Marc R. Dotson p 540 Fisher Hall, 2100 Neil Avenue Columbus, OH 43210 Cell Phone: (614) 935-7506 E-Mail: dotson.126@osu.edu Education Ph.D., The Ohio State University, 2016 (Expected) Marketing M.Sc. with Merit, The London School of Economics & Political Science, 2009 International Political Economy B.S. Summa Cum Laude, Southern Utah University, 2008 Political Science with Minors in Mathematics and Journalism Working Papers “A Hierarchical Bayesian Choice Model with Latent Dirichlet Allocation.” Job Market Paper. Abstract: Choice models produce part-worth estimates that tell us what individuals prefer but not why. To understand the why of preference we need to model consumer heterogeneity by specifying covariates that will explain cross-sectional variation in the part-worths. In this paper we demonstrate a way to estimate covariates for the upper level of a hierarchical Bayesian choice model that lead to an improvement in explaining part-worth heterogeneity. The covariates are estimated using latent Dirichlet allocation, a model for high-level interactions among discrete data. We find improvement in model fit using the covariates estimated with LDA over competing models using binary covariates. This provides an important step toward a continued synthesis between text analysis and choice modeling. “An Extended Model of Brand Choice: Incorporating Product Relevance and Perceived Efficacy,” with Greg M. Allenby. Abstract: Choice is the end result of a complex process with consumers proceeding through a variety of stages prior to purchase. Extended models of behavior take this staged process into account, but have not been frequently employed in the choice modeling literature. In this paper, we develop an extended model of behavior for brand choice and measure the effect of product relevance. A product is relevant if a respondent perceives it as being able to address his or her needs, where needs describe the motivating state of the individual. We find evidence that the effect of product relevance is manifest through preference certainty rather than its level. Implications for measuring consumer preferences are explored. Presentations “A Hierarchical Bayesian Choice Model with Latent Dirichlet Allocation” – Department of Marketing and Logistics at The Ohio State University, July 2015 “An Extended Model of Brand Choice: Incorporating Product Relevance and Perceived Efficacy” – Sawtooth Software Conference, March 2015 – Department of Marketing and Logistics at The Ohio State University, August 2014 – Haring Doctoral Symposium at Indiana University, April 2014 – Statistics, Psychology, and Marketing Seminar Series at The Ohio State University, March 2014 Teaching Marketing Research (Undergraduate) at The Ohio State University, 2014 (4.6 out of 5.0) Teaching Assistant, Bayesian Statistics and Marketing Seminar at The Ohio State University, 2014 Marketing Research (Undergraduate) at The Ohio State University, 2013 (3.5 out of 5.0) Industry Experience Statistical Analyst The Modellers Research Associate The Modellers Scholarships & Awards Mittelstaedt Doctoral Symposium Fellow, 2015 Haring Doctoral Symposium Fellow, 2014 Mittelstaedt Doctoral Symposium Fellow, 2013 Alpha Chi National Honor Society Robert W. Sledge Fellowship, 2008-2009 Professional Affiliations American Marketing Association INFORMS References Greg M. Allenby Helen Kurtz Chair of Marketing and Professor of Marketing and Statistics The Ohio State University Phone: (614) 292-9452 E-Mail: allenby.1@osu.edu Nino Hardt Professor of Marketing The Ohio State University Phone: (614) 688-2996 E-Mail: hardt.8@osu.edu Mingyu (Max) Joo Professor of Marketing The Ohio State University Phone: (614) 247-8845 E-Mail: joo.85@osu.edu August 2010-July 2012 Salt Lake City, UT January-August 2010 Salt Lake City, UT