Beyond authorship: Analyzing disciplinary patterns of contribution statements using the CRediT taxonomy

This is a preprint. It is undergoing peer review. The final published version may differ. 1 Beyond authorship: Analyzing disciplinary patterns of contribution statements using the CRediT taxonomy Elvira González-Salmón1, Victoria Di Césare1*, Aoxia Xiao2 and Nicolas Robinson-Garcia1 1Unit for Computational Humanities and Social Sciences (U-CHASS), University of Granada (Spain) 2School of Information Management, Wuhan University (China) *Corresponding author: [email protected] Abstract In this research article, we present the first cross-disciplinary descriptive analysis of the use of contribution statements. Our main objective is to obtain further insight on contributions by a variety of fields (Multidisciplinary, Health, Life, Physical, and Social Sciences) from the largest dataset used up to now. We examine more than 700,000 articles published between 2018 and 2023 in Elsevier and PLOS journals, in combination with bibliometric data extracted from the Scopus database. The descriptive analysis of the dataset focuses on the overall coverage of the merged data, the distribution of authorship and disciplines at paper level, and the interactions between contribution statements, author order, and disciplines. Our two main findings indicate that, on the one hand, looking at contributions and authorship order can enrich the way we understand science as a social endeavor. On the other hand, delving deeper into contributorship differences by field is key. We underscore the value of the Contributor Role Taxonomy (CRediT) in unveiling nuanced research dynamics and offering a more equitable framework for evaluation. Keywords: CRediT; Authorship; Contributorship; Division of labor; Contribution statements Introduction Team science is on the rise (Wuchty et al., 2007), challenging previous conceptions of what accounts for science and authorship (Birnholtz, 2006). This may require new norms that are prepared to deal with the new structure of science (Jabbehdari & Walsh, 2017) and has already led many to suggest replacing the concept of author with that of contributor, which acknowledges the distributed and collaborative nature of science as it is conducted in the 21st Century. Since their introduction in biomedical journals by the end of the 1990s (Rennie et al., 1997; Rennie et al., 2000), contribution statements are becoming more widespread in academia, especially thanks to the launch of the This is a preprint. It is undergoing peer review. The final published version may differ. 2 Contributor Roles Taxonomy (CRediT) (Brand et al., 2015). This taxonomy was developed to homogenize contribution statements across publications and facilitate discussions about authorship that help avoid author disputes (Allen et al., 2019). Since its launch, CRediT is now used by more than 40 different publishers across a wide range of disciplines (https://credit.niso.org/). Their expansion not only has the potential of overcoming the limitations of using authorship as a “credit” in research assessment but can help us understand how science is done, how researchers distribute tasks, how this influences author order, and how it relates to their prestige as reflected in the author byline of publications (Haeussler & Sauermann, 2013). Contributorship has been studied for some years now. Cronin et al. (2003) traced them even before they appeared in a special section within research articles by looking into the acknowledgements for Psychology and Philosophy throughout the 20th Century. Drawing on a multidisciplinary dataset of more than 80,000 documents, Larivière et al. (2016) examined the relationship between division of labor, contribution types, and authors’ seniority to provide evidence on the existence of conceptual contributions made by senior researchers and technical tasks performed by younger scholars. Later, Larivière et al. (2021) updated their analysis using CRediT for a set of more than 30,000 PLOS papers and studied the distribution of contributions across teams. Other relevant works have applied contributions to look into credit allocation (Ding et al., 2021; Sauermann & Haeussler, 2017), credit bias (Matheson, 2022), scientific trajectories (Robinson-Garcia et al., 2020), team dynamics (Xu et al., 2022), international collaboration (Wu et al., 2025), or gender differences (Macaluso et al., 2016; Sugimoto & Larivière, 2023), among others. However, much remains unknown since most analyses focus on specific disciplines and publications, mainly Biomedical Sciences and PLOS journals. In addition, from a bibliometric point of view, the possibilities of this still novel taxonomy are not fully explored. More than one decade since the birth of CRediT, we now have enough data to start getting a glimpse of its adoption, patterns, and prospects. Here, we present the first cross-disciplinary descriptive analysis of the use of contribution statements. The aim of this descriptive paper is to obtain further