Same data, different analysts: variation in effect sizes due to analytical decisions in ecology and evolutionary biology

BMC Biology, Journal Year: 2025, Volume and Issue: 23(1)

Published: Feb. 6, 2025

Although variation in effect sizes and predicted values among studies of similar phenomena is inevitable, such far exceeds what might be produced by sampling error alone. One possible explanation for results differences researchers the decisions they make regarding statistical analyses. A growing array has explored this analytical variability different fields found substantial despite analysts having same data research question. Many these have been social sciences, but one small "many analyst" study ecology. We expanded scope prior work implementing a large-scale empirical exploration model predictions generated ecology evolutionary biology. used two unpublished datasets, from (blue tit, Cyanistes caeruleus, to compare sibling number nestling growth) conservation (Eucalyptus, grass cover tree seedling recruitment). The project leaders recruited 174 analyst teams, comprising 246 analysts, investigate answers prespecified questions. Analyses conducted teams yielded 141 usable effects (compatible with our meta-analyses all necessary information provided) blue tit dataset, 85 Eucalyptus dataset. heterogeneity both although patterns differed between them. For analyses, average was convincingly negative, less growth nestlings living more siblings, there near continuous size large negative zero, even crossing traditional threshold significance opposite direction. In contrast, relationship only slightly not most ranged weakly positive, about third direction or other. However, were also several striking outliers zero. we variable selection random structures as well ratings methods peer reviewers, no strong any deviation meta-analytic mean. other words, analyses that mean likely dissimilar sets, use their models, receive poor reviews than those close existence analysis outcomes raises important questions how ecologists biologists should interpret published results, conduct future.

Language: Английский

---

Introducing the Journal of Robustness Reports

Published: April 16, 2025

The vast majority of empirical research articles report a single primary analysis outcome that is the result plan, executed by team (usually also designed experiment and collected data). However, recent many-analyst projects have demonstrated different teams generally adopt unique approach there exists considerable variability in associated conclusions. There appears to be no optimal statistical plausible plans need not lead same conclusion. A high outcomes signals conclusions are relatively fragile dependent on specifics plan. Crucially, without multiple analyzing data, it difficult gauge extent which robust. We recently proposed particular scientific interest or societal importance accompanied two three short reports summarize results alternative analyses conducted independent experts [F. Bartoš et al., Nat. Hum. Behav. (2025)]. In order showcase practical feasibility epistemic benefits this we founded Journal Robustness Reports, dedicated publishing reanalyses findings. This editorial describes scope workflow Reports including type format published articles. hope will help make findings norm across sciences.

Language: Английский

---

A necessary diversity of perspectives in decision-making regarding deep seabed mining: implications for science, people, and the environment

Frontiers in Sustainable Resource Management, Journal Year: 2025, Volume and Issue: 4

Published: April 23, 2025

The transition from the current fossil fuel-based economy toward one that relies on renewable sources of energy allegedly will require a set minerals for manufacturing batteries store this and power electric devices. Deep seabed mining (DSM) is an economic activity has potential to fill these material requirements as it collecting rich mineral resources bottom ocean. This brings enormous challenges regulation potentially irreversible impacts large scale. In addition, considered common heritage humankind, therefore, questions distributions burdens profits also emerge. We build premise social justice, legitimacy, participatory processes discuss six perspectives should be while dealing with DSM. claim DSM seen through wicked problem lens, acknowledging limits ignorance squared, inside scientific paradigm open possibility post-normal science. Participation center recognizing plural rationalities, ensuring justice capabilities, actively including global South. conclude DSM's legitimacy can enhanced by following perspective guidelines.

Language: Английский

---

Replicating a COVID-19 study in a national England database to assess the generalisability of research with regional electronic health record data

BMJ Open, Journal Year: 2025, Volume and Issue: 15(4), P. e093080 - e093080

Published: April 1, 2025

Objectives To assess the degree to which we can replicate a study between regional and national database of electronic health record data in UK. The original examined risk factors associated with hospitalisation following COVID-19 infection people diabetes. Design A replication retrospective cohort study. Setting Observational from primary secondary care sources used large, urbanised region (Greater Manchester Care Record, Greater Manchester, UK—2.8 m patients). This covering whole England, UK (NHS England’s Secure Data Environment service for accessed via BHF Science Centre’s CVD-COVID-UK/COVID-IMPACT Consortium—54 Participants Individuals diagnosis type 1 diabetes or 2 prior positive test result. matched controls (3:1) were individuals who had result, but did not have on date their Matching was done age at diagnosis, sex approximate test. Primary outcome measures Hospitalisation within 28 days Results We found that many effect sizes show statistically significant difference, some did. Where study, then they remained size same direction similar magnitude. Conclusions There is evidence findings studies smaller datasets be extrapolated larger, setting. However, there differences, therefore remain an essential part healthcare research.

Language: Английский

---

Provide proactive reproducible analysis transparency with every publication

Royal Society Open Science, Journal Year: 2025, Volume and Issue: 12(3)

Published: March 1, 2025

The high incidence of irreproducible research has led to urgent appeals for transparency and equitable practices in open science. For the scientific disciplines that rely on computationally intensive analyses large datasets, a granular understanding analysis methodology is an essential component reproducibility. This article discusses guiding principles computational reproducibility framework enables scientist proactively generate complete reproducible trace as unfolds, share data, methods executable tools part publication, allowing other researchers verify results easily re-execute steps investigation.

Language: Английский

---

No evidence for olfactory kin discrimination in begging blue tit nestlings

Animal Behaviour, Journal Year: 2025, Volume and Issue: unknown, P. 123131 - 123131

Published: March 1, 2025

Language: Английский

---

Reply to Pitogo et al.: No single silver bullet to simply understand war-biodiversity conflict in the Philippines

npj Biodiversity, Journal Year: 2025, Volume and Issue: 4(1)

Published: May 5, 2025

Sociopolitical conflicts have significant but often overlooked impacts on biodiversity. In our reply, we reaffirm key findings from previous work and directly address the Matters Arising raised by Pitogo colleagues. Additionally, present fine-scale analyses that further support original conclusions. We emphasise need for continued research to fully unravel complex relationship between conflict environmental impacts.

Language: Английский

---