3 Day Doctoral Seminar with Prof. Tom Meyvis (New York University, USA)
Day 1: Responding to the replication crisis in behavioral science by running more powerful studies
Day 2: Moderation, mediation, and covariates: Manipulating, measuring, and controlling third variables to establish causal mechanisms.
Day 3: Affective forecasting errors: Our surprising inability to predict how our feelings
This seminar will first address the nature and causes of the current replication crisis in behavioral science, as well as the most popular proposed solutions for the problem.
I will discuss the limitations of these solutions and propose augmenting these with strategies to conduct more powerful experiments.
Most of the seminar will focus on practical guidelines for increasing the power of experiments (reducing type II error) in ways that do not inflate the risk of false positives (type I error) - and also effectively communicate that to readers and reviewers.
This seminar will explore the different ways in which you can design experiments to fulfill their basic function: establishing whether and how a factor causes an effect.
We will cover the basics of avoiding confounds, testing for causal mechanisms through moderation, mediation, and mediated moderation, and controlling for noise using ANCOVA.
We will discuss how to execute these analyses, but we will also discuss the strengths, weaknesses, and conditions for each approach, as well as when they are meaningful vs. meaningless.
One of our key concerns when making choices (as consumers, investors, voters, and anyone pursuing happiness) should be how the outcomes will make us feel.
If I go on vacation to Spain (vs. Paris), buy a big-screen TV (vs. a smaller one), buy a volatile stock (vs. invest in a mutual fund), vote for a candidate who ends up winning (vs. losing) the election, spend more time acquiring goods (vs. experiences), how will that affect my happiness, satisfaction, regret?
Even though these affective predictions are obviously important, and even though we have plenty of experience observing outcomes of similar decisions we’ve made in the past, we are remarkable bad at it.
I will discuss different streams of research that document and explain these errors, as well as the consequences of these errors (for consumers and managers), and the reasons for their persistence.
Tom Meyvis is a Professor of Marketing at the Stern School of Business, New York University.
He received his Ph.D. in Marketing from the University of Florida and his Licentiaat in Experimental Psychology from the Catholic University of Leuven, Belgium.
His research and teaching interests focus on consumer behavior, specifically consumer decision making and strategies to increase consumer enjoyment. His research has been published in leading journals, including the Journal of Consumer Research, Journal of Marketing Research, Psychological Science, and Organizational Behavior and Human Decision Processes.
Previous research projects have studied branding issues (such as consumers' evaluations of brand extensions and the effect of irrelevant product information), pricing issues (such as the biased processing of store price comparisons and the role of anticipated regret in purchase timing decisions), and the marketing of hedonic experiences (such as the effect of commercial interruptions on the enjoyment of television shows).
Current research projects focus on issues related to consumers' enjoyment: how consumers predict their enjoyment (and why they are often not very good at this) and how consumers can structure their experiences to maximize their enjoyment.
Branding, Consumer Decision Making, Managing Consumer Experiences, Pricing