insight into contributions by a variety of fields, covering Health, Life, and Physical Sciences, as well as Multidisciplinary and Social Sciences, from the largest dataset used up to now, to the best of our knowledge. This broader disciplinary range is what differentiates our research from previous analyses, contributing to a more detailed understanding of CRediT data that allows for policy recommendations tailored by field. In order to offer a more nuanced image of the taxonomy, we analyze a set of more than 700,000 research articles belonging to Elsevier and PLOS journals from all fields of science, which we combine with bibliometric data extracted from the Scopus database. This combination of data has not been done before. We look into disciplinary differences in the use of contributions, their overall coverage, and their relationship with author position in papers. We conclude by discussing the potential of this data to open new venues of research on career trajectories, scientific impact and recognition, and the social organization of the sciences. This is a preprint. It is undergoing peer review. The final published version may differ. 3 Data & Methods In this study, we examine a total of 710,722 journal articles published between 2018 and 2023. The data was facilitated by the ICSR Lab from Elsevier as well as by PLOS. Elsevier was one of the first publishers adopting CRediT within their journal portfolio (Elsevier, 2024). Since then, they provide the option to include such information in their Editorial Manager, and they are working on offering it as well through the ScienceDirect journal platform (Genova, 2023). From the data provided by the ICSR Lab, we selected a total of 629,507 unique articles from 1,921 journals, authored by a list of 1,838,423 unique author profiles, ranging from 2020 to 2023. We conducted a manual quality check to ensure the reliability of the dataset, which is available in the Appendix (see Tables A1 and A2). Based on the results of this data validation, we removed years 2017-2019, and 2024 from the analysis. Most of the errors in those years involved the incorrect assignment of CRediT contributions to authors, the omission of some contributions or authors, and cases where the original paper used contribution statements that did not follow CRediT. PLOS data was facilitated by the publisher through a data use agreement. PLOS has stood out as one of the main drivers of the expansion of contribution statements by liberating their bibliographic data and allowing the scientometric community to explore it to better understand team dynamics. Examples of such efforts are the studies conducted by Larivière et al. (2016, 2021), Sauermann and Haeussler (2017), Macaluso et al. (2016), and Robinson-Garcia et al. (2020), to name just a few. In this case, they provided contribution data from their journal portfolio related to 97,819 publications for the 2018-2023 period, from which 83.2% belonged to PLOS One. Along with the author contribution statements assigned to every author, this dataset includes each article’s Digital Object Identifier (DOI). We enriched this set with author position data through their common DOIs and obtained a total of 92,280 unique PLOS articles, although matching each author to its corresponding CRediT statement and position was not straightforward, since shared author IDs were lacking. To overcome this limitation, we designed different name-surname identification strategies and iterated them in subsequent rounds to search for: 1) unique surnames per DOI, 2) unique name initials and surnames combinations per DOI, 3) unique full names and surnames combinations per DOI, and 4) coincidences per DOI through the separation of compound surnames. A total of 81,215 records were correctly matched, corresponding to 415,014 disambiguated authors (based on Scopus Author Profile). In our resulting PLOS subset, we only kept the records that had information on all the authors involved, where these identification strategies allowed us to determine their specific CRediT statements and positions per unique DOI. After a manual validation check (see Appendix, Table A3), the PLOS subset was combined with the already described ScienceDirect data. Our final dataset contains a total of 1,935 journals and 710,722 bibliographic records, authored by a total of 2,167,783 unique Scopus Author Profiles and published between 2018 and 2023 (Figure 1). The full Python code developed to merge, compute, and This is a preprint. It is undergoing peer review. The final published version may differ. 