Brian Wansink is the head of the Cornell University’s food psychology research unit, the Food and Brand Lab. He was regarded as a social science star, because he’s written many articles about why and how we eat. This articles also gained a lot of media attention from prestigious newspapers and magazines. The core of his articles was that weight loss is possible for anyone by just making a few small changes to their environment without strict dieting or intense exercising. So, it were claims that the mass would like to belief.But some oddities came to light when Özge Sigirci, a young scientist from Turkey, came to visit the Cornell University. She immediately got an assignment from Wansink. He gave her a dataset from an experment in an Italian restaurant and asked her to analyze this dataset and to come up with nice results. Eventually this is what she did, she analyzed the data over and over to come up with some nice results. Wansink reported this in his blog and publicly praised Sigirci.But, statisticians and other researchers have argued that this is not how science should be done. They say that researchers should first come up with specific hypotheses before they conduct a study. This is different than what Wansink has done, he retroactively created hypotheses to fit the data after he conducted an experiment. So he got a lot of criticism on his blog post. For instance, over the last 14 months, critics have created “the Wansink Dossier” with a list of errors and inconsistencies which suggest that me manipulated his data. Cornell opened an investigation and since then five of his papers were retracted and 14 were corrected.It seems that Wansink and his team were aggressive manipulators of data. They even joked about mining datasets for impressive-looking results and that they should target journals with low standards. They also adapted their findings in the hope that the media would use it so that they would go “virally big time”.Many...