The 7590 most puissant researchers of the year

Our biannual list recognises the most influential researchers, the true pioneers of science, the giants on whose shoulders the rest of us stand like so many flakes of dandruff, the ones who made the most impact, the ones that were on the telly all the time, who had a scienceornature paper (or two or three) and got interviewed about that thing with the hyperventilating press release, who got the most citations, or caused a stir, the movers and shakers who moved things and er… shook, because we felt like they didn’t have enough attention already.

Advertisement

The library where I grew up

What do you remember of the library where you grew up?

— Heather Christle (@heatherchristle) November 11, 2021

The library where I grew up was like a building from the future, with exciting organic shapes cast in concrete and a brightly coloured spiral staircase. Every wall was covered in shelves so sunlight crept in at ceiling level through a narrow strip of window. On summer afternoons, the sun came through at an angle like laser beams that made the dust motes boil in the still air.

The alleged global warming hiatus

Tags

review

Theodore G. Shepherd has a new paper “Bringing physical reasoning into statistical practice in climate-change science” which looks at the disconnect between physical reasoning and statistical practice and bringing the former into the latter. It rails against Null Hypothesis Significance Testing. There’s a grand tradition of papers doing this and they are invariably interesting to read even though the message is usually a variation on a well known theme. This paper talks more generally about statistical rituals of which NHST is but one and Shepherd notes that he’s occasionally to be seen performing a ritual or two himself (who isn’t?):

I suspect I am not alone in admitting that most of the statistical tests in my own papers are performed in order to satisfy these rituals, rather than as part of the scientific discovery process itself.

It outlines instances where rituals can get you into trouble (multiple testing, base rate neglect, prosecutor fallacy etc.) and this is all fair enough and all familiar ground. The interesting variations involve delving into the assumptions of NHST. Science can be complex, particularly climate science, so care should be taken not to over simplify. At the same time, complexity isn’t necessarily to be favoured.

It’s all very complex.

Or simple.