But here it is, courtesy of bustle.com: "A new study by Nextions discovered that proofreading is not the objective task we may have thought it was. When partners in a law firm were given a research memo to evaluate, the lawyers who thought the memo was written by a Caucasian male found fewer mistakes and gave the memo higher ratings than the lawyers who thought the exact same memo was written by an African-American male.
'Tomas Meyer,' the imaginary, race-shifting author of the memo, made a total of 22 errors: 7 concerning spelling or grammar, 6 that were “substantive writing” errors, 5 factual errors, and 4 analytic errors. What’s extra creepy about the study is that even the obvious, objective errors like spelling mistakes fell victim to the racial bias of the readers. When Tomas Meyer was Caucasian, readers found an average of 2.9 spelling/grammar errors (OUT OF 7!) in his work. When Tomas Meyer was African American, they found an average of 5.8 spelling/grammar errors. So not only does this show us that lawyers can’t proofread, it means that when we believe an author is white, we may actually fail to see the objective mistakes that they’ve made.
Tomas Meyer also received much crueler comments on his work when he was African American. While Caucasian Tomas Meyer 'has potential,' lawyers 'can’t believe' that African American Tomas Meyer 'went to NYU.'
Perhaps surprisingly, the readers’ race didn’t appear to affect the outcome of their proofreading bias. According to Nextions, 'There was no significant correlation between a partner’s race/ethnicity and the differentiated patterns of errors found between the two memos.' They did find, however, that female partners spotted more of the errors than the men did. Looks like the unbiased, objective art of proofreading could use some editing itself."