing rather than the complete dataset. We have deliberately avoided citing specific percentages, sample sizes, or statistical-significance figures because we could not independently verify them against the full text. Readers who need the underlying numbers — effect sizes, exact survey items, discipline-by-discipline breakdowns — should consult the primary source directly rather than rely on secondary summaries, including this one.
With that caveat, the study’s own framing is that proactive education and policymaking can promote integrity in collaborative scholarly work and help recalibrate local authorship norms over time — a more optimistic, and more rigorously tested, claim than most institutions are able to make about their own research-integrity training, which is rarely evaluated with a pre/post design spanning several years.
Why the faculty-student gap keeps showing up
The persistence of a faculty-student perception gap is not a new finding on its own. Related work — for instance, a 2020 Science and Engineering Ethics study on undergraduate research experiences found authorship expectations were frequently “not taught and not caught” in day-to-day mentoring — has repeatedly pointed to the same underlying mechanism: authorship criteria are often assumed to be common knowledge rather than explicitly taught, and the person with the least power in the relationship (typically the student or junior researcher) is the one least likely to raise a disagreement before submission.
This is precisely the dynamic CASRAI’s own guide on student-advisor authorship disputes and power imbalance addresses: without an explicit, criteria-based framework agreed early in a project, ambiguity tends to resolve in favor of whoever has more institutional authority, not necessarily whoever made the qualifying contribution. Formal criteria — such as the four-part test in the ICMJE authorship criteria — exist precisely to replace informal, role-based assumptions with a contribution-based standard that can be checked and discussed openly.
What this means for research-integrity offices and PIs
- Don’t assume shared understanding. If faculty and students in the same lab can hold systematically different views of what “deserves” authorship, an unstated or purely verbal authorship convention is a documented risk factor for later disputes, not a neutral default.
- Written policy plus training, not either alone. The intervention evaluated in this study paired a formal institutional policy with active training — consistent with the general research-integrity principle that policy documents alone rarely change behavior without accompanying education, and training alone rarely sticks without an institutional reference point to point back to.
- Discuss authorship early, and revisit it. Both the ICMJE framework and CASRAI’s guide on resolving authorship-order disputes recommend agreeing on anticipated contributions — and hence anticipated authorship — before a project is well underway, then revisiting that agreement if actual contributions diverge from what was planned.
- Discipline matters. Because the study explicitly examined disciplinary context alongside role and gender, it reinforces a point CASRAI’s guide to authorship-order conventions by discipline makes independently: there is no single universal authorship norm across fields, which makes an institution-wide, criteria-based policy more useful than reliance on informal disciplinary custom, especially in interdisciplinary collaborations.
Where this fits in the broader authorship-integrity landscape
The study sits alongside a growing body of applied research-integrity work testing whether structured education actually changes authorship behavior, rather than simply raising awareness of the problem. A related 2026 report in the same broad literature — an educational intervention for graduate students that paired online training with a structured authorship-agreement form used in mentor-mentee discussions (indexed on PubMed / PMC) — points in a similar direction: structured, document-anchored conversations about authorship, introduced early and reinforced institutionally, are one of the few interventions in this space with any prospective, pre/post evidence behind them, as opposed to relying solely on cross-sectional surveys of attitudes at a single point in time.
For CASRAI’s audience of research-integrity officers, graduate program directors, and PIs, the practical takeaway is less about this single study’s exact effect size and more about the design pattern it tests: a written, contribution-based authorship policy, reinforced by training, evaluated with a real before/after comparison rather than assumed to work. That combination — policy plus training plus follow-up measurement — is a reasonable template for any institution trying to move authorship practice from informal custom toward an explicit, defensible standard such as the CRediT taxonomy or the ICMJE criteria.
Frequently asked questions
Is this a peer-reviewed study?
Yes — it is published in Accountability in Research, a peer-reviewed journal covering research ethics and integrity, via Taylor & Francis (DOI 10.1080/08989621.2026.2651757).
What exactly was the “intervention”?
Per the abstract, it combined training on ethical authorship practices with the adoption of a formal, written institutional authorship policy, sustained over roughly three years before the follow-up survey.
Does the study give exact numbers on how much perceptions changed?
The publicly available abstract describes “notable shifts” in awareness, expectations, and attitudes without specifying figures in the summary itself. This page does not report specific statistics because we could not verify them against content behind the journal’s paywall — consult the full article for exact results.
How does this relate to CRediT or ICMJE criteria?
Both are examples of the kind of explicit, contribution-based authorship standard this research suggests institutions should be actively teaching and reinforcing, rather than assuming researchers absorb informally. See CASRAI’s guides to the ICMJE authorship criteria and the CRediT contributor roles taxonomy.
Editorial note: this page summarizes findings as described in the study’s published abstract and journal listing. The full text was not accessible for independent verification at the time of writing; specific statistics have been deliberately omitted or hedged rather than estimated, per CASRAI’s sourcing standards.
