a 2017 study of 'binaural beats' - an auditory phenomenon in which sounds appear to oscillate between the ears - was retracted after it was revealed that 48 of the original 88 datasets had not been included in the original report  

original articles: Colzato et al., 2021 (Experimental Brain Research), Colzato et al., 2017 (Experimental Brain Research),

this story was in episode 6 #retraction #beat #responder #phacking

the error bar says

in a minor win for the transparency and quality of science, a paper published in 2017 was retracted after a University of Leiden investigation found that the first author of the paper had not included 48 of the original 88 participants in the published dataset. the first author disagreed with the retraction, while the other two authors agreed

the paper involved the acoustic phenomenon of binaural beats - the perception of two sounds oscillating between the two ears caused by differences in the frequencies of the sounds

this removal or selective analysis of data is a form of manipulation that researchers may sometimes get involved in. it's always bad. sometimes it's necessary if the original data quality is poor, or something didn't work, but it should always be reported fully and transparently in the paper


the system works