Statistical tools used to identify scientific misconduct in mobile phone research (REFLEX program)
Abstract
A severe case of scientific misconduct was discovered in a paper from 2005 allegedly showing harmful effects (DNA breakage) of non-thermal mobile phone electromagnetic field exposure on human and rat cells. Here we describe the way how the fraudulent data were identified. The low variations of the reported biological data are shown to be below theoretical lower limits (multinomial distributions). Another reason for doubts was highly significant non-equal distributions of last digits, a known hint towards data fabrication. The Medical University Vienna, where the research was conducted, was informed about these findings and came to the conclusion that the data in this and another, related paper by the same group were fabricated, and that both papers should be retracted.
Cite
@article{arxiv.0807.2554,
title = {Statistical tools used to identify scientific misconduct in mobile phone research (REFLEX program)},
author = {Alexander Lerchl and Adalbert FX Wilhelm},
journal= {arXiv preprint arXiv:0807.2554},
year = {2015}
}
Comments
This paper has been withdrawn by the authors due to concerns with referenced data sets. No further revisions.