Skip to playerSkip to main content
  • 1 year ago
Published scientific studies can motivate research, inspire products, and inform policy. However, recent studies that examined dozens of published pharmaceutical papers managed to replicate the results of less than 25% of them — and similar results have been found in other scientific disciplines. How do we combat this crisis of scientific irreproducibility? Matt Anticole investigates.

Lesson by Matt Anticole, animation by Brett Underhill.

Category

📚
Learning
Transcript
00:00In 2011, a team of physicists reported a startling discovery.
00:11Neutrinos traveled faster than the speed of light,
00:14by 60 billionths of a second,
00:16in their 730-kilometer trip from Geneva to a detector in Italy.
00:20Despite six months of double-checking,
00:23the bizarre discovery refused to yield.
00:25But rather than celebrating a physics revolution,
00:28the researchers published a cautious paper arguing for continued research
00:32in an effort to explain the observed anomaly.
00:35In time, the error was tracked to a single incorrectly connected fiber-optic cable.
00:41This example reminds us that real science is more than static textbooks.
00:45Instead, researchers around the world are continuously publishing their latest discoveries,
00:51with each paper adding to the scientific conversation.
00:54Published studies can motivate future research,
00:57inspire new products,
00:59and inform government policy.
01:01So, it's important that we have confidence in the published results.
01:04If their conclusions are wrong,
01:06we risk time, resources, and even our health in the pursuit of false leads.
01:11When findings are significant,
01:13they are frequently double-checked by other researchers,
01:16either by reanalyzing the data or by redoing the entire experiment.
01:21For example, it took repeated investigation of the CERN data before the timing error was tracked down.
01:28Unfortunately, there are currently neither the resources nor professional incentives
01:33to double-check the more than one million scientific papers published annually.
01:38Even when papers are challenged, the results are not reassuring.
01:42Recent studies that examined dozens of published pharmaceutical papers
01:46managed to replicate the results of less than 25% of them,
01:50and similar results have been found in other scientific disciplines.
01:54There are a variety of sources for irreproducible results.
01:57Errors could hide in the original design, execution, or analysis of the data.
02:03Unknown factors such as patient's undisclosed condition in a medical study
02:07can produce results that are not repeatable in new test subjects.
02:11And sometimes, the second research group can't reproduce the original results
02:15simply because they don't know exactly what the original group did.
02:20However, some problems might stem from systematic decisions in how we do science.
02:25Researchers, the institutions that employ them,
02:28and the scientific journals that publish findings
02:31are expected to produce big results frequently.
02:34Important papers can advance careers, generate media interest,
02:39and secure essential funding.
02:41So there's slim motivation for researchers to challenge their own exciting results.
02:45In addition, little incentive exists to publish results unsupportive of the expected hypothesis.
02:51That results in a deluge of agreement between what was expected and what was found.
02:56In rare occasions, this can even lead to deliberate fabrication,
03:00such as in 2013 when a researcher spiked rabbit blood with human blood
03:05to give false evidence that his HIV vaccine was working.
03:09The publish or perish mindset can also compromise academic journals' traditional peer review processes,
03:15which are safety checks where experts examine submitted papers for potential shortcomings.
03:20The current system, which might involve only one or two reviewers, can be woefully ineffective.
03:25That was demonstrated in a 1998 study where eight weaknesses were deliberately inserted into papers,
03:32but only around 25% were caught upon review.
03:35Many scientists are working toward improving reproducibility in their fields.
03:40There's a push to make researchers' raw data, experimental procedures,
03:44and analytical techniques more openly available in order to ease replication efforts.
03:50The peer review process can also be strengthened to more efficiently weed out weak papers prior to publication.
03:57And we could temper the pressure to find big results
04:00by publishing more papers that failed to confirm the original hypothesis,
04:04an event that happens far more than current scientific literature suggests.
04:08Science always has and always will encounter some false starts
04:12as part of the collective acquisition of new knowledge.
04:15Finding ways to improve the reproducibility of our results
04:18can help us weed out those false starts more effectively,
04:21keeping us moving steadily toward exciting new discoveries.
04:25An example of the
Comments

Recommended