Prologue
This is the story of an academic researcher who identified a massive error in her own published work. She realised that she had reached the wrong conclusions because of the error.
The paper and the mistake
Julia Strand is an assistant professor of psychology at Carleton College. In 2018, she and 2 co-authors published a paper in the Psychonomic Bulletin & Review. It is a paper about how people strain to listen in crowded spaces. The paper is entitled Talking points: A modulating circle reduces listening effort without improving speech recognition.
Her team pre-registereed the analyses. They shared the materials, data and code on the Open Science Framework.
She obtained a grant to carry out follow-up researches. Her paper was the basis of a student’s masters project.
Several months later, they started a follow-up study and tried to replicate the research. The findings did not replicate. She found out that they had unintentially programmed the timing clock wrongly in the original research. That means that the discovery in the original research is a programming bug. She discovered the mistake alone at her laptop in her home in the evening. The bug was hard to identify.
What to do
On the next day, she notified the co-authors, the editor and the publisher, the NIH program officer overseeing her grant, her department and a student who had used the data for her master thesis. She notified the editor and the publisher to initiate a retraction. She notiied the student to cancel her master thesis defence. She notified her department because they oversee her tenure review. She also notified her research students.
The editor and publisher did not opt to retract the paper. They decided to publish a revised version of the article, linked to the original paper.
She shared her experience here.
The Retract Watch wrote an article on her encouraging experience. This is because this is the way things should be in science.
Epilogue
“Each of us builds our discoveries on the work of others; if that work is false, our constructions fall like a house of cards and we must start all over again.” — Nobel Laureate Michael Bishop.
“Over many centuries, researchers have developed professional standards designed to enhance the progress of science and to avoid or minimize the difficulties of research.… Researchers have three sets of obligations that motivate their adherence to professional standards. First, researchers have an obligation to honor the trust that their colleagues place in them.… Second, researchers have an obligation to themselves. Irresponsible conduct in research can make it impossible to achieve a goal.… Third, because scientific results greatly influence society, researchers have an obligation to act in ways that serve the public.” — On Being a Scientist: A Guide to Responsible Conduct in Research. National Academies. 1988. From John F. Ahearne. “Honesty: Ultimately, ethics in scientific publishing, as in life, comes down to one word.” American Scientist. Volume 99, Number 2.