Online Program Home
My Program

Abstract Details

Activity Number: 22 - The World of Data Analysis Professionals
Type: Topic Contributed
Date/Time: Sunday, July 29, 2018 : 2:00 PM to 3:50 PM
Sponsor: Section for Statistical Programmers and Analysts
Abstract #329035 Presentation
Title: Statistics at Consumer Reports
Author(s): Michael Saccucci*
Companies: Consumer Reports
Keywords: Consumer Reports; Product ratings; Text mining; Machine learning; Experimental Design

Consumer Reports (CR) is an Independent, Unbiased, Nonprofit Testing and Information Organization that has been rating products and services since 1936. CR purchases all products we test and receives no special treatment from manufacturers. Our model has historically been to test a representative group of available models and make recommendations on which products and services offer the best value. Over the past 20 years, the statistics group has worked together with testers to increase the number of tested models. As a result, we have seen a significant increase in both the number of tested models and a reduction in the time to publish. However, as search engines have improved and competition has increased, consumers no longer just want our recommendations. Today's savvy shopper wants information about models they are interested in purchasing. This requires the use of text mining and machine learning algorithms to augment our traditional testing model. This presentation will focus on how the role of statistics has changed over the years to meet this new testing paradigm.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program