Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

Sample Size Calculation01:19

Sample Size Calculation

6.7K
Knowledge of the sample size is the first requirement to conduct random sampling or an experiment. The sample size is the total number of units, observations, or groups (in some cases) used to get the data to estimate a population parameter. As the name suggests, the sample size is that of the sample drawn from the population and differs from the population size.
The sample size for the given experiment or sampling effort is fundamental to any study design. Sample size decides the number of...
6.7K
One-Way ANOVA: Unequal Sample Sizes01:15

One-Way ANOVA: Unequal Sample Sizes

6.8K
One-way ANOVA can be performed on three or more samples of unequal sizes. However, calculations get complicated when sample sizes are not always the same. So, while performing ANOVA with unequal samples size, the following equation is used:
6.8K
One-Way ANOVA: Equal Sample Sizes01:15

One-Way ANOVA: Equal Sample Sizes

4.1K
One-Way ANOVA can be performed on three or more samples with equal or unequal sample sizes. When one-way ANOVA is performed on two datasets with samples of equal sizes, it can be easily observed that the computed F statistic is highly sensitive to the sample mean.
Different sample means can result in different values for the variance estimate: variance between samples. This is because the variance between samples is calculated as the product of the sample size and the variance between the...
4.1K
Criteria for Causality: Bradford Hill Criteria - II01:28

Criteria for Causality: Bradford Hill Criteria - II

1.3K
The Bradford Hill criteria serve as guidelines for establishing causative links in epidemiological research. Beyond Strength, Consistency, Specificity, and Temporality, key criteria also include Biological Gradient, Plausibility, Coherence, Experiment, and Analogy. These principles assist scientists in assessing the likelihood of causation in complex biological contexts. Below is a summary of these concepts:
1.3K
Criteria for Causality: Bradford Hill Criteria - I01:30

Criteria for Causality: Bradford Hill Criteria - I

1.1K
The Bradford Hill criteria are a group of principles that provide a framework to determine a causal relationship between a specific factor and a disease. There are nine criteria that are pivotal in assessing causality in epidemiological studies. Here's a closer look at Strength, Consistency, Specificity, and Temporality criteria with definitions and examples:
1.1K
Binary Fission01:26

Binary Fission

3.2K
Binary fission is the primary mode of asexual reproduction in prokaryotes, such as bacteria. It results in the production of two genetically identical daughter cells. This highly efficient process ensures the rapid propagation of bacterial populations under favorable conditions and involves coordinated cellular and molecular events.DNA Replication and SeparationThe process begins with the replication of the bacterial chromosome. The circular DNA molecule unwinds at a specific origin of...
3.2K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Author Correction: The STARD-AI reporting guideline for diagnostic accuracy studies using artificial intelligence.

Nature medicine·2026
Same author

Prognostic factor research: why it matters in orthopaedics and how we do it better.

Bone & joint open·2026
Same author

Cost-effectiveness of osteoporotic fracture risk assessment in people with intellectual disabilities: a UK NHS modelling study.

BMJ open·2026
Same author

Signal or noise? Evaluating commonly used attribution methods for explaining deep neural networks in electrocardiogram classification.

European heart journal. Digital health·2026
Same author

Effectiveness of drug interventions to prevent delirium after surgery for older adults: systematic review and network meta-analysis of randomised controlled trials.

BMJ (Clinical research ed.)·2026
Same author

Reporting Completeness of Systematic Reviews of Nutrition or Diet-Related Randomised Controlled Trials: A Meta-Research on Adherence to PRISMA Items.

Nutrition reviews·2026

Related Experiment Video

Updated: Feb 8, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.8K

Sample size for binary logistic prediction models: Beyond events per variable criteria.

Maarten van Smeden1, Karel Gm Moons1, Joris Ah de Groot1

  • 11 Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht, The Netherlands.

Statistical Methods in Medical Research
|July 4, 2018
PubMed
Summary

The Events Per Variable (EPV) criterion is not a reliable measure for developing clinical prediction models. Sample size, number of predictors, and events fraction are better indicators of model performance.

Keywords:
EPVLogistic regressionprediction modelspredictive performancesample sizesimulations

More Related Videos

Modeling the Size Spectrum for Macroinvertebrates and Fishes in Stream Ecosystems
07:41

Modeling the Size Spectrum for Macroinvertebrates and Fishes in Stream Ecosystems

Published on: July 30, 2019

8.0K
Porcine Liver Transplantation Without Veno-Venous Bypass As an Extended Criteria Donor Model
12:49

Porcine Liver Transplantation Without Veno-Venous Bypass As an Extended Criteria Donor Model

Published on: August 17, 2022

3.1K

Related Experiment Videos

Last Updated: Feb 8, 2026

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma
04:09

Predicting Treatment Response to Image-Guided Therapies Using Machine Learning: An Example for Trans-Arterial Treatment of Hepatocellular Carcinoma

Published on: October 10, 2018

8.8K
Modeling the Size Spectrum for Macroinvertebrates and Fishes in Stream Ecosystems
07:41

Modeling the Size Spectrum for Macroinvertebrates and Fishes in Stream Ecosystems

Published on: July 30, 2019

8.0K
Porcine Liver Transplantation Without Veno-Venous Bypass As an Extended Criteria Donor Model
12:49

Porcine Liver Transplantation Without Veno-Venous Bypass As an Extended Criteria Donor Model

Published on: August 17, 2022

3.1K

Area of Science:

  • Statistics
  • Biostatistics
  • Clinical Epidemiology

Background:

  • Binary logistic regression is widely used for clinical prediction models.
  • The Events Per Variable (EPV) criterion, particularly EPV ≥10, is commonly used to determine sample size and predictor numbers.
  • Existing criteria may not adequately reflect the complexities of prediction model development.

Purpose of the Study:

  • To investigate the influence of EPV and other factors on the out-of-sample predictive performance of binary logistic regression models.
  • To evaluate the appropriateness of the EPV criterion in prediction model development.
  • To propose improved methods for sample size determination in prediction modeling.

Main Methods:

  • An extensive simulation study was conducted.
  • The study examined the impact of EPV, events fraction, number of predictors, variable characteristics, and predictor effects.
  • Model performance (calibration, discrimination, probability error) was assessed before and after regression shrinkage and variable selection.

Main Results:

  • EPV showed a weak relationship with predictive performance metrics.
  • EPV is not a suitable criterion for binary prediction model development.
  • Out-of-sample predictive performance is better approximated by the number of predictors, total sample size, and events fraction.

Conclusions:

  • The EPV criterion should not be relied upon for sample size determination in prediction model development.
  • Future sample size criteria should incorporate the number of predictors, total sample size, and events fraction.
  • These findings offer guidance for improving sample size determination in clinical prediction modeling.