NUS has increasingly urged faculty members to contribute towards interdisciplinary research, recognizing many complex problems and advancements require knowledge and approaches from various domains. However, evaluating promotion and tenure (P&T) for interdisciplinary researchers can be challenging because their work may not neatly fit into the traditional disciplinary boundaries or evaluation criteria. Often such interdisciplinary research requires team effort, making it difficult to accurately attribute credit for specific contributions to individual researchers. The P&T Framework for interdisciplinary research and team science aims to create a more inclusive and equitable evaluation system that recognizes and rewards the diverse contributions of scholars working across disciplinary boundaries, fostering collaboration, innovation, and addressing complex challenges. The NUS promotion system is comprehensive and rigorous where a candidate is reviewed by external experts, department evaluation committee (DEC), head of the department, faculty promotion and tenure committee (FPTC), dean, university promotion and tenure committee (UPTC), and NUS senior management. The UPTC consists of 13 independent scholars representing diverse disciplines across NUS including STEM, humanities, and social sciences; therefore, this committee is well-positioned to recognize and assess interdisciplinary research. Also, interdisciplinary researchers are generally jointly appointed by two or more departments, thereby being evaluated by the respective heads of the departments and the deans. At NUS, interdisciplinary research is highly valued and encouraged, and its evaluation for P&T requires special consideration and attention from the reviewing parties. The following serves as a reference for reviewing parties and a guide for applicants engaged in interdisciplinary research to prepare their dossier.
The general principles of P&T apply to interdisciplinary candidates. A candidate working across disciplinary boundaries has to build a coherent research programme and needs a clear research signature that is recognized by external reviewers as well as by the primary department. Publications in top journals of several disciplines should contribute to building this coherent signature and establish a clear space in which the candidate can demonstrate leadership and obtain recognition. For instance, consider a candidate working in fields X and Y. The experts from field X justify the promotion based on the candidate’s contributions in field Y but do not recognize their contributions in field X; on the other hand, the experts from field Y justify the promotion based on the candidate’s contributions in field X but do not recognize their contributions in field Y. In such a scenario, it is difficult to justify the promotion. In contrast, it is not necessary for a candidate to establish a research signature in each of the fields to which he/she contributes. This would amount to asking interdisciplinary candidates to do double or triple the work of single disciplinary candidates. For instance, a candidate may be highly recognized in field X by importing methods/theories from field Y that resolve important questions in field X and receive invitations for keynote lectures in recognition of these contributions. But field Y may not recognize the contributions of the candidate to field X and find the research much less interesting from their disciplinary point of view. The candidate should still be promoted based on their contributions to field X. It is important for the peer review process to involve experts from relevant disciplines to assess the quality, rigor, and significance of interdisciplinary work.
The composition of the DEC should reflect the interdisciplinary nature of the candidate’s work. One of the DEC members may be chosen from another department related to the field of which the candidate’s research contributes.
Similarly, the candidate and the DEC should choose two types of external reviewers (ER): (1) ERs with a clear anchoring in the candidate’s primary discipline (this is important to collect views on the candidate’s research signature) and (2) ERs who also engage in similar interdisciplinary research. Their primary discipline could be the same as the candidate’s primary discipline, or other disciplines to which the candidate contributed. It is not advisable to solicit external reviews from researchers who are exclusively working in the secondary disciplines to which the candidate contributes, as this would again amount to requiring the candidate to establish research signatures in each field to which he/she contributes.
The publication profile of an interdisciplinary researcher may be atypical vis-à-vis their primary or secondary disciplines. Reviewing parties need prioritize assessing the quality and impact of the candidate’s work in an interdisciplinary context. While the quality standards for publications should remain consistent whether a candidate conducts disciplinary or interdisciplinary research, it is essential to establish the quality tiers of publications in each of the disciplines involved. Department benchmarking reports from the respective disciplines provide a good starting point. The reviewing parties may establish equivalence in terms of quality among outlets from different disciplines. This could involve comparing journals within a particular discipline to reputable cross-disciplinary journals (e.g., Science, Nature, Proceedings of the National Academy of Sciences), between journals of different disciplines, or between different forms of outlets (e.g., journal publications and conferences). Using absolute impact factors as a metric is discouraged, as lower-ranked journals in one discipline can have higher impact factors than top journals in other disciplines. When assessing impact, there is a risk of undervaluing interdisciplinary articles through over-reliance on citation counts. For example, social scientists regularly publish important articles in top medical journals such as the New England Journal of Medicine (NEJM). However, the bibliometrics of these papers may not match with medical papers that are cited with higher velocity than in the social sciences. Similarly, choosing peers for benchmarking poses a challenge. The most suitable benchmarks are researchers with a similar interdisciplinary profile, as they experience the same citation patterns. Comparisons with NUS-internal peers who were recently promoted in the candidate’s primary or secondary department should be approached with caution unless these comparators also similarly engaged in interdisciplinary research.
Team science is a collaborative effort between two or more independent researchers to address problems that require multiple areas of expertise because of the interdisciplinary nature of a project. In general, these projects lead to publications with multiple authors, including first, last and corresponding authors, in which it is important to ascertain the contribution of each author for the purpose of P&T decisions. The guiding principle here is that the authorship order should always be judged according to the disciplinary norms.
The weight depends on the contribution analysis, and it decreases with the number of authors on the publication.
In some domains, authorship order is not necessarily based on contributions. For example, it is common in the Humanities and Social Sciences, Computer Science etc. for students and junior researchers to be placed first, or for authors to be sequenced by alphabetical order. It is important that individual contributions are identified in such cases. The candidate should provide a clear statement of their contributions in these publications. If a reviewing party is in doubt, a statement from the other team authors might be requested. Otherwise, the guidelines provided in this section (“Authorship order based on nature of contribution”) will apply.
The citation rates of publications are positively associated with the number of authors; in extreme cases, publications with hundreds of authors can result in very high citation metrics. Participation in such publications without being a first/last/corresponding author suggests marginal involvement. Reviewing parties should therefore evaluate citation metrics relative to the number of authors.
The same principles outlined above should apply to other research contributions, including papers in conference proceedings, invention disclosures, patents, designs, and models.
The most critical measures of successful interdisciplinary research are the generation of cutting-edge innovations that: