ABCS Training Events

Effect Size, p-value, and Accuracy

Presented By

Brian Luke

Watch Recording View Slides

Event Details

Presenter: Brian Luke, PhD, Advanced Biomedical Computational Science 

This presentation will explain the difference between the mean and standard deviation of a set of values and the standard error of the mean. The parameters involved in comparing two normally distributed populations (Cases and Controls) relative to a single value (biomarker) are the sample size, the effect size, the standard deviations of the distributions, the significance level (alpha), and the power (1 – beta). You need four of the five parameters to estimate the fifth, and in many cases the researcher is interested in the sample size. Increasing the sample size will, in general, not change the effect size (difference in means of the populations), or the standard deviations of the populations, but will increase the significance (decrease the p-value) of the effect size. This means that increasing the population size will not increase the accuracy of the classification relative to the biomarker, i.e., the ability to distinguish Cases from Controls. 

This will be a hybrid event. To register, please visit the following link: 

This session will be recorded, and materials will be shared with attendees a few days after the event. 

For additional details and questions, please contact Natasha Pacheco (, Advanced Biomedical Computational Science group, Frederick National Laboratory for Cancer Research.

Event Details

Tue, May 14

12:00 PM - 1:00 PM

Building 549, Conference Room B, NCI-Frederick