According to a just released study, law partners graded the same writing sample based at least in part on the race of the writer. The study published by the consulting firm Nextions, “Written in Black and White: Exploring Confirmation Bias in Racialized Perception of Writing Skills,” used perhaps a somewhat flawed methodology. and does not purport to have statistical significance. On its face, though, the study does seem to confirm “confirmation bias,” which it describes as “[a] mental shortcut – a bias – engaged by the brain that makes one actively seek information, interpretation and memory to only observe and absorb that which affirms established beliefs while missing data that contradicts established beliefs.”
Five partners from five different law firms wrote a legal research paper supposedly prepared by two, third year litigation department associates. To it were added spelling or grammar mistakes and substantive technical writing errors. The memo was distributed to sixty law firm partners at twenty-two different law firms who had been told they were part of a writing analysis study. Twenty-one were members of a racial/ethnic minority and thirty-nine were Caucasian. Everyone received the exact same memorandum to analyze
This is where it gets interesting. Fifty percent of the partners received a memo written by a “Thomas Meyer,” a white graduate of New York University. The other half were informed that the author was “Thomas Meyer,” a black graduate from the same school. The reviewers also received identical research material supporting the memorandum.
The black Thomas Meyer averaged a 3.2/5.0 rating whereas the white Thomas Meyer averaged a 4.1/5.0 rating. Reviewer comments were consistently more positive for the white Thomas Meyer. For example, reviewers of the supposedly white=written memorandum made comments like “generally good writer,” “has potential” and “good analytic skills.” Comments for the black Mr. Meyer included “needs lots of work,” “average at best” and “can’t believe he went to NYU.”
Sure, the methodology of the study can be criticized, but the results are reality based. Racism exists. The results might not confirm outright racial animus. They do highlight confirmation bias in writing skills; in other words, a bias that makes one look for information that affirms established beliefs while glossing over data that contradicts them — confirmation bias.