4 visualize the data is available (Di Césare and Xiao, 2024), as well as all the supplementary material (González-Salmón et al., 2025). Figure 1. Flowchart of the data merging process. To investigate whether author positions in the byline are associated with their participation in different CRediT categories, we employed the chi-square test of independence. In this analysis, the categorical variable “author position” (first, middle, last) was cross-tabulated with each CRediT category, where participation was recorded as a binary variable (involved vs. not involved). The tests were conducted not only for the overall dataset but also separately within each research field. Two levels of analysis were conducted. First, an overall test was performed using a contingency table that simultaneously included author positions and all CRediT categories, in order to assess whether the overall distribution of contribution types is independent of author position. The detailed results of this analysis are presented in the Appendix (Table A5). Second, to provide a more exhaustive examination, separate chi-square tests were conducted for each individual CRediT category, comparing the distributions of involvement (yes/no) across different author positions within each field. The detailed results of these tests are reported in the Appendix (Table A6). Next, we conduct a descriptive analysis of the dataset focusing on the overall coverage of the merged data, the distribution of authorship and disciplines at paper level, and the interactions between contribution statements, author order, and disciplines. This is a preprint. It is undergoing peer review. The final published version may differ. 5 Results General overview Figure 2 provides an overview of the magnitude of the dataset analyzed in comparison with the overall size of Scopus. The first two years of the period present low coverage, but the years 2020 to 2023 represent on average 6% of Scopus’ content, with 2022 reaching the highest coverage point at 7.2% (208,207 articles). Table S1 in the supplementary material (González-Salmón et al., 2025) shows this comparison by discipline and field. The list of fields, disciplines, and their corresponding acronyms is included in the Appendix (Table A4). Figure 2. Coverage of our merged dataset compared to Scopus by year. Both in absolute and relative terms, the most represented field in our dataset and compared to Scopus is Physical Sciences (501,549 articles, 6% of Scopus’ content). Within it, the disciplines Engineering (197,177), Materials Science (160,637), and Chemistry (129,390) stand out with the highest numbers of articles over the whole period. Multidisciplinary as a field is also prominent because it presents the highest percentages of shared articles between our dataset and Scopus throughout the years (19% on average) (Table S1, supplementary material, González-Salmón et al., 2025). In addition, Table S2 (supplementary material, González-Salmón et al., 2025) lists the 1,935 journals that form our merged dataset with the number of articles each one contains per year. The journal covering the largest number of publications from our dataset is PLOS One, with almost 10% of the total articles. This is to be expected since we are using PLOS data. This is a preprint. It is undergoing peer review. The final published version may differ. 6 Paper level analysis Table 1 shows descriptive measures of the number of authors per paper, overall and by major field. The mode for the whole set is 4 authors per article (n = 113,504). When looking at the number of authors across fields, we observe narrow differences. For Social Sciences, the most common number of authors is 3 (16,085 papers), while for both Multidisciplinary (9,765 papers) and Physical Sciences (85,361 papers), it is 4. In Health Sciences and Life Sciences most papers are written by 5 authors (9,411 and 23,107 articles, respectively). The maximum number of authors found in a single article changes significantly, with Physical Sciences showcasing the highest value (271 co-authors in a paper) and Social Sciences the lowest (75 co-authors in a paper). This is in line with other research that finds that Social Sciences are less collaborative than other fields (Parish et al., 2018) and research groups are less common in such disciplines (Kyvik & Reymert, 2017). Table 1. Descriptive measures of the number of authors per article, overall and by field. Measures Field Overall Health Sciences Life Sciences Multidisciplinary Physical Sciences Social Sciences Mean 5.6 6.9 6.3 6.4 5.3 4.2 Median 5.0 6.0 6.0 6.0 5.0 4.0 Mode 4.0 5.0 5.0 4.0 4.0 3.0 Minimum 1 1 1 1 1 1 Maximum 271 144 144 100 271 75 1º quartile 3.0 4.0 4.0 4.0 3.0 3.0 3º quartile 7.0 9.0 8.0 8.0 7.0 5.0 Standard deviation 3.2 4.2 3.5 3.9 2.8 2.4 Similarly, in Table 2 we examine the descriptive measures of the number of contribution types per article, overall and by field. All fields except for Multidisciplinary follow a similar pattern, in which the average number of contribution types per paper is around 9 contribution types. In Social Sciences, the most common number of contribution types per paper is 8, whereas for Physical Sciences it is 9, for Health Sciences and Life Sciences 10, and for Multidisciplinary it is 11 contribution types. Thus, Social Sciences is the field with the fewest contribution types, while Multidisciplinary presents almost 40% more contribution types. This may be related to Table 1, since it showed a lower number of authors per publication for the Social Sciences. This is a preprint. It is undergoing peer review. The final published version may differ. 