Photo Credit: Getty Images
They also aren't particularly reliable or trustworthy, which means surveys are problematic. In the two surveys compared for this Criminology study, however, respondents were asked to estimate earnings twice using different measures of time (days, weeks, or months), helping to weed out faulty or exaggerated answers. "I was pretty surprised at how consistent the findings were across the two datasets," researcher Holly Nguyen of Penn State University tells Quartz.
Read the full story on Newser.com
More From Newser