Futility of “Colorblindness” and Confirmation Bias: Can Poison Workplace Interaction of a Supervisor and Subordinate Employee; These Interactions Must Be Covered under Equal Employment Opportunity Law

In the United States, the manager of an organization is give great power to hire, supervise, instruct, discipline, and terminate employees, according to the laws relating to employment (indexed as master and servant) law.

Perhaps in a sufficient number of cases, manager’s assessment in an applicable discretionary power is correct, but in a situation of implied bias–that the supervisor is prone to disfavor the work of a subordinate just because of his or her skin color, such discretion becomes racially discriminatory and unfair. The subordinate cannot do anything to avoid the effects of a manager who applies a unconscious bias to such subordinate.

As a result, broad assertions (even with manager-supplied documented “proof”) of poor performance, must be viewed with a skeptical eye and with a view of ensuring that the manager’s assessment and documentation is itself fair and equitable to the subordinate and not just accepting the manager’s compilation of errors is the final word. This point is a definite weakness of the Grote forced distribution system (governed by manager’s unchecked “opinion”), covered separately in this blog.

Exploring the issue of unconscious bias, Dr. Arin N. Reeves, a researcher with Nextions, a consulting organization, investigated the effect of confirmation bias. Specifically, Dr. Reeves was investigating the reason from prior research finding that supervising lawyers perceived African American lawyers to be sub par in their writing skills in comparison to their Caucasian counterparts.

Dr. Reeves and her team performed a study, studying the issue from a perspective of unconscious or implicit bias. Five law partners were asked to write an legal research memo from a hypothetical third-year litigation associate that focused on trade secrets in Internet start-ups. The researchers then introduced 22 errors–spelling and grammar (7), substantive technical writing errors (6), errors of fact (5), and errors of analysis (4).

This error-amended memo was provided to 60 partners of law firms; one-half of them were told that the memo was written by an African American associate, and the other one-half of them were told that the author was Caucasian.

The result–from 53 law partners (24 having reviewed the African American associate’s memo and 29 having reviewed the Caucasian associates memo)–is as follows:

African American Associate Caucasian Associate
Overall quality of the memo (rated from 1 (poor) to 5 (extremely well written) 3.2/5.0 4.1/5.0
Qualitative comments
  • “needs lots of work”,
  • “can’t believe he went to NYU”,
  • average at best”
  • “generally good writer but needs to work on…”,
  • “has potential”,
  • “good analytical skills”
Specific errors
Spelling and grammar errors found 5.8/7.0 2.9/7.0
Technical writing errors 4.9/6.0 4.1/6.0
Errors in fact 3.9/5.0 3.2/5.0
Errors in analysis Rated better overall because he had fewer critical comments.
Formatting (not requested by research group, but 41 comments received) 29 comments / 41 11 comments / 41

The researchers noted that there was no significant correlation between a partners race/ethnicity and the differentiated patters of errors found between the two memos. In addition, there was no significant correlation between a partner’s gender and the differentiated patterns of errors found between the two memos, the researchers continued. The researchers stated that they did find that female partners generally found more errors and wrote longer narratives than the male partners.