7 Table 2. Descriptive measures of the number of CRediT contribution types per article, overall and by field. Measures Field Overall Health Sciences Life Sciences Multidisciplinary Physical Sciences Social Sciences Mean 9.2 9.2 9.4 10.3 9.0 8.8 Median 9.0 9.0 10.0 10.0 9.0 9.0 Mode 9.0 10.0 10.0 11.0 9.0 8.0 Minimum 1 1 1 1 1 1 Maximum 14 14 14 14 14 14 1º quartile 7.0 7.0 8.0 8.0 7.0 7.0 3º quartile 11.0 11.0 12.0 12.0 11.0 11.0 Standard deviation 2.8 2.9 2.8 2.7 2.8 2.8 The average proportion of contribution types by discipline can be seen in Figure 3. Some particular statements, like Conceptualization, Methodology, Writing – original draft, and Writing – review & editing, are clearly present in all disciplines at a high rate. Others, such as Resources, Project administration, and Funding acquisition, are conversely seldom present. In between, we find a few mixed scenarios where the same contribution, for instance Software, Data curation, and Formal analysis, has a high proportion of usage in some disciplines but very low in others. Here we also observe differences across major fields. We can observe a lower proportion of Formal analysis in Physical Sciences, and Supervision and Validation in Social Sciences, in comparison with other fields. Moreover, there is a slightly lower proportion of Investigation in Health and Social Sciences, and of Software in Health and Life Sciences. These results align with the understanding of the Social Sciences as a field in which all researchers can carry out most tasks, as there is a lower division of labor and, consequently, a lower need for Supervision and Validation. Figure 3 also reflects a higher proportion of Funding acquisition in Life Sciences, and a slightly higher use of Project administration and Resources in Health and Life Sciences, likely reflecting greater access to external funding than other fields, which leads to the need to manage budgets and teams (Tian et al., 2024). The proportion of Conceptualization, Data curation, Methodology, Visualization, Writing - original draft, and Writing - review & editing remains similar across fields. This is a preprint. It is undergoing peer review. The final published version may differ. 8 Figure 3. Average proportion of CRediT contribution types by discipline. This analysis at the paper level shows differences across fields, mainly between Social Sciences and Multidisciplinary, and the remaining fields. There could be many reasons for this, but these findings align with previous ones in the literature, such as smaller team sizes in the Social Sciences (Kyvik & Reymert, 2017) or the usage of qualitative methods in many Social Sciences disciplines that would lead to underuse of CRediT categories like Software or Visualization. Moreover, Social Sciences and Humanities researchers have historically focused on more localized issues and methods (Sivertsen, 2016), which could lead them to share a common overall understanding of the phenomenon under study and thus to contribute to the same tasks. Author level analysis Figure 4 shows the average number of contribution types each author contributes with, on average per field. In this way, we show how distributed tasks are by field. In Health, Life, and Physical Sciences, the most common scenario is that of 2 contributions per author (26.57%, 27.82% and 30.43% respectively), while in Multidisciplinary and Social Sciences, each author performs 3 tasks in 20.29% and 23.19% of the papers. If we look at higher numbers of contributions per author, we find that Health, Life, and Physical Sciences are less represented in those groups (in less than 6% of their articles, authors contribute with 7 or more different tasks), while Multidisciplinary and Social Sciences have a slightly more significant representation there (14.50% and 10.67%). Therefore, we observe a higher distribution of tasks in fields such as Health, Life, and Physical Sciences. On the opposite side, we find the Multidisciplinary and Social Sciences fields, where authors are involved in a higher number of tasks on average, showing a lower dependence on research groups (Parish et al., 2018). These results correspond to those obtained at the paper level regarding the specificity of the Social Sciences. This is a preprint. It is undergoing peer review. The final published version may differ. 