Advanced

Bayes Factors: A ‘re-volution’ in psychology

Patching, Geoffrey LU (2018) Bayes@Lund 2018
Abstract
In psychology, Bayes Factors (BFs) are being increasingly reported as a complement to p-values. However, just as the mindless computation of p-values encourages a simple dichotomization of study results, the mindless computation of BFs with an overemphasis on hypothesis testing detracts from more useful approaches to interpreting study results, such as parameter estimation. BFs may serve to facilitate the transition from frequentist to Bayesian data analysis, but let us not repeat mistakes of the past. In this short talk, I shall argue that Bayesian parameter estimation is preferred with an example from a recent Master thesis.
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to conference
publication status
published
subject
conference name
Bayes@Lund 2018
conference location
Lund, Sweden
conference dates
2018-04-12 - 2018-04-12
language
English
LU publication?
yes
id
bcaef1cd-a832-45ef-bbd3-129aa327a9f2
alternative location
https://bayesat.github.io/lund2018/bayes_at_lund_2018.html
date added to LUP
2018-10-02 12:29:23
date last changed
2018-11-21 21:41:57
@misc{bcaef1cd-a832-45ef-bbd3-129aa327a9f2,
  abstract     = {In psychology, Bayes Factors (BFs) are being increasingly reported as a complement to p-values. However, just as the mindless computation of p-values encourages a simple dichotomization of study results, the mindless computation of BFs with an overemphasis on hypothesis testing detracts from more useful approaches to interpreting study results, such as parameter estimation. BFs may serve to facilitate the transition from frequentist to Bayesian data analysis, but let us not repeat mistakes of the past. In this short talk, I shall argue that Bayesian parameter estimation is preferred with an example from a recent Master thesis.},
  author       = {Patching, Geoffrey},
  language     = {eng},
  title        = {Bayes Factors: A ‘re-volution’ in psychology},
  url          = {https://bayesat.github.io/lund2018/bayes_at_lund_2018.html},
  year         = {2018},
}