Duke University

EXECUTIVE SUMMARY AND STRATEGIC VISION

Duke University presents a robust scientific integrity profile, with an overall risk score of -0.081 indicating a performance that is well-aligned with, and in several key areas surpasses, national standards. The institution demonstrates exceptional strengths in maintaining very low-risk levels for multiple affiliations, publication in discontinued journals, and use of institutional journals, reflecting a solid foundation of ethical research practices. However, areas requiring strategic attention emerge in the medium-risk categories of retracted output, hyper-authored publications, dependency on external collaboration for impact, and the presence of hyperprolific authors. These vulnerabilities, while moderate, stand in contrast to the University's mission to uphold "high ethical standards" and "integrity." The institution's global leadership, evidenced by its top-tier SCImago Institutions Rankings in critical fields such as Medicine, Biochemistry, Genetics and Molecular Biology, Arts and Humanities, and Psychology, could be undermined if these integrity risks are not proactively managed. To fully realize its goal of providing "real leadership in the educational world," Duke University is encouraged to leverage its foundational strengths to develop targeted governance mechanisms that address these specific vulnerabilities, ensuring its outstanding scholarly contributions are built upon an unimpeachable bedrock of scientific integrity.

ANALYSIS BY INDICATOR

Rate of Multiple Affiliations

The institution exhibits a Z-score of -0.916, significantly lower than the national average of -0.514. This result indicates an exemplary and consistent low-risk profile, suggesting that the institution's affiliation practices are even more conservative than the already low-risk national standard. While multiple affiliations can be a legitimate outcome of collaboration, the university's very low rate provides strong assurance against strategic attempts to inflate institutional credit or engage in “affiliation shopping,” reinforcing a culture of clear and transparent academic contribution.

Rate of Retracted Output

With a Z-score of 0.173, the institution shows a moderate risk level, which represents a notable deviation from the low-risk national average of -0.126. This greater sensitivity to retraction events compared to national peers warrants a review of internal processes. A rate significantly higher than the average can alert to a vulnerability in the institution's integrity culture. This suggests that quality control mechanisms prior to publication may be facing challenges, potentially indicating recurring methodological issues or a lack of rigorous supervision that requires immediate qualitative verification by management to safeguard the university's reputation.

Rate of Institutional Self-Citation

The institution's Z-score of -0.771 is below the national average of -0.566, both of which fall within the low-risk category. This prudent profile demonstrates that the university manages its citation practices with more rigor than the national standard. A certain level of self-citation is natural, but the institution's lower rate indicates a healthy avoidance of scientific isolation or 'echo chambers.' This suggests that the institution's academic influence is validated by the broader global community rather than being inflated by internal dynamics, reflecting strong external engagement.

Rate of Output in Discontinued Journals

The institution's Z-score of -0.475 is slightly lower than the national average of -0.415, with both metrics firmly in the very low-risk range. This signifies a state of total operational silence regarding this risk, where the absence of signals is even more pronounced than the already excellent national benchmark. This extremely low rate indicates that the institution exercises exceptional due diligence in selecting dissemination channels, effectively protecting its research and reputation from the risks associated with 'predatory' or low-quality publishing practices.

Rate of Hyper-Authored Output

The institution registers a Z-score of 1.078, which is notably higher than the national average of 0.594, though both are in the medium-risk category. This indicates a high exposure to this risk, suggesting the institution is more prone to publishing works with extensive author lists than its peers. While legitimate in 'Big Science' contexts, this elevated rate outside of those fields can signal author list inflation, which dilutes individual accountability. It serves as a signal to distinguish between necessary massive collaboration and potentially problematic 'honorary' authorship practices.

Gap between Impact of total output and the impact of output with leadership

The institution's Z-score of 0.913 is significantly higher than the national average of 0.284, placing both in the medium-risk tier but highlighting the institution's high exposure to this particular vulnerability. This wide positive gap suggests that the university's overall scientific prestige may be more dependent on external collaborations where it does not hold intellectual leadership. This pattern signals a potential sustainability risk, inviting reflection on whether the institution's high-impact metrics are derived from genuine internal capacity or from strategic positioning in partnerships that do not fully leverage its own research leadership.

Rate of Hyperprolific Authors

With a Z-score of 0.207, the institution presents a medium-risk signal that moderately deviates from the low-risk national context (Z-score of -0.275). This finding suggests a greater sensitivity to this risk factor than its national peers. The presence of authors with extreme publication volumes challenges the perceived limits of meaningful intellectual contribution and alerts to potential imbalances between quantity and quality. This dynamic points to risks such as coercive authorship or the assignment of authorship without real participation, which prioritize metrics over the integrity of the scientific record.

Rate of Output in Institutional Journals

The institution's Z-score of -0.268 is almost identical to the national average of -0.220, demonstrating integrity synchrony and total alignment with a national environment of maximum security in this area. Both scores are in the very low-risk category. This indicates a strong commitment to avoiding academic endogamy and potential conflicts of interest. By predominantly choosing external, independent peer-reviewed journals, the institution ensures its scientific production undergoes standard competitive validation, maximizing its global visibility and credibility.

Rate of Redundant Output

The institution shows a Z-score of -0.019, positioning it in the low-risk category, which contrasts favorably with the medium-risk national average of 0.027. This demonstrates institutional resilience, as internal control mechanisms appear to be effectively mitigating a systemic risk present in the wider national environment. A low value in this indicator suggests that the university successfully discourages the practice of dividing a coherent study into minimal publishable units to artificially inflate productivity, thereby promoting the generation of significant new knowledge over sheer volume.