But Don’t Worry, Racism No Longer Exists

A new study confirms how much race still affects even those who undoubtedly do not view themselves as racist and are not viewed that way by others (which means most of us, of course). It was a rather clever study where they submitted the same writing sample to 60 partners in law firms, with half of them being told that the applicant was white and half that he was black. The predictable results:

In Written in Black and White, selected law firm partners were asked to evaluate a single research memo into which 22 different errors were deliberately inserted – 7 spelling/grammar errors, 6 substantive writing errors, 5 errors in fact, and 4 analytic errors. Half of the partner evaluators were told that the hypothetical associate author was African American and half were told that the author was Caucasian.

Sadly, you know what’s coming.

On a five point scale, reviews for the exact same memo averaged a 3.2 for the “African American” author and 4.1 for the “Caucasian” author. More surprising were the findings of “objective” criteria such as spelling. The partner evaluators found an average of 2.9 spelling and grammar errors for the “Caucasian” authors and 5.8 such errors for the “African American” authors. Overall the memo presumed to have been written by a “Caucasian” was “evaluated to be better in regards to the analysis of facts and had substantively fewer critical comments.”

The comments from the partners who thought the applicant was white talked about his “potential” and that he had “good analytical skills.” The black applicant was “average at best” and one partner “can’t believe he went to NYU.” There are a vast number of studies similar to this that show the same phenomenon, of course. And lest we think we are free from such subconscious biases, you can take some tests yourself on it.

"Literal (sorry?) nazis in political high office, literal KKK defenders in high court .. sounds ..."

Unqualified Judicial Nominee Defended ‘First KKK’
"I think that's being a bit cavalier there!"

Unqualified Judicial Nominee Defended ‘First KKK’
"How about written accounts by three women (and a witness of one of these rapes ..."

Trump’s Blatant Hypocrisy on Sexual Harassment
"Wow, how do you sign up to be in the Illuminati? Powerful enough to direct ..."

Taylor: The Illuminati Sent the Hurricanes ..."

Browse Our Archives

Follow Us!

What Are Your Thoughts?leave a comment
  • tbp1

    Haven’t there been studies where identical CVs were sent out for job openings, but with some of them bearing names like DeShawn or Shaniqua and others more WASPy sounding names? And of course, those with typically white names got WAY more bites than those with black sounding names.

    Too lazy to look it up, bit I’m pretty sure about this.

  • alanb


    If my memory cells are still operating properly, I think that study was done about a decade ago. Obviously, not much has changed since then.

  • timpayne


    How can you possibly say not much has changed since then? Last year John Roberts declared racism in the US to be over as he gutted the Voting Rights Act. In fact, I don’t even know why Ed posts on these studies when we already know the right answer.

  • http://www.facebook.com/den.wilson d.c.wilson

    I wonder what the results would be if they repeated this study with current and former Supreme Court justices instead of law partners.

  • http://drx.typepad.com Dr X

    In two more conditions, all of the partners were told that this was a test to see if they were racists. Results: all of the partners described the “black’ memo as “articulate.” No such mention was made when they were told the writer was white. [/sarc]

  • Childermass

    And I heard a very similar story about professors on NPR either yesterday or the day before that.

  • http://anexerciseinfutility.blogspot.com tommykey

    Part of it is that many people assume a racist is someone who spews racist epithets, is a Klansman or skin head, and so forth, instead of recognizing that racist beliefs exist on a spectrum. A person can be consciously against racism, consider racial epithets to be repugnant, etc., and yet still harbor beliefs of superiority of his own race and assumptions of inferiority of other races.

  • moarscienceplz

    I just took an IAT that measured whether I automatically think of European-race people or Asian-race people as being more “American”. Since i am white and grew up among mostly white people, I expected to have some pro-European prejudices. But I was surprised and pleased that I show ” little or no association between European American and Asian American with American and Foreign”.


  • eric

    So, how do we fix this? I think we can reasonably assume that the partners were not trying to discount the talents of a good prospective employee. They were likely just trying to find the best person for the job. Telling them “don’t be racist” probably isn’t going to help because they are already trying not to be. Some initial thoughts, but I welcome feedback and criticism:

    -In some cases, formal employment criteria an a decision-making heuristic can help. But those sorts of systems are also subject to gaming and are not necesasrily approriate or realistic for all jobs.

    -Awareness of potential racist tendencies probably does help, even if pithy orders like “don’t be racist” don’t. Probably, having evaluators be aware that we all (they included) have racist tendencies may help them guard against their own biases and flaws. An analogy would be someone telling me “eric, you make a lot of typos” before I write a document – doing that might make me more aware of the typos that I make.

    -Acculturation. Probably not a viable “workplace” strategy, and not something you can really force, but in general, I expect that the more cosmopolitan a person is, the less influence these tendencies will have. So even if your work can’t or shouldn’t make you do it, get out there on your own and meet people not like you!

    -Other ideas?

  • Crimson Clupeidae

    Good questions, Eric. I guess recognizing is at least a decent first step.

    I suspect a similar study with women’s/men’s names would yield similar results.

  • Alverant

    How did the reviewers know the race of the applicant? Was there a checkbox or a name? Wouldn’t that make them aware of any biases that would affect the results?

  • http://cheapsignals.blogspot.com Gretchen

    So, how do we fix this?

    Having people in positions of power read studies like this would be a good start.

  • http://en.uncyclopedia.co/wiki/User:Modusoperandi Modusoperandi

    To be fair, they only rated the black candidates lower because their office already had one.

  • http://en.uncyclopedia.co/wiki/User:Modusoperandi Modusoperandi

    Gretchen “Having people in positions of power read studies like this would be a good start.”

    There are, but those that thought the the studies were by black people discounted them.

  • http://drx.typepad.com Dr X

    The study, which was conducted by a leadership consulting firm, did make recommendations.


    One interesting finding: race and gender of the evaluator did not change the results, though women found more errors across the board and wrote more lengthy feedback.

  • martinc

    The link Ed recommended under the words “ take some tests yourself”> ” is extremely interesting. I would recommend taking 15 minutes to do one of the tests.

  • scienceavenger

    So, how do we fix this?

    Blinding every aspect of the evaluation process that can be for starters. If we don’t trust experimental scientists to be unbiased towards impersonal data…