Children as young as 5 may show bias against unfamiliar accents: Canadian study

Bias against unfamiliar accents can start young, according to a Canadian study that found children as young as five years old were more likely to prefer a teacher who had a local accent over a teacher with a regional or different accent.

It was a result that researchers hadn’t expected. Previous studies in the U.S. and France had shown children had accent bias, but researchers believed that Canadian kids who had been exposed to a wider variety of accents would have less bias.

In the study, published earlier this year in the Quarterly Journal of Experimental Psychology, researchers asked 144 children aged five and six to rate who they would like to be their teacher after hearing different speakers with a range of accents read a brief sample text.

The children showed a preference for teachers who spoke with the locally dominant Canadian accent, and rated speakers with French, Australian and Dutch accents lower when asked how good of a teacher they thought each speaker might be.

“The whole thing started with us trying to prove Canadian kids were more accepting than American kids, and it didn’t work,” Elizabeth Johnson, one of the authors of the study, said in an October press release.

“It was something I was really surprised about.”

The children were all from southern Ontario. Even children who had greater exposure to a variety of accents in their home and school life still picked the speaker with a local Canadian accent more times, the study found.

Johnson, a psychology professor with the University of Toronto, Mississauga campus, said that the idea for the study came from reflecting on student evaluations of faculty in higher education.

There are often criticisms leveraged against certain professors in these evaluations that have less to do with their teaching style or their capability and more to do with an implicit bias a student holds, she explained.

“The biases become stronger with age,” Johnson said. “We read [professor] evaluations and they might be obviously gendered in some way, or they might obviously be speaking like a non-native accent.”

Often this bias is ignored, but it can have consequences for the teachers in question, she pointed out.

“We don’t frequently pay attention to these sort of language issues when we look at how students evaluate their professors. And it’s a big deal because we have tons of professors whose first language isn’t English,” Johnson said.

“We wanted to know, where does that come from?”

To investigate how early this bias could emerge, researchers presented children aged five to six with pairs of adult speakers, whom they had to assess.

In each trial, one of the speakers would have the locally dominant Canadian accent, while the other speaker had either a British, Australian, Dutch or non-Canadian French accent.

All speakers read the same short piece in English.

Children were then asked to pick which speaker they’d rather have as their teacher, and also rate “how good of a teacher” they thought each speaker would be.

The children consistently chose the speaker with the local Canadian accent as the one they’d prefer to be their teacher.

They also showed stronger preference for the Canadian accent when the alternative was the French accent compared to the Australian accent, although no difference in preference strength was noted when it was a British versus a Dutch speaker as the other choice.

On average, children chose the speaker with the Canadian accent more than 60 per cent of the time, but when the alternative was a speaker with an accent from France, children chose the Canadian accent speaker more than 80 per cent of the time.

Researchers also recorded how much exposure to different accents each child had in their regular life, but found that there was no difference in how children scored based on their accent exposure.

While the children consistently rated the speaker with the local Canadian accent higher when asked to assess “how good” the speaker would be as a teacher, they didn’t rate any of the speakers as “bad” or “very bad” teachers.

The study also suggests that the preference isn’t purely due to an issue with comprehension — although it did play a factor.

Because the paired speakers were reading the same thing out loud, researchers theorized that if the non-local accent would be preferred more often if they talked second, as the same piece would’ve already been read by the local speaker.

In the experiment, children did show a stronger preference for the local Canadian speaker if the speaker with the non-local accent spoke first, “providing further support for the notion that difficulty in comprehending the non-local speakers may have contributed, at least in part, to children’s judgements in this task,” the study noted.

However, they still preferred the local Canadian accent 66 per cent of the time even when the non-local speaker spoke second and there was no comprehension issue.

Researchers say the consistency with which children chose the Canadian accent over the others shows that students weren’t merely choosing a local accent because they had trouble understanding the other accents.

“There is definitely social bias going on there,” Johnson says. “You really have to worry about this much younger than you thought.”

The study didn’t look at the cause of this, and researchers acknowledged that there could be numerous factors at play when children selected a preferred teacher.

“Given the binary nature of selection data, we acknowledge that this type of analysis may oversimplify the rich social decisions that children are making while evaluating others,” the study noted.

Researchers also pointed out that children did not rate any of the speakers poorly on their competence as a teacher, and that the children were hearing these speakers for the first time and may have had different assessments if they spent more time with the people behind the voices.

Johnson said they hope to research further to unpack some of these potential confounders.

“We’re also interested in better understanding what types of experiences might mitigate the initial formation or maintenance of negative linguistic biases in young children,” she said.

Determining the source of accent bias, she added, is “important for society and the way we function, the way we make decisions about people’s competency, and the way we choose to portray people in the media.”