Almost all retailers provide user ratings on their websites, the researchers said.
For the study, researchers examined user ratings for 1,272 products across 120 product categories, such as car seats, bike helmets, sunblock, air filters, smoke alarms and blood pressure monitors.
The analyses show a very low correspondence between average user ratings of products on an e-commerce portal and product ratings based on objective tests found in consumer reports.
"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," said Bart de Langhe, professor at the University of Colorado Boulder in US.
"A correspondence of 50 per cent would be random, so user ratings provide very little insight about objective product performance," de Langhe said.
The study also found that the user ratings do not predict the resale value of used products.
"Products with better reliability and performance retain more of their value over time," said de Langhe.
"If average user ratings reflect objective quality, they should correlate positively with resale values. The fact that they don't casts more doubt on the validity of user ratings," he said.
Researchers also examined what information consumers rely on when judging the quality of products on the e-commerce website.
They found that consumers rely very heavily on the average user rating, which is presented front and centre.
They do this regardless of whether the average rating is based on a small or a large sample of consumers.
"This is a mistake. Oftentimes, there are just not enough ratings for a product or there is too much disagreement among reviewers," said de Langhe.
"In this case, consumers should not trust the average very much, but they do nonetheless," he said.
The study was published in the Journal of Consumer Research.