9 Figure 4. Average number of CRediT contribution types by author involved per article and field. Co-occurrence of contributions by paper Here we investigate the level of co-occurrence between contribution types, that is, how common it is for a contribution to be conducted by the same author in the same publication (Figures 5 and 6). Overall, the most common contribution co-occurrences take place between Conceptualization and Methodology (39.7%), Writing - review & editing (35.0%), and Writing - original draft (33.8%). Furthermore, Writing - original draft also tends to co-occur with Methodology (34.4%) and Formal analysis (31.2%). The contributions that appear together the least are mostly related to Software and other statements (e.g., Funding acquisition, Supervision, and Project administration). The pairs formed by Funding acquisition and Data curation, Funding acquisition and Visualization, and Supervision and Data curation are also not usual contributions done by the same author within a publication (around 7% across all of them). There seems to be a higher co-occurrence of conceptual contributions whereby those who are involved in management-related contributions (e.g., Resources and Project administration) do not typically perform technical tasks (e.g., Data curation and Software). This could be related to academic age, given that junior authors tend to do more technical tasks while senior researchers are more involved in management (Larivière et al., 2016). When looking into specific fields, we observe that Methodology and Conceptualization is the most common combination in the Social (51.6%), Health (40.9%), and Physical Sciences (39.0%), while Writing - review & editing together with Conceptualization co-occurs the most in Multidisciplinary (46.0%) and Life Sciences (36.5%). In general, there is a higher co-occurrence of contributions in the Social Sciences and Multidisciplinary fields than in the rest. This finding is consistent with the results shown in Figure 4, which indicate that authors in these fields are more likely to contribute to multiple roles simultaneously and to work in smaller research groups. This is a preprint. It is undergoing peer review. The final published version may differ. 16 fairer alternative for evaluating scientific outputs, one that ensures that recognition is appropriately distributed among authors. Second, delving deeper into contributorship differences by field is key. The Social Sciences’ particular dynamics highlight the need to evaluate each field by applying ad hoc criteria that make varied uses of the contributions. In a research world where interdisciplinarity is increasingly common (Bolduc et al., 2023), the same approach could be taken for the Multidisciplinary field. The analysis of this dataset led us to consider many potential areas that could be investigated further. For instance, in combination with data on gender, country of affiliation, or academic age, among others, contribution statements could provide a fuller understanding of the intersection between labor division and inequalities within science. Since it is not possible to gain access to every laboratory in order to study working patterns, the CRediT taxonomy could also be seen as an accessible and broad alternative to ethnographic approaches. CRediT statements are still recent so, for the time being, it is not possible to conduct historical analyses. But thinking ahead, we find it very interesting to be able to witness the changes and trends that the use of this taxonomy is beginning to reveal. If the CRediT taxonomy were to be widely adopted across all disciplines and integrated into research evaluation frameworks, several challenges and opportunities would need to be addressed. A key challenge lies in the fact that not all fields apply the taxonomy in the same way. For instance, its use in the Social Sciences may require differentiated criteria compared with other domains. Moreover, not all categories of the taxonomy are equally valued across different research spheres. Treating all categories as equivalent risks flattening the diversity of contributions and obscuring the relative significance of different roles within the research process. On the opportunity side, the taxonomy has the potential to make visible and recognise diverse researcher profiles. Current evaluation systems tend to privilege the single archetype of the man-leader, while undervaluing other roles, such as supportive or collaborative contributions, which are often undertaken by women or researchers from less prestigious institutions or scientifically advanced countries. By enabling a more nuanced understanding of how science operates and of the value of different roles and their interplay, the CRediT taxonomy could broaden the recognition of outputs and career paths, fostering greater diversity and openness. Through such an approach, evaluative decisions could become fairer, helping to sustain a healthy, inclusive research ecosystem. Acknowledgments We would like to thank the ICSR Lab from Elsevier and PLOS for providing access to ScienceDirect, Scopus, and PLOS data on author contribution statements and bibliographic metadata. This is a preprint. It is undergoing peer review. The final published version may differ. 