Content and Context: Identifying the Impact of Qualitative Information on Consumer Choice

Managers and researchers alike suspect that the vast amounts of qualitative information in blogs, reviews, news stories, and experts’ advice influence consumer behavior. But, does qualitative information impact or rather reflect consumer choices? We argue that because message content and consumer choice are endogenous, nonrandom selection and conflation of awareness and persuasion complicate causal estimation of the impact of message content on outcomes. We apply Latent Dirichlet Allocation to characterize the topics of transcribed content from 2,397 stock recommendations provided by Jim Cramer on his show Mad Money. We demonstrate that selection bias and audience prior awareness create measurable biases in estimates of the impact of content on stock prices. Comparing recommendation content to prior news, we show that he is less persuasive when he uses more novel arguments. The technique we develop can be applied in a variety of settings where marketers can present different messages depending on what consumers know.

[1]  A. Lo,et al.  Stock Market Prices Do Not Follow Random Walks: Evidence from a Simple Specification Test , 1987 .

[2]  Panagiotis G. Ipeirotis,et al.  The Dimensions of Reputation in Electronic Markets , 2009 .

[3]  Werner Antweiler,et al.  Do Us Stock Markets Typically Overreact to Corporate News Stories? , 2006 .

[4]  Michael I. Jordan,et al.  Latent Dirichlet Allocation , 2001, J. Mach. Learn. Res..

[5]  Mike Y. Chen,et al.  Yahoo! for Amazon: Sentiment Extraction from Small Talk on the Web , 2001 .

[6]  Yu Jeffrey Hu,et al.  From Niches to Riches: Anatomy of the Long Tail , 2006 .

[7]  David M. Blei,et al.  Supervised Topic Models , 2007, NIPS.

[8]  William L. Moore,et al.  Impact of Mad Money Stock Recommendations: Merging Financial and Marketing Perspectives , 2009 .

[9]  Francis R. Bach,et al.  Online Learning for Latent Dirichlet Allocation , 2010, NIPS.

[10]  A. Pentland,et al.  Computational Social Science , 2009, Science.

[11]  David M. Pennock,et al.  Predicting consumer behavior with Web search , 2010, Proceedings of the National Academy of Sciences.

[12]  Jonah Berger,et al.  Positive Effects of Negative Publicity: When Negative Reviews Increase Sales , 2009, Mark. Sci..

[13]  H. Varian,et al.  Predicting the Present with Google Trends , 2009 .

[14]  Brendan T. O'Connor,et al.  From Tweets to Polls: Linking Text Sentiment to Public Opinion Time Series , 2010, ICWSM.

[15]  Guido W. Imbens,et al.  An efficient method of moments estimator for discrete choice models with choice-based sampling , 1992 .

[16]  Paul C. Tetlock.  Giving Content to Investor Sentiment: The Role of Media in the Stock Market , 2005, The Journal of Finance.

[17]  J. Stein,et al.  A Unified Theory of Underreaction, Momentum Trading and Overreaction in Asset Markets , 1997 .

[18]  John J. Neumann,et al.  Does Mad Money make the market go mad , 2007 .

[19]  Marshall Van Alstyne,et al.  Networks, Information and Brokerage: The Diversity-Bandwidth Tradeoff , 2010 .

[20]  Joseph Engelberg,et al.  Market Madness? The Case of Mad Money , 2010, Manag. Sci..

[21]  Jonah A. Berger,et al.  Positive Effects of Negative Publicity , 2010 .

[22]  S. Cosslett,et al.  Maximum likelihood estimator for choice-based samples , 1981 .

[23]  Gary King,et al.  Logistic Regression in Rare Events Data , 2001, Political Analysis.

[24]  Alexander Galetovic,et al.  Information or Opinion? Media Bias as Product Differentiation , 2007 .

[25]  Joel Peress,et al.  Media Coverage and the Cross-Section of Stock Returns , 2008 .

[26]  Jesse M. Shapiro,et al.  Media Bias and Reputation , 2005, Journal of Political Economy.

[27]  Bryan Lim,et al.  The Performance and Impact of Stock Picks Mentioned on 'Mad Money' , 2008 .

[28]  Chong Wang,et al.  Reading Tea Leaves: How Humans Interpret Topic Models , 2009, NIPS.

[29]  F. Vella.  Estimating Models with Sample Selection Bias: A Survey , 1998 .

[30]  John G. Lynch,et al.  Toward a Reconciliation of Market Power and Information Theories of Advertising Effects on Price Elasticity , 1995 .

[31]  Panagiotis G. Ipeirotis,et al.  Show me the money!: deriving the pricing power of product features by mining consumer reviews , 2007, KDD '07.

[32]  J. Heckman.  Sample selection bias as a specification error , 1979 .

[33]  E. Fama,et al.  Common risk factors in the returns on stocks and bonds , 1993 .

[34]  R. C. Merton,et al.  Presidential Address: A simple model of capital market equilibrium with incomplete information , 1987 .

[35]  A. Craig MacKinlay,et al.  Stock Market Prices Do Not Follow Random Walks: Evidence from a Simple Specification Test , 1988 .

[36]  Zhi Da,et al.  In Search of Attention , 2009 .

[37]  Noah A. Smith,et al.  Predicting Response to Political Blog Posts with Topic Models , 2009, NAACL.

[38]  Claudio Cioffi-Revilla,et al.  Computational social science , 2010 .

[39]  Erik Brynjolfsson,et al.  The Future of Prediction: How Google Searches Foreshadow Housing Prices and Quantities , 2009, ICIS.

[40]  G IpeirotisPanagiotis,et al.  Deriving the Pricing Power of Product Features by Mining Consumer Reviews , 2011 .

[41]  E. Brynjolfsson,et al.  The Future of Prediction: How Google Searches Foreshadow Housing Prices and Sales , 2013, ICIS 2013.