Cargando…

P-curve won’t do your laundry, but it will distinguish replicable from non-replicable findings in observational research: Comment on Bruns & Ioannidis (2016)

p-curve, the distribution of significant p-values, can be analyzed to assess if the findings have evidential value, whether p-hacking and file-drawering can be ruled out as the sole explanations for them. Bruns and Ioannidis (2016) have proposed p-curve cannot examine evidential value with observati...

Descripción completa

Detalles Bibliográficos
Autores principales: Simonsohn, Uri, Nelson, Leif D., Simmons, Joseph P.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Public Library of Science 2019
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6411107/
https://www.ncbi.nlm.nih.gov/pubmed/30856227
http://dx.doi.org/10.1371/journal.pone.0213454
_version_ 1783402339594403840
author Simonsohn, Uri
Nelson, Leif D.
Simmons, Joseph P.
author_facet Simonsohn, Uri
Nelson, Leif D.
Simmons, Joseph P.
author_sort Simonsohn, Uri
collection PubMed
description p-curve, the distribution of significant p-values, can be analyzed to assess if the findings have evidential value, whether p-hacking and file-drawering can be ruled out as the sole explanations for them. Bruns and Ioannidis (2016) have proposed p-curve cannot examine evidential value with observational data. Their discussion confuses false-positive findings with confounded ones, failing to distinguish correlation from causation. We demonstrate this important distinction by showing that a confounded but real, hence replicable association, gun ownership and number of sexual partners, leads to a right-skewed p-curve, while a false-positive one, respondent ID number and trust in the supreme court, leads to a flat p-curve. P-curve can distinguish between replicable and non-replicable findings. The observational nature of the data is not consequential.
format Online
Article
Text
id pubmed-6411107
institution National Center for Biotechnology Information
language English
publishDate 2019
publisher Public Library of Science
record_format MEDLINE/PubMed
spelling pubmed-64111072019-04-01 P-curve won’t do your laundry, but it will distinguish replicable from non-replicable findings in observational research: Comment on Bruns & Ioannidis (2016) Simonsohn, Uri Nelson, Leif D. Simmons, Joseph P. PLoS One Formal Comment p-curve, the distribution of significant p-values, can be analyzed to assess if the findings have evidential value, whether p-hacking and file-drawering can be ruled out as the sole explanations for them. Bruns and Ioannidis (2016) have proposed p-curve cannot examine evidential value with observational data. Their discussion confuses false-positive findings with confounded ones, failing to distinguish correlation from causation. We demonstrate this important distinction by showing that a confounded but real, hence replicable association, gun ownership and number of sexual partners, leads to a right-skewed p-curve, while a false-positive one, respondent ID number and trust in the supreme court, leads to a flat p-curve. P-curve can distinguish between replicable and non-replicable findings. The observational nature of the data is not consequential. Public Library of Science 2019-03-11 /pmc/articles/PMC6411107/ /pubmed/30856227 http://dx.doi.org/10.1371/journal.pone.0213454 Text en © 2019 Simonsohn et al http://creativecommons.org/licenses/by/4.0/ This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/) , which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
spellingShingle Formal Comment
Simonsohn, Uri
Nelson, Leif D.
Simmons, Joseph P.
P-curve won’t do your laundry, but it will distinguish replicable from non-replicable findings in observational research: Comment on Bruns & Ioannidis (2016)
title P-curve won’t do your laundry, but it will distinguish replicable from non-replicable findings in observational research: Comment on Bruns & Ioannidis (2016)
title_full P-curve won’t do your laundry, but it will distinguish replicable from non-replicable findings in observational research: Comment on Bruns & Ioannidis (2016)
title_fullStr P-curve won’t do your laundry, but it will distinguish replicable from non-replicable findings in observational research: Comment on Bruns & Ioannidis (2016)
title_full_unstemmed P-curve won’t do your laundry, but it will distinguish replicable from non-replicable findings in observational research: Comment on Bruns & Ioannidis (2016)
title_short P-curve won’t do your laundry, but it will distinguish replicable from non-replicable findings in observational research: Comment on Bruns & Ioannidis (2016)
title_sort p-curve won’t do your laundry, but it will distinguish replicable from non-replicable findings in observational research: comment on bruns & ioannidis (2016)
topic Formal Comment
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6411107/
https://www.ncbi.nlm.nih.gov/pubmed/30856227
http://dx.doi.org/10.1371/journal.pone.0213454
work_keys_str_mv AT simonsohnuri pcurvewontdoyourlaundrybutitwilldistinguishreplicablefromnonreplicablefindingsinobservationalresearchcommentonbrunsioannidis2016
AT nelsonleifd pcurvewontdoyourlaundrybutitwilldistinguishreplicablefromnonreplicablefindingsinobservationalresearchcommentonbrunsioannidis2016
AT simmonsjosephp pcurvewontdoyourlaundrybutitwilldistinguishreplicablefromnonreplicablefindingsinobservationalresearchcommentonbrunsioannidis2016