17 Funding information This paper is part of the COMPARE project (Ref: PID2020-117007RA-I00) funded by the Spanish Ministry of Science (Ref: MCIN/AEI/10.13039/501100011033 FSE invierte en tu futuro). Elvira González-Salmón is currently supported by an FPU grant from the Spanish Ministry of Science (Ref: FPU2021/02320). Victoria Di Césare is supported by a FPI grant from the Spanish Ministry of Science (Ref: PRE2021-097022). Aoxia Xiao is supported by a scholarship by from the China State Scholarship Fund. Nicolas Robinson-Garcia is supported by a Ramón y Cajal grant from the Spanish Ministry of Science (Ref: RYC2019-027886-I). Authors contributions Elvira González-Salmón: Data curation, Investigation, Formal analysis, Project administration, Writing – original draft. Victoria Di Césare: Data curation, Formal analysis, Methodology, Software, Visualization, Writing – review & editing. Aoxia Xiao: Data curation, Formal analysis, Methodology, Software, Visualization. Nicolas Robinson-Garcia: Conceptualization, Methodology, Project administration, Resources, Supervision, Writing – review & editing. References Allen, L., O’Connell, A., & Kiermer, V. (2019). How can we ensure visibility and diversity in research contributions? How the Contributor Role Taxonomy (CRediT) is helping the shift from authorship to contributorship. Learned Publishing, 32(1), 71–74. https://doi.org/10.1002/leap.1210 Birnholtz, J. P. (2006). What does it mean to be an author? The intersection of credit, contribution, and collaboration in science. Journal of the American Society for Information Science and Technology, 57(13), 1758–1770. https://doi.org/10.1002/asi.20380 Bolduc, S., Knox, J., & Ristroph, E. B. (2023). Evaluating team dynamics in interdisciplinary science teams. Higher Education Evaluation and Development, 17(2), 70–81. https://doi.org/10.1108/HEED-10-2021-0069 Brand, A., Allen, L., Altman, M., Hlava, M., & Scott, J. (2015). Beyond authorship: Attribution, contribution, collaboration, and credit. Learned Publishing, 28(2), 151–155. https://doi.org/10.1087/20150211 This is a preprint. It is undergoing peer review. The final published version may differ. 18 Costas, R., & Bordons, M. (2011). Do age and professional rank influence the order of authorship in scientific publications? Some evidence from a micro-level perspective. https://doi.org/10.1007/s11192-011-0368-z Cronin, B., Shaw, D., & La Barre, K. (2003). A cast of thousands: Coauthorship and subauthorship collaboration in the 20th century as manifested in the scholarly journal literature of psychology and philosophy. Journal of the American Society for Information Science and Technology, 54(9), 855–871. https://doi.org/10.1002/asi.10278 Di Césare, V., & Xiao, A. (2024). CRediT descriptive code [Python]. https://github.com/vdicesare/CRediT-Descriptive Ding, J., Liu, C., Zheng, Q., & Cai, W. (2021). A new method of co-author credit allocation based on contributor roles taxonomy: Proof of concept and evaluation using papers published in PLOS ONE. Scientometrics, 126(9), 7561–7581. https://doi.org/10.1007/s11192-021-04075-x Elsevier (2024). CRediT author statement. https://www.elsevier.com/researcher/author/policies-andguidelines/credit-author-statement Escabias, M., & Robinson-Garcia, N. (2022). The relation between contribution statements and academic age. 26th International Conference on Science, Technology and Innovation Indicators (STI 2022), Granada, Spain. https://doi.org/10.5281/ZENODO.6974732 Fanelli, D., & Glänzel, W. (2013). Bibliometric Evidence for a Hierarchy of the Sciences. PLoS ONE, 8(6), e66938. https://doi.org/10.1371/journal.pone.0066938 Genova, Z. (May 16, 2023). CRediT where credit's due. Elsevier. https://www.elsevier.com/connect/credit-where-credits-due Giménez Toledo, E. (2020). Why Books are Important in the Scholarly Communication System in Social Sciences and Humanities. Scholarly Assessment Reports, 2(1), 6. https://doi.org/10.29024/sar.14 González-Salmón, E., Di Césare, V., Xiao, A., & Robinson-Garcia, N. (2025). Supplementary material of the manuscript "Beyond authorship: Analyzing disciplinary patterns of contribution statements using the CRediT taxonomy" [Data set]. Zenodo. https://doi.org/10.5281/zenodo.17531864 Haeussler, C., & Sauermann, H. (2013). Credit where credit is due? The impact of project contributions and social factors on authorship and inventorship. Research Policy, 42(3), 688–703. https://doi.org/10.1016/j.respol.2012.09.009 Henriksen, D. (2016). The rise in co-authorship in the social sciences (1980–2013). Scientometrics, 107(2), 455–476. https://doi.org/10.1007/s11192-016-1849-x Huang, Y., Liu, X., Li, R., & Zhang, L. (2023). The science of team science (SciTS): An emerging and evolving field of interdisciplinary collaboration. El Profesional de La Información, e320204. https://doi.org/10.3145/epi.2023.mar.04 This is a preprint. It is undergoing peer review. The final published version may differ. 