Exam proctoring

The class for which I TA had their exam today. I helped Prof. Heller out by going over the exam questions ahead of time, just to make sure that the poor little babies weren't overly challenged. (We wouldn't want to hurt their confidence going into exam week, now would we?)

For one of the questions, Heller wanted to use some data that was actually obtained by one of the students in the class. I dug it up for him but it wasn't exactly to his liking. "Do you think you could find a spectrum which has minor impurity peaks in it?" I told him I couldn't, but I said I would have no problem inventing the data in excel and producing a fake spectrum which was indistinguishable from the real thing. This is apparently not something your PhD adviser likes to hear, although he wasn't so appalled as to not use the spectrum. It was a little frightening how easy it was to make a plausible looking spectrum from completely falsified data. A test for the reader....three of the following spectra are real, the fourth is invented. Can you tell the difference? (click on the image to enlarge it)