The Love of Large Numbers: A Popularity Bias in Consumer Choice

Date
Fri February 3rd 2017, 3:15 - 4:30pm
Event Sponsor
Department of Psychology
Location
Jordan Hall (Building 420), Room 050

Abstract: Social learning—the ability to learn from observing the decisions of others and their outcomes—is fundamental to human evolutionary and cultural success. The Internet now provides social evidence on an unprecedented scale. However, properly utilizing this evidence requires a capacity for statistical inference, without which simplistic “herd behavior” may lead to suboptimal decisions. I'll discuss a recent project in which my collaborators and I examined how people interpret online review scores with different numbers of reviews—a potential indicator both of an item’s popularity and of the precision of the average review score. Our task was designed to pit statistical inference against a simple decision bias for popular items. We modeled optimal statistical inference using empirical prior information from millions of reviews posted on Amazon.com. Under certain conditions, people preferred a product with more reviews over one with fewer reviews even though the statistical model indicated that the latter was likely to be of higher quality than the former.