19 Jabbehdari, S., & Walsh, J. P. (2017). Authorship Norms and Project Structures in Science. Science, Technology, & Human Values, 42(5), 872–900. https://doi.org/10.1177/0162243917697192 Kulczycki, E., Engels, T. C. E., Pölönen, J., Bruun, K., Dušková, M., Guns, R., Nowotniak, R., Petr, M., Sivertsen, G., Istenič Starčič, A., & Zuccala, A. (2018). Publication patterns in the social sciences and humanities: Evidence from eight European countries. Scientometrics, 116(1), 463–486. https://doi.org/10.1007/s11192-018-2711-0 Kyvik, S., & Reymert, I. (2017). Research collaboration in groups and networks: Differences across academic fields. Scientometrics, 113(2), 951–967. https://doi.org/10.1007/s11192-017-2497-5 Larivière, V., Desrochers, N., Macaluso, B., Mongeon, P., Paul-Hus, A., & Sugimoto, C. R. (2016). Contributorship and division of labor in knowledge production. Social Studies of Science, 46(3), 417–435. https://doi.org/10.1177/0306312716650046 Larivière, V., Gingras, Y., Sugimoto, C. R., & Tsou, A. (2015). Team size matters: Collaboration and scientific impact since 1900. Journal of the Association for Information Science and Technology, 66(7), 1323–1332. https://doi.org/10.1002/asi.23266 Larivière, V., Pontille, D., & Sugimoto, C. R. (2021). Investigating the division of scientific labor using the Contributor Roles Taxonomy (CRediT). Quantitative Science Studies, 2(1), 111–128. https://doi.org/10.1162/qss_a_00097 Macaluso, B., Larivière, V., Sugimoto, T., & Sugimoto, C. R. (2016). Is Science Built on the Shoulders of Women? A Study of Gender Differences in Contributorship. Academic Medicine, 91(8), 1136–1142. https://doi.org/10.1097/ACM.0000000000001261 Matheson, A. (2022). Attribution and credit bias in publication ethics. Indian Journal of Medical Ethics, 1–8. https://doi.org/10.20529/IJME.2022.023 Nederhof, A. J. (2006). Bibliometric monitoring of research performance in the Social Sciences and the Humanities: A Review. Scientometrics, 66(1), 81–100. https://doi.org/10.1007/s11192-0060007-2 Parish, A. J., Boyack, K. W., & Ioannidis, J. P. A. (2018). Dynamics of co-authorship and productivity across different fields of scientific research. PLOS ONE, 13(1), e0189742. https://doi.org/10.1371/journal.pone.0189742 Rennie, D., Yank, V., & Emanuel, L. (1997). When authorship fails: a proposal to make contributors accountable. Journal of the American Medical Association, 278(7), 579-585. https://doi.org/10.1001/jama.278.7.579 Rennie, D., Flanagin, A., & Yank, V. (2000). The Contributions of Authors. Journal of the American Medical Association, 284(1), 89–91. Robinson-Garcia, N., Costas, R., Sugimoto, C. R., Larivière, V., & Nane, G. F. (2020). Task specialization across research careers. eLife, 9, e60586. https://doi.org/10.7554/eLife.60586 Sauermann, H., & Haeussler, C. (2017). Authorship and contribution disclosures. Science Advances, 3(11), e1700404. https://doi.org/10.1126/sciadv.1700404 This is a preprint. It is undergoing peer review. The final published version may differ. 20 Sivertsen, G. (2016). Patterns of internationalization and criteria for research assessment in the social sciences and humanities. Scientometrics, 107(2), 357–368. https://doi.org/10.1007/s11192016-1845-1 Sugimoto, C. R., & Larivière, V. (2023). Equity for Women in Science. Harvard University Press. Thelwall, M., & Maflahi, N. (2022). Research coauthorship 1900–2020: Continuous, universal, and ongoing expansion. Quantitative Science Studies, 3(2), 331–344. https://doi.org/10.1162/qss_a_00188 Tian, W., Cai, R., Fang, Z., Xie, Q., Hu, Z., & Wang, X. (2024). Research funding in different SCI disciplines: A comparison analysis based on Web of Science. Quantitative Science Studies, 5(3), 757–777. https://doi.org/10.1162/qss_a_00315 Wu, R., Esposito, C., & Evans, J. (2025). Shifting power asymmetries in scientific teams reveal China’s rising leadership in global science. Proceedings of the National Academy of Sciences, 122(44), e2414893122. https://doi.org/10.1073/pnas.2414893122 Wuchty, S., Jones, B. F., & Uzzi, B. (2007). The Increasing Dominance of Teams in Production of Knowledge. Science, 316(5827), 1036–1039. https://doi.org/10.1126/science.1136099 Xu, F., Wu, L., & Evans, J. (2022). Flat teams drive scientific innovation. Proceedings of the National Academy of Sciences, 119(23), e2200927119. https://doi.org/10.1073/pnas.2200927119 Appendix For the validation of the Scopus data, we manually checked 384 random articles. They constitute a representative sample with a 95% confidence and 5% margin of error. The sample was stratified by field and year. We checked whether the contributions that our data stated matched those in the original publication, and obtained the results shown in Table A1. Table A1. Validation check of a random set of articles from Elsevier journals. Field Year % of correct CRediT statements Nº of articles with correct statements Health Sciences 2017 0% 0/1 2018 0% 0/1 2019 0% 0/1 2020 87.5% 7/8 2021 75% 6/8 2022 87.5% 7/8 2023 87.5% 7/8 Life Sciences 2017 0% 0/1 2018 0% 0/1 2019 100% 1/1 2020 81.3% 13/16 2021 86.7% 13/15 This is a preprint. It is undergoing peer review. The final published version may differ. 