A study led by the University of California, Berkeley (UC Berkeley) suggests big-data analysis may yield inaccurate representations about people.
“If you want to know what individuals feel or how they become sick, you have to conduct research on individuals, not on groups,” contends UC Berkeley’s Aaron Fisher.
Joint research by UC Berkeley, Drexel University, and the University of Groningen in the Netherlands used statistical models to compare data on hundreds of people, including healthy persons and those with various disorders. Their analysis of data via online and smartphone self-report polls and heart-rate measurements consistently determined that group conclusions may not necessarily hold true for individuals.
In one example, group analysis of people with depression revealed that they worry a lot, but individual analysis found wide variations ranging from no worrying to agonizing far above the group average.
From Berkeley News
View Full Article
Abstracts Copyright © 2018 Information Inc., Bethesda, Maryland, USA