Biology subjects

Ioannidis, J. P. A.

Publications and source records attributed to Ioannidis, J. P. A..

5 recordsLinked to original sources

bioRxiv2026

Uptake and Implementation of Multiverse-style Analyses Across 613 Studies

Empirical conclusions can depend on the many individual choices researchers make when analyzing data. Multiverse-style analyses address this by computing results across a set of specifications rather than a single one, but it is unclear how widely they are being used and how they are being implemented. We searched the Web of Science Core Collection (May 2026) for articles citing six foundational papers on different variants of multiverse-style methods and classified each of the citing articles as implementing such methods or only discussing them. For implementations, we recorded the framework used, the number of specifications, which of four decision nodes (measurement, data processing, modeling, and estimation) were varied, and other aspects such as how results were visualized and interpreted. Of the 1545 classifiable articles, 613 (39.7%) implemented a multiverse-style analysis (primarily multiverse n = 336, specification curve n = 175, vibration of effects n = 20, multimodel n = 59, and multi-analyst/many-analyst n = 23). Uptake spanned many disciplines, most often psychology (39.8%), the social sciences (20.7%), and medicine (19.2%). The number of specifications ranged from fewer than ten to more than ten thousand (median = 144, IQR 24-1248). Modeling (75%) and data-processing (62%) choices were included most often. Interpretation was predominantly descriptive, whereas formal inference (10.5%), preregistration (8.6%), and explicit attention to the defensibility of specifications (3.9%) were uncommon. Multiverse-style analyses are increasingly becoming established across the quantitative sciences, but some implementation practices can be strengthened so that these analyses become more fully transparent and more genuinely informative about the robustness of research findings.

scientific communication and education↗

bioRxiv2025

Gender imbalances of retraction prevalence among highly cited authors and among all authors

BackgroundScientific retractions remain rare but have become increasingly common. We have previously incorporated retraction data into Scopus-based databases of top-cited (top 2%) scientists to facilitate linkage of retractions with impact metrics at the individual scientist level. Here, we set out to explore whether gender disparities in the likelihood of having retractions exist, both among highly-cited authors and among all authors with [≥]5 publications. MethodsWe conducted a descriptive cross-sectional bibliometric analysis of a Scopus-based authors database. We used NamSor to assign gender, retaining only results with a confidence >85%. We examined the demographics of scientists with and without retractions among highly cited authors (career-long impact: n = 217,097) and among all other authors (n = 10,361,367). We stratified by publication age, field, country income level, and publication volume, and calculated gender-specific retraction rates and the prevalence ratio (R) of women versus men to have at least one retraction. Multivariable logistic regression were fitted to estimate adjusted associations, reported as odds ratios (OR) and 95% confidence intervals (CI). ResultsGender could be classified for 8,267,888 scientists. Among highly cited authors, 3.3% of men and 2.9% of women had at least one retraction; among all authors, the rate was 0.7% for both genders. Differences varied by field: womens rates were at least one-third lower than mens (R < 0.7) in Biology, Biomedical Research, and Psychology (RD D1.3) in Economics, Engineering, and Information and Communication Technologies. Among highly cited authors, younger cohorts showed increasingly higher rates among men (4.2% men vs. 3.0% women in those starting to publish in 2002-2011; 8.7% men vs. 4.9% women in those starting post-2011). Country-level differences among highly cited authors were pronounced in some countries, as in Pakistan (28.7% men vs. 14.3% women). In multivariable analyses, differences were minimal, if any, for women versus men in having [≥]1 retraction (among all authors OR:D0.96, 95% CI: 0.94-0.97) among highly cited authors OR:D0.96, 95% CI: 0.89-1.03), while there were very strong associations of retractions with career age, country income, publication volume, and specific disciplines. ConclusionOur analysis shows no major gender differences in retraction rates overall, but field-specific differences may exist. Field, career age, country, and publication volume are stronger correlates of retraction. Structural and contextual factors likely drive retraction patterns and warrant further investigation.

scientific communication and education↗

bioRxiv2023

Twitter and Mastodon presence of highly-cited scientists

Social media platforms have an increasing influence in biomedical and other disciplines of science and public health. While Twitter has been a popular platform for scientific communication, changes in ownership have led some users to consider migrating to other platforms such as Mastodon. We aimed to investigate how many top-cited scientists are active on these social media platforms, the magnitude of the migration to Mastodon, and correlates of Twitter presence. A random sample of 900 authors was examined among those who are at the top-2% of impact based on a previously validated composite citation indicator using Scopus data. Searches for their personal Twitter accounts were performed in early December 2022, and re-evaluations were performed at 2 weeks, 4 weeks, and 2 months (February 6, 2023). 262/900 (29.1%) of highly-cited scholars had Twitter accounts, and only 9/800 (1%) had Mastodon accounts. Female gender, North American and Australia locations, younger publication age, and clinical medicine or social science expertise correlated with higher percentages of Twitter use. The vast majority of highly-cited author users of Twitter had few followers and tweets. Only 6 had more than 10,000 followers and none had more than 100,000. One limitation of our study is that it is possible that some accounts, especially with Mastodon, could not be detected. However, the study suggests that Twitter remains the preferred social media platform for highly-cited authors, and Mastodon has not yet challenged Twitters dominance. Moreover, most highly-cited scientists with Twitter accounts have limited presence in this medium.

scientific communication and education↗

bioRxiv2020

Assessment of transparency indicators across the biomedical literature: how open is open?

Recent concerns about the reproducibility of science have led to several calls for more open and transparent research practices and for the monitoring of potential improvements over time. However, with tens of thousands of new biomedical articles published per week, manually mapping and monitoring changes in transparency is unrealistic. We present an open-source, automated approach to identify five indicators of transparency (data sharing, code sharing, conflicts of interest disclosures, funding disclosures and protocol registration) and apply it across the entire open access biomedical literature of 2.75 million articles on PubMed Central. Our results indicate remarkable improvements in some (e.g. conflict of interest disclosures, funding disclosures), but not other (e.g. protocol registration, code sharing) areas of transparency over time, and map transparency across fields of science, countries, journals and publishers. This work has enabled the creation of a large, integrated, and openly available database to expedite further efforts to monitor, understand and promote transparency and reproducibility in science.

scientific communication and education↗

bioRxiv2017

Modeling Science Trustworthiness Under Publish Or Perish Pressure

The scientific endeavor pivots on the accurate reporting of experimental and theoretical findings, and consequently scientific publication is immensely important. As the number of active scientists continues to increase, there is concern that rewarding scientists chiefly on publication creates a perverse incentive where careless and fraudulent research can thrive. This is compounded by the predisposition of top-tier journals towards novel or positive findings rather than negative results or investigations that merely confirm a null hypothesis, despite their intrinsic value, potentially compounding a reproducibility crisis in several fields. This is a serious problem for both science and public trust in scientific findings. To date, there has been comparatively little mathematical modeling on the factors that influence science trustworthiness, despite the importance of quantifying the problem. In this work, we present a simple phenomenological model with cohorts of diligent, careless and unethical scientists with funding allocated based on published outputs. The results of this analysis suggest that trustworthiness of published science in a given field is strongly influenced by the false positive rate and the pressures from journals for positive results, and that decreasing available funding has negative consequences for the resulting trustworthiness. We also examine strategies to combat propagation of irreproducible science, including increasing fraud detection and awarding diligence, discussing the implications of these findings.

scientific communication and education↗