21 2022 86.4% 19/22 2023 81.3% 13/16 2024 0% 0/1 Multidisciplinary 2019 0% 0/1 2021 100% 1/1 2022 100% 1/1 2023 100% 1/1 Physical Sciences 2018 0% 0/1 2019 100% 1/1 2020 86.4% 38/44 2021 84.5% 49/58 2022 88.33% 68/77 2023 89.1% 49/55 2024 0% 0/1 Social Sciences 2019 0% 7/1 2020 100% 7/7 2021 87.5% 7/8 2022 88.3% 9/11 2023 100% 7/7 2024 100% 1/1 Given that the years 2017, 2018, 2019, and 2024 were barely represented in our sample, we conducted an additional validation focusing on them. We aimed to check at least 50 random cases, distributed again by field. The year 2017 presented less than 50 cases, so we examined all of them. Table A2 displays the results of this extra manual check. Table A2. Additional validation of Elsevier journal articles. Field Year % of correct CRediT statements Nº of articles with correct statements Health Sciences 2017 20% 1/5 2018 0% 0/8 2019 94.4% 17/18 Life Sciences 2017 25% 1/4 2018 25% 4/16 2019 50% 6/12 2024 37.5% 9/24 Multidisciplinary 2018 45.5% 10/22 2019 80% 4/5 Physical Sciences 2018 100% 1/1 2019 100% 3/3 2024 80% 8/10 Social Sciences 2019 85.7% 6/7 2024 66.7% 8/12 The most common mistakes we found on the contributions, in order of occurrence, included the following (more than one could happen in the same paper): This is a preprint. It is undergoing peer review. The final published version may differ. 22 ● Contributions from authors are missing or wrongly assigned in our data. ● Authors are missing from our data. ● When the data were not normalized in Scopus (that is, they included contributions that are not part of the CRediT taxonomy, such as “Problem designing” or “Wrote the manuscript”), our data had trouble identifying the contribution they belonged to. ● Our data has extra authors who do not appear in Scopus. ● Contributions were sometimes duplicated in our data (for instance, an author having “Data curation” twice as their contributions). For the validation of the PLOS data, we checked 50 random papers for each year (2018-2023) to analyze its robustness. As we can see in Table A3, PLOS data had higher percentages of accuracy than that from Scopus. Table A3. Validation check of a random set of articles from PLOS journals. Year % of correct CRediT statements Nº of articles with correct statements 2018 100% 50/50 2019 94% 47/50 2020 98% 49/50 2021 100% 50/50 2022 98% 49/50 2023 100% 50/50 Table A4. Grouping of disciplines and fields. Field Discipline Abbreviation Health Sciences Dentistry DENT Health Professions HEAL Medicine MEDI Nursing NURS Veterinary VETE Life Sciences Agricultural and Biological Sciences AGRI Biochemistry, Genetics and Molecular Biology BIOC Immunology and Microbiology IMMU Neuroscience NEUR Pharmacology, Toxicology and Pharmaceutics PHAR Physical Sciences Chemical Engineering CENG Chemistry CHEM Computer Sciences COMP Earth and Planetary Sciences EART Energy ENER Engineering ENGI Environmental Science ENVI Materials Science MATE Mathematics MATH Physics and Astronomy PHYS This is a preprint. It is undergoing peer review. The final published version may differ. 23 Social Sciences Arts and Humanities ARTS Business Management and Accounting BUSI Decision Sciences DECI Economics, Econometrics and Finance ECON Psychology PSYC Social Sciences SOCI Multidisciplinary MULT The chi-squared statistic measures the magnitude of deviation between observed and expected frequencies. A larger value of the chi-square statistic indicates a greater deviation. Table A5. Results of chi-square tests of independence between author position and CRediT statements. Field chi-square statistic (χ²) p_value df Overall 1618538.998 0.000 26 Health Sciences 176008.858 0.000 26 Life Sciences 461125.574 0.000 26 Multidisciplinary 125086.840 0.000 26 Physical Sciences 1222241.785 0.000 26 Social Sciences 83714.325 0.000 26 Note. All chi-square tests were conducted with 26 degrees of freedom. All p-values are 0.000, indicating statistically significant associations between author position and overall CRediT category participation in each field. Table A6. Results of chi-square tests (χ²) examining the relationship between author position and each CRediT statement across fields. CRediT statements Overall Health Sciences Life Sciences Multidisciplinary Physical Sciences Social Sciences Conceptualization 543506.5 84718.8 174857.4 69065.6 349430.1 42733.8 Data curation 226756.9 20609.1 60200.0 32352.8 155523.3 16230.6 Formal analysis 354040.5 52550.8 105365.8 59391.1 208227.6 36379.6 Funding acquisition 324168.3 48507.6 138608.6 46620.3 186274.7 8798.9 Investigation 316457.3 22527.6 80898.6 28597.5 240162.2 18642.3 Methodology 437605.5 41574.8 90134.2 32035.5 356593.5 43804.0 Project administration 228209.1 34926.5 99542.6 37912.1 139668.8 6167.3 Resources 57995.7 4437.3 19498.3 7511.3 43621.0 1812.9 Software 188771.1 13488.3 31575.3 16418.3 151821.1 21787.7 Supervision 546400.3 79377.1 188436.4 61063.4 349898.3 20577.2 Validation 47014.8 5197.1 13980.1 14757.9 28410.9 2312.7 Visualization 222653.5 32707.9 69417.9 40647.4 128752.3 19119.6 Writing – original draft 1449275.5 158225.1 363522.4 140773.4 1049999.8 90007.3 Writing – review & editing 137873.3 14407.7 49750.6 12529.4 101214.0 5772.9 Note. All chi-square tests were conducted with 2 degrees of freedom. All p-values are 0.000, indicating statistically significant associations between author position and CRediT statements in each field.