First the link to this week’s complete list as HTML and as PDF.
***
I have seen and commented on many sociological and psychological findings of dubious substance, relevance, and (non-statistical) significance, but Schilke et al. have set a new low. In a reply to comment they seriously cite a correlation of less than r2 = 0.01 to bolster their argument. How many zeroes behind the decimal point do you need before you can call a correlation nonexistent?
***
Bohannon throws a glaring highlight on a theme I’ve been banging on about for ages. Have any of those pseudonymous co-authors even seen the final draft of the article and has any one of them been given the right to question the conclusions and to demand changes? If so, they are authors, bear responsibility for the content and need to be named. If not, they’re not authors and their rightful place is in the acknowledgements section. Traditionally not only technicians and students doing the work and providers of samples are mentioned here but also the contributors of ideas, often central and important ideas. The point here is, that these ideas were filtered through the main author and his grasp of them and the originators had no say in what has become of them in the final script. Heeding this simple and very obvious rule, the whole issue of pseudonyms becomes a non-problem.
***
When farmland is given up and let fall fallow, habitat and ecology changes. That much is trivial. But why has change and adaptation in the past always been beneficial and is change in the future always detrimental, as Normile claims again?
***
Rosario-Ortiz et al. is an article one could easily base a high-school exam on.
“In the Netherlands […] pipe networks are, on average, 33 to 37 years old. Although there are regional differences, an estimated 22% of the pipes in the United States are more than 50 years old; the average age of pipe at failure is 47 years, and only 43% of pipes are considered to be in good or excellent condition.”
Questions:
1) Is the average age of US pipes higher or lower than of Netherlands ones?
2) Can that question even be answered from the data given?
3) Does such a hodgepodge of unrelated numbers make any sense at all?
4) Do the writers have the slightest understanding of what they’re writing about?
***
Tsetsos et al. reminds me of similar results for the efficiency of prejudices. New information you receive may well be just noise as discussed here and it pays to stand by your previous considered opinions and prejudices until sufficiently strong proof arrives. This is particularly valid for scientific fads and fashions. A classic example is the Neolithic migration, which became totally out of date with cultural diffusionism holding sway until recent DNA results completely vindicated the old and old-fashioned view. A simpler case, much nearer to what this article actually deals with, is the choice of ice cream. It is far less effort just to say e.g. Stracciatella every time than to look at and evaluate all the choices on offer, even if you were to miss a slightly better choice every once in a while that way. Human choice is seldom a luxury item and mostly has to be made quickly, in a hurry and on incomplete data, so Tsetsos et al.’s optimisation sounds entirely plausible and worthwhile.