Results: attitudes at work
Does accent bias interfere with professional judgements of expertise?
Our previous research found systematic patterns of bias against certain accents among the UK public. Would professionals from an elite occupation like law exhibit the same biases? This study showed that when legal professionals are asked to assess candidates in a controlled setting and pay attention to the quality of responses, they are able to switch off biases and focus on the content of what a candidate is saying.
Overview
The Social Mobility Commission report suggested a presence of some forms of accent bias in the legal sector – and our research with accent labels and recorded voices found systematic patterns of bias among the general public against certain accents in England. Can such biases affect a person’s success in a job interview, and act as a potential obstacle to social mobility and success in life for certain groups?
61 lawyers and graduate recruiters in leading UK-based international law firms completed our mock hiring exercise. We examined whether ‘good’ and ‘poor’ quality mock interview answers were clearly identified as such regardless of accent, or whether the accent these answers were spoken in affected the rating of their objective quality.
Our results for this aspect of trainee hiring show that lawyers in fact have a clear ability to suppress accent biases and focus on the quality of responses.
Ignoring accent and focusing on expertise
When assessing expert knowledge, lawyers did not show significant preferences for Received Pronunciation (RP) or General Northern English (GNE), nor did they show a consistent dispreference for working class or non-white accents.
Instead, they showed a consistent ability to rate responses according to whether the response was higher or lower quality, regardless of accent (see Fig. 1). Their responses very closely matched the ratings that a pre-test group of legal professionals gave when they saw only the written version of the mock responses.
Thus, at least in a test setting, legal professionals may have been inhibiting their routine accent biases in order to attend to the content of responses.
It is worth noting that the two accents most downgraded in the Nationwide Survey – the two working class accents from the South East, Estuary English (EE) and Multicultural London English (MLE) – received by far the lowest individual ratings when giving high quality responses. This is not the majority pattern, but it may indicate the presence of pockets of negative bias against these voices.
Fig. 1 - Ratings of high and low quality answers
The distribution of evaluations for low and high quality answers by legal professionals.
A higher bar for Received Pronunciation speakers?
Although there is a general absence of biased responses, we do in fact see a subtle accent effect, especially when assessing personal social attractiveness or likeability. Interestingly, this in the reverse direction to the traditional status of Received Pronunciation (RP): Lawyers reported lower likability ratings for RP across the board.
The slight downgrading of RP responses may be due to an implicitly higher expectation for more educated respondents, which still can point to differential expectations of knowledge and ability.
In relation to likeability ratings, the marginally lower rating of RP may also indicate high status but low solidarity associations. For example, one of our preliminary interviewees noted “I might be prejudiced if someone spoke with a certain kind of extremely posh, aristocratic voice.”
Lawyers not influenced by other factors
Several factors that had a strong effect among the general public (in our surveys of attitudes to accent labels and recorded voices) did not exert an effect on lawyers.
The age and regional origin of legal professionals did not affect how they responded to job candidates, unlike what we found among the general public. The number of years an individual had worked in the legal sector also had no discernible effect.
Motivation to Control a Prejudiced Response (MCPR) – a psychological factor that had a strong effect on how listeners behaved in our nationwide survey – also did not significantly influence professionals when they were rating a candidate’s competence and expertise. However it did affect their ratings of personal social attractiveness.
Listeners who were concerned that they should not seem prejudiced gave higher overall ratings for personal likeability in particular. But this does not influence them as much when they were judging professional competence.
Together these findings further support the conclusion that legal professionals were paying close attention to what job candidates said, not how they were saying it, reflecting the judicious exercise of professional judgement when conscious of a need for fairness.
Conclusion
Our study shows that when asked to assess candidates in a controlled setting, with the ability to listen carefully to the content of responses and perhaps with heightened awareness of a need to be fair, legal professionals have the ability to switch off biases and attend very well to the quality of content.
However, the present study simulates just one small part of trainee hiring, focusing exclusively on the recognition of competence in responses. It does not provide information on the potential for accent bias in other aspects of professional life, e.g. informal interaction during the interview, post-hiring experiences of candidates from non-traditional backgrounds, or progression through more senior career stages.
We therefore cannot conclude that RP speakers would not ultimately still have a better chance at being hired. Recall that the 2015 Social Mobility Commission report points to many other aspects of ‘cultural fit’ in hiring decisions, and our surveys with the general public confirm the continued elevated status of RP.
In continuing work, we are now examining whether our sample professionals’ ability to avoid bias develops with experience, and also whether training interventions such as reminding recruiters not to let personal accent preferences drive judgements of professional potential have any effect