Star ratings can mislead shoppers and impact sales, UBC study shows
Consumers are more likely to round up star ratings and round down numerical ones, researchers find
If you're deciding between two products, both rated a 3.5, but one shows the score with stars while the other uses plain numbers, odds are you'll pick the one with stars.
That's the key finding from a recent study by the University of B.C. Sauder School of Business, which reveals that consumers consistently overestimate ratings shown as stars and underestimate the same values when shown as numbers.
"Whenever I see a star rating, especially if it's a fractional one, which leads to incomplete star pictures like 3.5 or 3.7, I take a moment to ask myself, am I being fooled?" said Deepak Sirwani, an assistant professor at UBC and co-author of the study.
When a product rating is shown using stars, a score like 3.5 is typically displayed as three full stars followed by a half-filled fourth star.
Visually, this creates the impression of an incomplete image, says Sirwani.
"So we tend to round up and complete the incomplete picture with the visual ratings," he explained during an interview with CBC's On The Coast.
In contrast, numbers prompt consumers to fixate on the first digit — a phenomenon known as the "left-digit effect."
"We read from left to right, especially in Western civilization and a lot of other civilizations, so we focus more on three in the number 3.5 … and round down."
That gap in perception, according to the research, can have a massive effect.
Impact on sales
"When you show a product using a star rating, your sales might triple compared to the same rating shown using numbers," Sirwani said.
But the study also warns that star ratings might lead to inflated expectations — and eventually, disappointed customers.
"[Consumers] might feel they were over promised and under delivered so that may increase negative reviews," the study's co-author said.
Researchers recommend using "visually complete stars" — ones that only show a star's filled-in part, not its overall border — to minimize bias.
Vancouver-based retail strategist David Ian Gray called the findings "clever" and "unique."
"I don't think it's anything we're consciously thinking about day-to-day," he told CBC News. "But reviews are incredibly influential on consumer behaviour, spending choices."
Gray says brands hosting their own internal reviews might consider switching formats — though he cautions that many consumers look for third-party ratings to avoid bias.
"There are a lot of websites with dubious reviews, possibly paid for," he said. "Consumers are looking for objectivity."
Gray says studies such as UBC's can make people more alert.
"I think consumers, when they hear about this, are going to be a little bit more conscious of how they're intuitively reacting, and maybe double-check their thinking."
The researchers note the study's implications stretch far beyond product reviews, stating that similar visual biases may impact how people interpret battery life on a phone, fuel gauges, fitness trackers, and safety ratings on vehicles, among other things.
The study is calling for industry standards and regulatory guidelines, not only to level the playing field between retailers, but to ensure consumers aren't being misled — even accidentally.
"Whether it's pictorial formats or numbers, ratings matter, and platforms should probably think about it more carefully than they have," said Sirwani.
With files from On The Coast and Shaurya Kshatri