Malcolm Macleod

@maclomaclee.bsky.social

Meta-researcher, Neurologist, Runner ... aspiring to get better at all these and more!

It's now 4½ years since I first thought of writing it, and 3½ years since I actually started writing it, but today I finally got around to submitting the preprint. Say hello to SPIV. 🧵

SPIV analysis: A heuristic meta-method for detecting non-random patterns in study data
Nicholas J. L. Brown, PhD
Kyle Sheldrick, MD, PhD
Ben W. Mol, MD, PhD
Lyle C. Gurrin, PhD

Please address correspondence to nicholasjlbrown@gmail.com.

Abstract
The increasing prevalence of open data is changing the ways in which forensic meta-science is done. Previously it was possible to infer what had or had not taken place only by examining its published summary statistics. Today meta-researchers wishing to reanalyse published results can also examine author-supplied individual participant data (IPD) to identify irregularities, either because they are directly published or because an editor has asked the authors to supply them. Here, we discuss some of the ways in which one might go about trying to detect anomalies in typical datasets in the biomedical or social sciences by examining the relationship between consecutive participant records within a dataset. We refer to these as “SPIV analyses”, where SPIV stands for Sequences in Purportedly Independent Values. SPIV is based on the principle that the data records for consecutive participants in a randomized study — and, potentially, some study designs that do not use randomisation — should be independent of each other for the great majority of measured variables. Using SPIV techniques, we can identify data that do not satisfy this condition, perhaps as a result of having been falsified or generated manually by unscrupulous researchers.

Who do you review for? Should you boycott commercial publishers? Should you boycott specific publishers? Does it matter for your career? We are orgainizing a one-hour session with lightning talks and an open discussion. Join us on May 26th 12-13 CEST online. Registration at t1p.de/wdirf

Bild

Join us at #WCRI2026 for a networking session for anyone working on #PaperMills and #GenAI on the context of #ResearchIntegrity. There'll be lightning talks, networking and probs drinks/ dinner afterwards. May 4, 6.00-7.30 pm, all welcome!

a man wearing glasses and a black shirt that says may 4 on it

ALT: a man wearing glasses and a black shirt that says may 4 on it

media.tenor.com

World Conferences on Research Integrity@wcrifoundation.bsky.social · 5mo ago

More ways to connect at #WCRI2026 in Vancouver: 🔹 Pre-conference workshops (May 3) 🔹 NERQ Reception (May 4, 9–11 PM) 🔹 Paper Mills & Generative AI rapid insights session (May 4) Learn more and join us in May. wcri2026.org/pre-conferen...

There is so much to be disappointed about in this story--from the institutional response to Miles Hewstone himself. We still have a lot of work to do to make academic environments inclusive spaces where everyone can thrive, and where harassers are held accountable.

Nils Reimer@reimtime.bsky.social · 9mo ago

This—on my former university, department, and advisor—is harrowing but required reading for all social psychologists. www.bloomberg.com/news/feature...