‘Bad science’ spreads through natural selection, says study

New research also finds that the statistical power of studies in social and behavioural science has not improved in half a century

September 21, 2016
Dr Frankenstein and Igor in laboratory
Source: Getty
Bad science: laboratory methods can propagate directly, through the production of graduate students who go on to start their own labs, or indirectly

“Bad science” spreads through universities in a process similar to Darwinian natural selection, a paper has argued.

It coins the phrase “the natural selection of bad science”, whereby laboratories that use potentially shaky methods that lead to lots of publication produce successful graduate students who spread these methods when they themselves open new labs.

“Laboratory methods can propagate either directly, through the production of graduate students who go on to start their own labs, or indirectly, through prestige-biased adoption by researchers in other labs,” the paper argues.

“Methods which are associated with greater success in academic careers will, other things being equal, tend to spread.”

A “replication crisis” in science, in which researchers are not able to reproduce others’ results, has been blamed on a number of types of “bad science”, including a reluctance to double-check findings and statistical manipulations to get a publishable result.  

This latest paper focuses on the problem of statistically underpowered studies: using too few subjects in an experiment for the results to be reliable.

It found that statistical power in the social and behavioural sciences has not improved since the first warnings of underpowered experiments in the early 1960s.

“Statistical power is quite low”, found "The natural selection of bad science", published in Royal Society Open Science, and “more importantly, statistical power shows no sign of increase over six decades”.

“The data are far from a complete picture of any given field or of the social and behavioural sciences more generally, but they help explain why false discoveries appear to be common,” it says.

Last year, an attempt to reproduce 100 prominent psychology papers found that only about a third managed to replicate statistically significant results.

david.matthews@tesglobal.com

You've reached your article limit

Register to continue

Registration is free and only takes a moment. Once registered you can read a total of 6 articles each month, plus:

  • Sign up for the editor's highlights
  • Receive World University Rankings news first
  • Get job alerts, shortlist jobs and save job searches
  • Participate in reader discussions and post comments
Register

Have your say

Log in or register to post comments

Featured Jobs

Research Assistant LONDON SCHOOL OF ECONOMICS & POLITICAL SCIENCE LSE
Lecturer in University Study Skills UNIVERSITY OF HAFR AL BATIN
Lecturer in English Language UNIVERSITY OF HAFR AL BATIN

Most Commented

question marks PhD study

Selecting the right doctorate is crucial for success. Robert MacIntosh and Kevin O'Gorman share top 10 tips on how to pick a PhD

India, UK, flag

Sir Keith Burnett reflects on what he learned about international students while in India with the UK prime minister

Pencil lying on open diary

Requesting a log of daily activity means that trust between the institution and the scholar has broken down, says Toby Miller

Application for graduate job
Universities producing the most employable graduates have been ranked by companies around the world in the Global University Employability Ranking 2016
Construction workers erecting barriers

Directly linking non-EU recruitment to award levels in teaching assessment has also been under consideration, sources suggest