THE LACK OF REPRODUCIBILITY IN RESEARCH HOW STATISTICS CAN ENDORSE RESULTS Academic Article

abstract

• 2015, Universitat de Valencia. All rights reserved. Scientific research is validated by reproduction of the results, but efforts to reproduce spurious claims drain resources. We focus on one cause of such failure: false positive statistical test results caused by random variability. Classical statistical methods rely on p-values to measure the evidence against null hypotheses, but Bayesian hypothesis testing produces more easily understood results, provided one can specify prior distributions under the alternative hypothesis. We describe new tests, UMPBTs, which are Bayesian tests that provide default specification of alternative priors, and show that these tests also maximize statistical power.

authors

• Johnson, Valen

published proceedings

• METODE-POPULAR SCIENCE JOURNAL

altmetric score

• 3

author list (cited authors)

• Goddard, S. D., & Johnson, V. E.

citation count

• 0

complete list of authors

• Goddard, Scott D||Johnson, Valen E

publication date

• January 2015

publisher

• Universitat de Valencia  Publisher

keywords

• Bayesian Analysis
• Hypothesis Test
• Statistical Evidence
• Uniformly Most Powerful Bayesian Tests

Digital Object Identifier (DOI)

• 10.7203/metode.83.3913

start page

• 175

end page

• 179

volume

• 0

issue

• 5

URL

• http%3A%2F%2Fdx.doi.org%2F10.7203%2Fmetode.83.3913