Decision contamination in the wild: Sequential dependencies in Yelp review ratings

Abstract

Current judgments are systematically biased by prior judgments. Such biases occur in ways that seem to reflect the cognitive system’s ability to adapt to the statistical regularities within the environment. These cognitive sequential dependencies have been shown to occur under carefully controlled laboratory settings as well as more recent studies designed to determine if such effects occur in real world scenarios. In this study we use these well-known findings to guide our analysis of over 2.2 million business review ratings. We explore how both within-reviewer and within-business (between reviewer) ratings are influenced by previous ratings. Our findings, albeit exploratory, suggest that current ratings are influenced in systematic ways by prior ratings. This work is couched within a broader program that aims to determine the validity of laboratory findings using large naturally occurring behavioral data.


Back to Table of Contents