Name-Brand vs. Off-Brand: A Twist on Taste Testing for a Mathematical Statistics Course

• Activity Number: 342
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistical Education
• Author(s): Eric Reyes*+
• Keywords: Mathematical Statistics ; Maximum Likelihood ; Forced Binary Choice

Abstract

The Mathematical Statistics course caters to students with a strong mathematics background including calculus, probability, some programming, and often an introductory statistics course. We present an in-depth activity that asks students to make inference on the probability that an individual can actually discern a name-brand product from its off-brand counterpart in a blind taste test. The twist: students are challenged to account for the fact that a correct response from a participant may be the result of a random guess, as opposed to a true discernment. The activity uses the taste test as the setting for bringing together several concepts taught throughout the course --- probabilistic modeling, conditional distributions, maximum likelihood estimation and inference --- as well as computational methods for implementation. We also discuss how the activity can be implemented to provide students with a miniature research experience.