Skip to content
DraconiticResearch
Search Sources API
203,203 studies31 sourcesUpdated 35 min ago
Peer-reviewedCC BYPLOS Medicine

Addressing threats to validity in large dataset research

  • Wonsuk Yoo1
  • John P A Ioannidis2,3,4,5
5 affiliations
  1. 1School of Nursing and Health Studies, Division of Biostatistics/Bioinformatics, Department of Public Health Sciences, TCO Program, Sylvester Comprehensive Cancer Center, University of Miami, Coral Gables, Florida, United States of America.
  2. 2Meta-Research Innovation Center at Stanford (METRICS), Stanford University, California, United States of America.
  3. 3Stanford Prevention Research Center, Department of Medicine, Stanford University School of Medicine, Stanford, California, United States of America.
  4. 4Department of Epidemiology and Population Health, Stanford University School of Medicine, Stanford, California, United States of America.
  5. 5SDepartment of Biomedical Data Science, Stanford University School of Medicine, Stanford, California, United States of America.
Read full textPDFDOICite

Abstract

In this Guidelines and Guidance article, Wonsuk Yoo and John P.A. Ioannidis outline why research using large datasets faces threats to validity, highlighting four main statistical pitfalls and providing guidance and worked examples on how to mitigate against them.

Related

  1. What Pilot Studies Can (and Cannot) Do for Validity in Psychological Research
    Yashvin Seetahul, Mahmoud Elsherif, Caroline Zygar-Hoffmann +7
    PsyArXivPreprintCC BY 4.0Full textStatisticsOSF
  2. Measurement reliability, construct validity, and transparent reporting in original and replication psychological research
    Cas Goos, Marjan Bakker, Jelte M. Wicherts +1
    PLOS OnePeer-reviewedCC BY 4.0Full textStatisticsPLOS
  3. Reproducible benchmarking of four machine-learning classifiers on a large synthetic heart-disease competition dataset
    Raed M. Ennab
    PLOS OnePeer-reviewedCC BY 4.0Full textStatisticsCardiovascular medicinePLOS
  4. Research on graph theory segmentation method and dataset construction for connected handwritten digits
    Zhilai Zhang, Yang Zhang, Yang Wang
    PLOS OnePeer-reviewedCC BYFull textStatisticsPLOSEurope PMCPMC
  5. Efficient Gaussian Process Modeling for Large Replicated and Censored Data in Industrial Research
    Ying Zhang, Emily L. Kang, Shuo Li
    arXivPreprintarXiv licenceFull textStatisticsarXiv

Cite

BibTeXRIS
BibTeX
@article{yoo2026addressing,
  title = {{Addressing threats to validity in large dataset research}},
  author = {Wonsuk Yoo and John P A Ioannidis},
  journal = {PLOS Medicine},
  year = {2026},
  month = nov,
  doi = {10.1371/journal.pmed.1005282},
  url = {https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005282},
  note = {CC BY},
}
RIS
TY  - JOUR
TI  - Addressing threats to validity in large dataset research
AU  - Wonsuk Yoo
AU  - John P A Ioannidis
PY  - 2026
DA  - 2026/11/01
JO  - PLOS Medicine
DO  - 10.1371/journal.pmed.1005282
UR  - https://journals.plos.org/plosmedicine/article?id=10.1371/journal.pmed.1005282
AB  - In this Guidelines and Guidance article, Wonsuk Yoo and John P.A. Ioannidis outline why research using large datasets faces threats to validity, highlighting four main statistical pitfalls and providing guidance and worked examples on how to mitigate against them.
SN  - 1549-1676
SN  - 1549-1277
ER  -