Bayesian inference of population prevalence
Within neuroscience, psychology, and neuroimaging, the most frequently used statistical approach is null hypothesis significance testing (NHST) of the population mean. An alternative approach is to perform NHST within individual participants and then infer, from the proportion of participants showin...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
eLife Sciences Publications Ltd
2021-10-01
|
Series: | eLife |
Subjects: | |
Online Access: | https://elifesciences.org/articles/62461 |
id |
doaj-56f9a507c79743bfae8f9f1f19f7305b |
---|---|
record_format |
Article |
spelling |
doaj-56f9a507c79743bfae8f9f1f19f7305b2021-10-06T13:49:22ZengeLife Sciences Publications LtdeLife2050-084X2021-10-011010.7554/eLife.62461Bayesian inference of population prevalenceRobin AA Ince0https://orcid.org/0000-0001-8427-0507Angus T Paton1Jim W Kay2Philippe G Schyns3School of Psychology and Neuroscience, University of Glasgow, Glasgow, United KingdomSchool of Psychology and Neuroscience, University of Glasgow, Glasgow, United KingdomDepartment of Statistics, University of Glasgow, Glasgow, United KingdomSchool of Psychology and Neuroscience, University of Glasgow, Glasgow, United KingdomWithin neuroscience, psychology, and neuroimaging, the most frequently used statistical approach is null hypothesis significance testing (NHST) of the population mean. An alternative approach is to perform NHST within individual participants and then infer, from the proportion of participants showing an effect, the prevalence of that effect in the population. We propose a novel Bayesian method to estimate such population prevalence that offers several advantages over population mean NHST. This method provides a population-level inference that is currently missing from study designs with small participant numbers, such as in traditional psychophysics and in precision imaging. Bayesian prevalence delivers a quantitative population estimate with associated uncertainty instead of reducing an experiment to a binary inference. Bayesian prevalence is widely applicable to a broad range of studies in neuroscience, psychology, and neuroimaging. Its emphasis on detecting effects within individual participants can also help address replicability issues in these fields.https://elifesciences.org/articles/62461statisticsgeneralisationinferenceprevalence |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Robin AA Ince Angus T Paton Jim W Kay Philippe G Schyns |
spellingShingle |
Robin AA Ince Angus T Paton Jim W Kay Philippe G Schyns Bayesian inference of population prevalence eLife statistics generalisation inference prevalence |
author_facet |
Robin AA Ince Angus T Paton Jim W Kay Philippe G Schyns |
author_sort |
Robin AA Ince |
title |
Bayesian inference of population prevalence |
title_short |
Bayesian inference of population prevalence |
title_full |
Bayesian inference of population prevalence |
title_fullStr |
Bayesian inference of population prevalence |
title_full_unstemmed |
Bayesian inference of population prevalence |
title_sort |
bayesian inference of population prevalence |
publisher |
eLife Sciences Publications Ltd |
series |
eLife |
issn |
2050-084X |
publishDate |
2021-10-01 |
description |
Within neuroscience, psychology, and neuroimaging, the most frequently used statistical approach is null hypothesis significance testing (NHST) of the population mean. An alternative approach is to perform NHST within individual participants and then infer, from the proportion of participants showing an effect, the prevalence of that effect in the population. We propose a novel Bayesian method to estimate such population prevalence that offers several advantages over population mean NHST. This method provides a population-level inference that is currently missing from study designs with small participant numbers, such as in traditional psychophysics and in precision imaging. Bayesian prevalence delivers a quantitative population estimate with associated uncertainty instead of reducing an experiment to a binary inference. Bayesian prevalence is widely applicable to a broad range of studies in neuroscience, psychology, and neuroimaging. Its emphasis on detecting effects within individual participants can also help address replicability issues in these fields. |
topic |
statistics generalisation inference prevalence |
url |
https://elifesciences.org/articles/62461 |
work_keys_str_mv |
AT robinaaince bayesianinferenceofpopulationprevalence AT angustpaton bayesianinferenceofpopulationprevalence AT jimwkay bayesianinferenceofpopulationprevalence AT philippegschyns bayesianinferenceofpopulationprevalence |
_version_ |
1716840604591194112 |