Menu

Psychology’s Failure to Replicate

Researchers react to the finding that most of 100 studies recently analyzed were not reproducible.

Aug 31, 2015
Amanda B. Keener

PIXABAY, FOUNDRY

In the wake of a report released last week (August 28) in Science, which showed that just 39 percent of 100 studies published in 2008 in the top three psychology journals could be successfully reproduced, researchers are ready to clean up the field’s literature. The results come from a collaborative effort of 270 scientists working for The Reproducibility Project based at the Center for Open Science in Charlottesville, Virginia.

“It’s like we’ve come clean,” Alan Kraut, the executive director of the Association for Psychological Science, told The New York Times. “This kind of correction is something that has to happen across science, and I’m proud that psychology is leading the charge on this.”

According The New York Times, the study authors noted that the lack of reproducibility for more than 60 percent of the studies was not linked to fraud or misconduct, but rather may reflect a collection of published conclusions that were stronger than their data. Another possibility is that the study reproduction results were wrong, or that small differences in study design made big impacts on the results of both analyses.

“This project is not evidence that anything is broken. Rather, it’s an example of science doing what science does,” study coauthor Cody Christopherson, a psychology researcher at Southern Oregon University, told Smithsonian Magazine. “It’s impossible to be wrong in a final sense in science. You have to be temporarily wrong, perhaps many times, before you are ever right,” he added.

Study coauthor Brian Nosek, a psychologist at the University of Virginia, told The Verge that the incentive to publish novel findings likely contributes to irreproducible results and discourages attempts to reproduce existing work. “If this occurs on a broad scale,” he said, “then the published literature may be more beautiful than reality.”

June 2019

Living with Bacteria

Can pathogens be converted to commensals?

Marketplace

Sponsored Product Updates

Best Practices: Calculating Cell Confluency
Best Practices: Calculating Cell Confluency
In this white paper, learn how to use a cell imager system to directly and accurately capture and calculate cell confluency!
LabTwin's AI-powered Digital Assistant Now Talks Back and Connects Data Sources in the Lab with New Open API
LabTwin's AI-powered Digital Assistant Now Talks Back and Connects Data Sources in the Lab with New Open API
LabTwin GmbH, the world's first voice and AI-powered digital lab assistant, today announced its new open API that will connect scientists with data sources both inside and outside of the lab. 
BCG Digital Ventures and Sartorius Help Launch the World's First Voice-powered Digital Assistant for Scientists
BCG Digital Ventures and Sartorius Help Launch the World's First Voice-powered Digital Assistant for Scientists
LabTwin GmbH, an independent company backed by Boston Consulting Group Digital Ventures (BCG Digital Ventures) and leading biopharma supplier, Sartorius, today announced the launch of the world's first voice and AI-powered digital lab assistant.
Understanding Transcriptomic or Proteomic Datasets to Reveal Biological Mechanisms
Understanding Transcriptomic or Proteomic Datasets to Reveal Biological Mechanisms
When analyzing large transcriptomics or proteomics datasets, we want to understand whether the phenomenon is unusual or commonplace and whether there are informative similarities to other areas of biology. To learn more about how Ingenuity Pathway Analysis (IPA®) and Analysis Match can help, download this white paper from QIAGEN!