Some months ago we decided to put Discretizations on hiatus while we tried to figure out what to do about the tsunami of scientific papers washing up on PubMed (which was already piled deep in the flotsam and jetsam of publish or perish) now that China, India, Korea, etc. are getting in on the fun. On the one hand we're sorely tempted to take advantage of a situation that presents us daily with awesome posting opportunities like: Hay fever causes testicular cancer! On the other hand we know that most statistically significant findings are in fact false - in no small part because "it is a habit of mankind to entrust to careless hope what they long for, and to use sovereign reason to thrust aside what they do not desire". So, rather than being a part of the problem by propagating noise and not signal we've decided to limit our Discretizations to papers that 1) report an actual observation and not just the spectre of one drawn from data dredging and statistical analysis; 2) reflect an effort to reproduce a previously reported finding (e.g. testing a hypothesis previously drawn from statistical inference but in a different population); or, 3) extend (marginally) some previously established finding. Here goes:
No evidence found to support the hypothesis that parental smoking (and the benzene exposure that goes with it) during pregnancy increases a child's risk of developing acute lymphoblastic leukemia (ALL)
Does the debate about segregating cystic fibrosis patients by lung/bronchial microbiome, going so far as to prevent some/most/all-but-one CF sufferers from attending indoor CF Foundation events, presage future debates as more and more diseases are found to have bacterial components? Here are the pros and cons of the CF debate.