Academic journal article Psychonomic Bulletin & Review

Publication Bias and the Failure of Replication in Experimental Psychology

Academic journal article Psychonomic Bulletin & Review

Publication Bias and the Failure of Replication in Experimental Psychology

Article excerpt

Published online: 4 October 2012

© Psychonomic Society, Inc. 2012

Abstract Replication of empirical findings plays a fundamental role in science. Among experimental psychologists, successful replication enhances belief in a finding, while a failure to replicate is often interpreted to mean that one of the experiments is flawed. This view is wrong. Because experimental psychology uses statistics, empirical findings should appear with predictable probabilities. In a misguided effort to demonstrate successful replication of empirical findings and avoid failures to replicate, experimental psychologists sometimes report too many positive results. Rather than strengthen confidence in an effect, too much successful replication actually indicates publication bias, which invalidates entire sets of experimental findings. Researchers cannot judge the validity of a set of biased experiments because the experiment set may consist entirely of type I errors. This article shows how an investigation of the effect sizes from reported experiments can test for publication bias by looking for too much successful replication. Simulated experiments demonstrate that the publication bias test is able to discriminate biased experiment sets from unbiased experiment sets, but it is conservative about reporting bias. The test is then applied to several studies of prominent phenomena that highlight how publication bias contaminates some findings in experimental psychology. Additional simulated experiments demonstrate that using Bayesian methods of data analysis can reduce (and in some cases, eliminate) the occurrence of publication bias. Such methods should be part of a systematic process to remove publication bias from experimental psychology and reinstate the important role of replication as a final arbiter of scientific findings.

Keywords Bayesian methods . Hypothesis testing . Meta-analysis . Publication bias . Replication

(ProQuest: ... denotes formulae omitted.)

Introduction

Imagine that you read about a set of 10 experiments that describe an effect (call it effect "A"). The experiments appear to be conducted properly, and across the set of experiments, the null hypothesis was rejected 9 times out of 10. Next, you read about another set of 19 experiments that describe effect "B." Again, the experiments appear to be conducted properly, and the null hypothesis was rejected 10 times out of 19 experiments. I suspect most experimental psychologists would express stronger belief in effect A than in effect B. After all, effect A is so strong that almost every test was statistically significant, while effect B rejected the null hypothesis only about half of the time. Replication is commonly used to weed out false effects and verify scientific truth. Unfortunately, faith in replication is unfounded, at least as science is frequently practiced in experimental psychology.

Effect A is based on the series of experiments by Bem (2011) that reported evidence of people using Psi ability to gain knowledge from the future. Even after hearing about this study's findings, most psychologists do not believe that people can get information from the future. This persistent disbelief raises serious questions about how people should and do interpret experimental findings in psychology. If researchers remain skeptical of a finding that has a 90% successful replication rate, it would seem that they should be skeptical of almost all findings in experimental psychology. If so, one must consider whether it is worthwhile to spend time and money on experiments that do not change anyone's beliefs.

Effect B is based on a meta-analysis of a set of experiments that describe the bystander effect (Fischer et al., 2011), which is the empirical observation that people are less likely to help someone in distress if there are other people around. I do not know of anyone who doubts that the bystander effect is real, and it is frequently discussed in introductory psychology textbooks (e. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.