What is the colloquial name for the "look elsewhere" effect? The particle physics community tends to report the significance of a discovery by first calculating the p-values over a range of hypotheses (i.e. values of mass for the hypothetical particle), and then correcting for the "look elsewhere effect". As far as I can tell, this is particle physics jargon referring to the tendency of multiple trials to produce a significant p-value, even if the null hypothesis is true. 
But this is such a basic principle in statistics. Sure, it's ignored (much to the annoyance of any competent scientist) for reasons ranging from incompetence to propaganda, but it's still extremely well known. 
The problem: I can't get any good results typing "look-elsewhere effect" into Google. There must be some other, more colloquial, name for this, anyone know what it is? 
 A: Try the Wiki page for Multiple Comparisons. Here's blog post by statistician Kaiser Fung where he equates the LEE and MC problem, and links to (WSJ Numbers Guy) Karl Bialik's column discussing the 5 sigma approach in particle physics.
The main two points they make is that physicists have lots of data from controlled experiments, and are looking for tiny differences. This explains why the standard for accepting a finding is so much higher in physics than in medicine, where data is scarce and so effects need to be larger. 
The second reason is the laws of physics are supposed to be immutable, so even tiny deviations can overturn a theory. Other fields, particularly non-experimental social sciences, are not so lucky, so small variations generally not destroy such theories.   
A: Here is a clear, concise, explanation:
http://cms.web.cern.ch/news/should-you-get-excited-your-data-let-look-elsewhere-effect-decide
A: Generally this problem is couched in terms of multiple comparisons. Wiki refers to it in these terms http://en.wikipedia.org/wiki/Multiple_comparisons.
The text we used in uni was Maxwell and Delaney Designing Experiments and Analysing Data which contains a chapter on the subject as applied to basic between subjects ANOVAs. While these models are basic, I do think that you should be able to generalise the concepts to more advanced models.
