Aug

31

 There's an interesting new academic paper in this month's Science journal regarding the reproducibility of psychology experiments.

The researchers tried to replicate 100 experiments and found that the results could not be replicated for many of them (between 30 and 60%, depending on p-value). I highlight this paper because Vic has previously opined on the flaws of laboratory psychological experiments and this new paper supports his view– and it will surely get a lot of attention both due to the results and the prestige of Science Journal.

Interested specs and people who believe in the results from Behaviorial Finance experiments should read the paper and consider whether it affects their belief system. (I'd add that there is an epistemology paradox in this paper since this paper's findings need to be replicated too! Hah)

Abstract:

Reproducibility is a defining feature of science, but the extent to which it characterizes current research is unknown. We conducted replications of 100 experimental and correlational studies published in three psychology journals using high-powered designs and original materials when available. Replication effects were half the magnitude of original effects, representing a substantial decline. Ninety-seven percent of original studies had statistically significant results. Thirty-six percent of replications had statistically significant results; 47% of original effect sizes were in the 95% confidence interval of the replication effect size; 39% of effects were subjectively rated to have replicated the original result; and if no bias in original results is assumed, combining original and replication results left 68% with statistically significant effects. Correlational tests suggest that replication success was better predicted by the strength of original evidence than by characteristics of the original and replication teams.


Comments

Name

Email

Website

Speak your mind

Archives

Resources & Links

Search