Honesty study was based on fabricated data
The study was originally published in 2012 and had five authors. The basic thrust of it was this: If people are asked to sign “an honesty declaration” at the top of a form instead of at the bottom, they are more likely to answer honestly. As Buzzfeed points out, a lot of organizations including the US government took the findings very seriously
https://hotair.com/john-s-2/2021/08/20/a-landmark-study-about-honesty-was-based-on-fabricated-data-n410485
the analysis of the fraud is here (in this blog post).
https://datacolada.org/98
A team of anonymous researchers downloaded it, and discovered that this field experiment suffers from a much bigger problem than a randomization failure: There is very strong evidence that the data were fabricated.
We’ll walk you through the evidence that we and these anonymous researchers uncovered, which comes in the form of four anomalies contained within the posted data file. The original data, as well as all of our data and code, are available on ResearchBox
Here's part of their damning conclusion:
The evidence presented in this post indicates that the data underwent at least two forms of fabrication: (1) many Time 1 data points were duplicated and then slightly altered (using a random number generator) to create additional observations, and (2) all of the Time 2 data were created using a random number generator that capped miles driven, the key dependent variable, at 50,000 miles.
A single fraudulent dataset almost never provides enough evidence to answer all relevant questions about how that fraud was committed. And this dataset is no exception. First, it is impossible to tell from the data who fabricated it. But because the fourth author has made it clear to us that he was the only author in touch with the insurance company, there are three logical possibilities: the fourth author himself, someone in the fourth author's lab, or someone at the insurance company. This footnote contains some supporting evidence: [14]. Second, we do not yet know exactly how the data were tampered with in order to produce the condition differences. Were the condition labels generated or altered after the mileage data were created? And we also don’t know the answer to other relevant questions, such as why the Calibri data were duplicated, or why the fabricator(s) generated condition differences at Time 1 [15]. Of course, we don’t need to know the answer to these questions to know that the data were fabricated. We know, beyond any shadow of a doubt, that they were.
= new reply since forum marked as read
