An NMR spectrum provides important data on what’s in the NMR sample. Students are taught to look for certain contaminants – TMS, CHCl3, H2O – as a matter of routine. As many know, however, removing the last traces of chromatography solvent, unreacted starting material, and so on, can be a difficult chore so some Japanese science students have apparently hit on an elegant solution: using the ‘Delete Peak’ option in the JEOL NMR software to remove unwanted peaks. Needless to say, the editors of the journals that had published these spectra, and the students’ research mentors (once they were informed about the spectra) were not amused (“Cleaning up the record,” C&ENews, April 21, 2014, p. 32-3).
‘Silent’ alteration of data (by ‘silent’ I mean that the alteration was not reported to a referee or a reader), even if it involves removing ‘unimportant’ peaks, or peaks due to impurities, is nothing less than scientific fraud. If you have questions about how to process your data, or what data and data processing procedures to report, always bring your questions to your research supervisor, the journal editor, and colleagues in the field. Scientists, and more often, science students, tend to keep such questions to themselves because they think they will lose their colleagues’ respect by asking a “stupid question.” In fact, the sure-fire way to lose this respect is to engage in fraudulent practices.