The average mileage in the study was reported between 24,000 and 27,000 miles per year, which is unusually high compared to the average American driver who drives around 13,000 miles a year. Ariely initially claimed the drivers were senior citizens in Florida, but the data provided by the insurance company showed no difference between those who signed at the top and those who signed at the bottom.
Bazerman felt overwhelmed by the evidence and acknowledged his involvement as a co-author on multiple papers with Francesca Gino. He was concerned about the potential personal and professional consequences and felt it was not his place to be the central figure in exposing the fraud.
The Data Colada team, consisting of Leif Nelson, Uri Simonsohn, and Joseph Simmons, started investigating academic research papers because they noticed a pattern of findings that didn't align with their intuition and statistical impossibilities in the data. They aimed to demonstrate how easy it is to produce significant but unreliable results through practices like p-hacking.
Academic fraud is a significant concern beyond the academic community because it can lead to public policies based on false or unreliable research, causing real-world harm. For example, the Wakefield scandal on vaccines and autism has had a corrosive impact on public health. Additionally, it can lead to wasted resources and undermine public trust in academic research.
The 'signing at the top' paper attracted significant attention and implementation because it presented a simple and easily implementable solution to reduce dishonesty. The effect was reported to be large and significant across multiple studies, and it was endorsed by prominent researchers like Dan Ariely and Francesca Gino. Many firms and government agencies, including Lemonade Insurance and the U.S. government, adopted the practice based on these findings.
Bazerman trusted his junior colleagues, particularly Francesca Gino, to handle the data collection and analysis. As a senior researcher, he focused on other aspects such as funding, administrative issues, and mentoring young scholars, which meant he did not closely examine the data himself.
The Data Colada team found the insurance data suspicious because the histogram of miles driven by the study subjects showed a nearly uniform distribution from zero to 50,000 miles, which is not what real data typically look like. Real data usually form a bell curve with most people clustered around the average. The insurance company later confirmed that the data Ariely used were significantly different from what they provided.
Simine Vazire believes psychology is in crisis due to a lack of integrity and credibility. The field is undergoing intense self-examination, and while many researchers are committed to scientific values, the prevalence of unreliable findings and the lack of robust replication efforts have raised concerns about the overall quality of research.
Bazerman and his team failed to replicate the 'signing at the top' effect multiple times, even after adjusting their methodology. When they conducted a large-scale replication using more than 10 times the original sample size, they still found no effect, suggesting the original findings were not robust.
The Data Colada team used a blog to share their findings because it allowed them to communicate more quickly and directly than the traditional journal review process. They wanted to provide short, accessible critiques and updates on their investigations without waiting years for peer review.
Some of the biggest names in behavioral science stand accused of faking their results. Last year, an astonishing 10,000 research papers were retracted. In a series originally published in early 2024, we talk to whistleblowers, reformers, and a co-author who got caught up in the chaos. (Part 1 of 2)
SOURCES:
Max Bazerman), professor of business administration at Harvard Business School.
Leif Nelson), professor of business administration at the University of California, Berkeley Haas School of Business.
Brian Nosek), professor of psychology at the University of Virginia and executive director at the Center for Open Science.
Joseph Simmons), professor of applied statistics and operations, information, and decisions at the Wharton School at the University of Pennsylvania.
Uri Simonsohn), professor of behavioral science at Esade Business School.
Simine Vazire), professor of psychology at the University of Melbourne and editor-in-chief of Psychological Science.
RESOURCES:
"More Than 10,000 Research Papers Were Retracted in 2023 — a New Record)," by Richard Van Noorden (*Nature, *2023).
"Data Falsificada (Part 1): 'Clusterfake),'" by Joseph Simmons, Leif Nelson, and Uri Simonsohn (*Data Colada, *2023).
"Fabricated Data in Research About Honesty. You Can't Make This Stuff Up. Or, Can You?)" by Nick Fountain, Jeff Guo, Keith Romer, and Emma Peaslee (*Planet Money, *2023).
Complicit: How We Enable the Unethical and How to Stop)*, *by Max Bazerman (2022).
"Evidence of Fraud in an Influential Field Experiment About Dishonesty)," by Joseph Simmons, Leif Nelson, and Uri Simonsohn (*Data Colada, *2021).
"False-Positive Psychology: Undisclosed Flexibility in Data Collection and Analysis Allows Presenting Anything as Significant)," by Joseph Simmons, Leif Nelson, and Uri Simonsohn (*Psychological Science, *2011).
EXTRAS:
"Why Do We Cheat, and Why Shouldn’t We?)" by *No Stupid Questions *(2023).
"Is Everybody Cheating These Days?)" by *No Stupid Questions *(2021).