Executive Summary
The research examines the ability of graphical presentation, experience, and training to reduce biases in data-based decisions. Managerial decision making relies on two fundamentally different types of information: belief-based and data-based information. Belief-based information is the set of beliefs that managers hold regarding what is generally true about the markets in which they compete. These beliefs stem from many sources, including formal training, personal experience, and the high-level strategies formulated by company executives and industry analysts. Such information contrasts with the data-based information that routinely flows through most organizations (e.g., sales and profit figures, budgets and forecasts, various technical and market research results). Some data may conflict with managerial beliefs, leading to either an appropriate updating of beliefs or a biased interpretation of the data in the direction of prior beliefs. Moreover, data frequently are processed only heuristically, and prior research has shown that decisions based on numerical data can be strongly biased by the cognitive heuristics used for the analysis.
The central role of numerical data in many managerial decisions prompts an important question about the extent to which biases can be reduced. Two sources of bias reduction are explicit training and real-world experience. In addition, many researchers have argued that presenting data using appropriate graphical formats can enhance data comprehension, which presumably helps reduce bias. This research examines the effects of graphical presentation, experience, and training on decision biases in data-based decisions.
Specifically, the authors investigate how data about prior advertising expenditures and sales outcomes function in budget allocation decisions, in an attempt to answer three important questions about data-based inferences. First, do biases exist that are strong enough to lead to seriously suboptimal decisions? Second, do graphical data displays, real-world experience, or explicit training reduce any observed biases? Third, are the observed biases well explained by a relatively small set of natural heuristics that managers use when making data-based allocation decisions? The results of the reported experiments lead the authors to conclude yes, no, and yes, respectively. They identify three broad classes of heuristics: difference-based heuristics (which assess causation by comparing changes in expenditures to changes in sales), trend-based heuristics (which assess causation by comparing overall trends in expenditures and sales), and exemplar-based heuristics (which emulate the allocation pattern of the observations with the highest sales). All three heuristics create biases in some situations. Overall, managers used exemplar-based heuristics most frequently, even though they are the most biasing of the three (sometimes allocating the most resources to an advertising medium uncorrelated with sales). Difference-based heuristics appeared less frequently but generated the most extreme allocations. Trend-based heuristics were the least frequently used. The observed biases appear to reflect visual perceptions; therefore, the authors offer some corresponding practical remedies.
Biography
J. Wesley Hutchinson is Stephen J. Heyman Professor and Professor of Marketing at the Wharton School of the University of Pennsylvania. He has degrees in Psychology from Stanford (PhD) and Duke (BS). His research interests focus on consumer and managerial decision making—in particular, the interrelationships among attention, learning, confidence, decision making, and expertise in repeated choice environments. He publishes articles in a wide variety of journals in business and psychology and teaches courses in new product development, the social impact of marketing, and research methods.
Joseph W. Alba is a Distinguished Professor of Marketing at the University of Florida. His research focuses on consumer knowledge, decision making, brand equity, and pricing. Joe has published in leading journals in marketing and psychology, including Journal of Marketing Research, Journal of Consumer Research, Journal of Marketing, and Journal of Experimental Psychology. His research has won numerous awards, and he has served as president of the Association for Consumer Research. He received his PhD from Temple University.
Eric M. Eisenstein is Assistant Professor of Marketing in the Fox School of Business at Temple University. His research focuses on high-cognition decision making and its psychological underpinnings, learning and the development of expertise, and debiasing and improving consequential decisions. Eric has published in a variety of journals in business. Before becoming a professor, he worked for four years at Mercer Management Consulting (now Oliver Wyman), where he focused on the management of technology and consumer research in the financial services and telecommunications industries.
Journal of Marketing Research, Volume 47, Number 4, August 2010
View Table of Contents