Buying stuff online? Don't always rely on ratings
For the study, researchers investigated user ratings for over 1,200 products across 120 product categories, from blood pressure monitors and bike helmets to air filters. Their analyses reveal a very low correlation between average user ratings of items on Amazon.com and product ratings, based on objective tests, found in consumer reports.
Further, the likelihood that an item with a higher user rating performs objectively better than an item with a lower user rating is only 57 percent, which casts some doubt on the validity of user ratings.
The study also examined what information consumers rely on when judging the quality of products on Amazon.com. Consumers depend heavily on the average user rating, regardless of whether the average rating is based on a small or a large sample of consumers.
But this is wrong-headed in that there are not enough ratings for a product or there is wide disagreement among reviewers. In this case, consumers should not trust the average very much, but they do nevertheless.
The authors recommend consumers be more cautious when they make inferences about product quality based on user ratings.
I always take online ratings with a grain of salt, especially when there are few of them for a given product. I always encourage my friends to compare online ratings on at least two websites when contemplating which product to buy. For example, if you're in the market for a new laptop, you may want to eyeball the user ratings on Amazon as well as CNET. That may help lessen the chances of making a misguided purchase decision.