StATS: What is a beta level?

The beta level (often simply called beta) is the probability of making a Type II error (accepting the null hypothesis when the null hypothesis is false). It is directly related to power, the probability of rejecting the null hypothesis when the null hypothesis is false. Power plus beta always equals 1.0.

Frequently researchers will select a sample size and decision rule to insure that beta is 0.20 or less (or equivalently power is 0.80 or more). Some researchers prefer to insure that the beta level is 0.10 or less.

Example: In a study of two interventions to prevent hypothermia during liver transplantation, the researchers designed the study so that the beta level was 0.2.

A minimum sample size of 24 total subjects was calculated as needed to be enrolled and analyzed to detect a clinically relevant difference in the primary outcome of mean rectal or esophageal temperatures (temperature difference 0.5°C between the groups) according to the power analysis based on following parameters: type II error rate- (beta = 0.2), type I error rate (alpha = 0.05, d = 0.5, sigma = 0.4). -- Water warming garment versus forced air warming system in prevention of intraoperative hypothermia during liver transplantation: a randomized controlled trial [ISRCTN32154832]. Janicki PK, Stoica C, Chapman WC, Wright JK, Walker G, Pai R, Walia A, Pretorius M, Pinson CW. BMC Anesthesiol 2002: 2(1); 7. [Medline] [Abstract] [Full text] [PDF]

This work is licensed under a Creative Commons Attribution 3.0 United States License. It was written by Steve Simon on 2007-04-05, edited by Steve Simon, and was last modified on 2010-04-01. This page needs minor revisions. Category: Definitions, Category: Hypothesis testing.