A Power Calculator for the Classical Twin Design

• Published in: Behavior genetics Vol. 47; no. 2; pp. 255 - 261
• Main Author: Verhulst, Brad
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.03.2017; Springer Nature B.V
• Subjects: Algorithms; Analysis; Behavioral Science and Psychology; Brief Communication; Clinical Psychology; Financing; Genetics; Health Psychology; Humans; Models, Genetic; Power; Power structure; Psychology; Public Health; Review boards; Statistical analysis; Statistics; Statistics as Topic - methods; Studies; Twin studies; Twins; Twins - statistics & numerical data; United States > US; Virginia; Biometrical genetics; Variance components; Twin study; Power
• ISSN: 0001-8244; 1573-3297
• DOI: 10.1007/s10519-016-9828-9

Power is a ubiquitous, though often overlooked, component of any statistical analyses. Almost every funding agency and institutional review board requires that some sort of power analysis is conducted prior to data collection. While there are several excellent on line power calculators for independent observations, twin studies pose unique challenges that are not incorporated into these algorithms. The goal of the current manuscript is to outline a general method for calculating power in twin studies, and to provide functions to allow researchers to easily conduct power analyses for a range of common twin models. Several scenarios are discussed to demonstrate the importance of various factors that influence the power within the classical twin design and to serve as examples for the provided functions.