Hypothesis testing, power and sample size determination for between group comparisons in fMRI experiments

Dulal K. Bhaumik; Anindya Roy; Nicole A. Lazar; Kush Kapur; Subhash Aryal; John A. Sweeney; Dave Patterson; Robert D. Gibbons

doi:10.1016/j.stamet.2008.05.003

Hypothesis testing, power and sample size determination for between group comparisons in fMRI experiments

Dulal K. Bhaumik, Anindya Roy, Nicole A. Lazar, Kush Kapur, Subhash Aryal, John A. Sweeney, Dave Patterson, Robert D. Gibbons

Research output: Contribution to journal › Article › peer-review

12 Scopus citations

Abstract

Modern methods for imaging the human brain, such as functional magnetic resonance imaging (fMRI) present a range of challenging statistical problems. In this paper, we first develop a large sample based test for between group comparisons and use it to determine the necessary sample size in order to obtain a target power via simulation under various alternatives for a given pre-specified significance level. Both testing and sample size calculations are particularly critical for neuroscientists who use these new techniques, since each subject is expensive to image.

Original language	English
Pages (from-to)	133-146
Number of pages	14
Journal	Statistical Methodology
Volume	6
Issue number	2
DOIs	https://doi.org/10.1016/j.stamet.2008.05.003
State	Published - Mar 2009

Keywords

Brain imaging
Equality of proportion
False discovery rate
Large sample test
Regions of interest

Access to Document

10.1016/j.stamet.2008.05.003

Cite this

@article{ebe366687427486abbb41a2e467f251e,

title = "Hypothesis testing, power and sample size determination for between group comparisons in fMRI experiments",

abstract = "Modern methods for imaging the human brain, such as functional magnetic resonance imaging (fMRI) present a range of challenging statistical problems. In this paper, we first develop a large sample based test for between group comparisons and use it to determine the necessary sample size in order to obtain a target power via simulation under various alternatives for a given pre-specified significance level. Both testing and sample size calculations are particularly critical for neuroscientists who use these new techniques, since each subject is expensive to image.",

keywords = "Brain imaging, Equality of proportion, False discovery rate, Large sample test, Regions of interest",

author = "Bhaumik, {Dulal K.} and Anindya Roy and Lazar, {Nicole A.} and Kush Kapur and Subhash Aryal and Sweeney, {John A.} and Dave Patterson and Gibbons, {Robert D.}",

note = "Funding Information: The authors thank an associate editor and two referees for their insightful comments that significantly improved the quality of this article. This work was supported in part by a grant from the National Institute of Mental Health (1P50HD055751-01).",

year = "2009",

month = mar,

doi = "10.1016/j.stamet.2008.05.003",

language = "English",

volume = "6",

pages = "133--146",

journal = "Statistical Methodology",

issn = "1572-3127",

publisher = "Elsevier",

number = "2",

}

TY - JOUR

T1 - Hypothesis testing, power and sample size determination for between group comparisons in fMRI experiments

AU - Bhaumik, Dulal K.

AU - Roy, Anindya

AU - Lazar, Nicole A.

AU - Kapur, Kush

AU - Aryal, Subhash

AU - Sweeney, John A.

AU - Patterson, Dave

AU - Gibbons, Robert D.

N1 - Funding Information: The authors thank an associate editor and two referees for their insightful comments that significantly improved the quality of this article. This work was supported in part by a grant from the National Institute of Mental Health (1P50HD055751-01).

PY - 2009/3

Y1 - 2009/3

N2 - Modern methods for imaging the human brain, such as functional magnetic resonance imaging (fMRI) present a range of challenging statistical problems. In this paper, we first develop a large sample based test for between group comparisons and use it to determine the necessary sample size in order to obtain a target power via simulation under various alternatives for a given pre-specified significance level. Both testing and sample size calculations are particularly critical for neuroscientists who use these new techniques, since each subject is expensive to image.

AB - Modern methods for imaging the human brain, such as functional magnetic resonance imaging (fMRI) present a range of challenging statistical problems. In this paper, we first develop a large sample based test for between group comparisons and use it to determine the necessary sample size in order to obtain a target power via simulation under various alternatives for a given pre-specified significance level. Both testing and sample size calculations are particularly critical for neuroscientists who use these new techniques, since each subject is expensive to image.

KW - Brain imaging

KW - Equality of proportion

KW - False discovery rate

KW - Large sample test

KW - Regions of interest

UR - http://www.scopus.com/inward/record.url?scp=61849161157&partnerID=8YFLogxK

U2 - 10.1016/j.stamet.2008.05.003

DO - 10.1016/j.stamet.2008.05.003

M3 - Article

AN - SCOPUS:61849161157

SN - 1572-3127

VL - 6

SP - 133

EP - 146

JO - Statistical Methodology

JF - Statistical Methodology

IS - 2

ER -

Hypothesis testing, power and sample size determination for between group comparisons in fMRI experiments

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this