J-Shaped Review Distribution

J-Shaped Review Distribution

Reviewed by Faisal Zaidi

The pattern in which most products receive high ratings, a smaller group receives the lowest, and few fall in between.

What it is

Nan Hu, Jie Zhang and Paul Pavlou discussed it in 2009 using data on books, DVDs and videos collected in 2005 from two large online retailers. On this site it is read alongside Choice Overload.

In effect

Between 72 and 78 per cent of the products in their data had ratings of four stars or more, and the skew persisted with over 2,000 reviews per product. The authors argued that the average rating is a poor proxy for quality and sales and traced the pattern to biases in review systems.

What it does not say

It is not the same thing as Choice Overload, which sits close to it on this site but describes a different part of how people decide. It also does not say that the finding reaches every product, buyer or setting; where this page goes beyond the cited research, the step is an inference and is labelled as one.

Sources

  1. Nan Hu is first author, with Jie Zhang and Paul Pavlou, of Overcoming the J-shaped distribution of product reviews, published in Communications of the ACM in 2009.
  2. Jie Zhang is co-author, with Nan Hu and Paul Pavlou, of Overcoming the J-shaped distribution of product reviews, published in Communications of the ACM in 2009.
  3. Paul A. Pavlou is co-author, with Nan Hu and Jie Zhang, of Overcoming the J-shaped distribution of product reviews, published in Communications of the ACM in 2009.