Featured Product
This Week in Quality Digest Live
Customer Care Features
Bryan Christiansen
And when to hire one
Etienne Nichols
What medical device manufacturers need to know
Kate Zabriskie
Don’t get distracted by dazzle
Trupti Dhere
Technology helps doctors make more informed decisions, minimize adverse events, improve patient care
Greg Kihlström
First, ask the right questions

More Features

Customer Care News
Yotrio and SunVilla to provide interactive, 3D-enabled assembly via BILT app
Algorithms protect data created and transmitted by IoT and other small electronics
Customers can place orders in a self-service configurator
Application now supports 14 leading industrial robot manufacturers
New digital service supports all Fastems systems
Key Technology’s system boosts quality, throughput
Driving advances across the electronics ecosystem
Weighing supply and customer satisfaction
Showcasing tech, solutions, and services at Gulfood Manufacturing 2022

More News

Dave Gilson

Customer Care

Use a Scorecard to Evaluate People More Fairly

A written framework is an easy way to hold everyone to the same standard

Published: Wednesday, October 26, 2022 - 12:02

Like most of us, lawyers think they can be impartial when they rate other people’s work. “They say, ‘Who writes a brief doesn’t matter. A brief is a brief; it stands on its own merit,’” explains Lori Nishiura Mackenzie, the lead strategist for diversity, equity, and inclusion at Stanford Graduate School of Business.

She cites an experiment in which 60 law firm partners were given a legal memo peppered with errors. All were told that a young lawyer had drafted it. Half were told that the writer was white; the other half were told he was Black.

When the partners’ evaluations of the memo came back, the imaginary “white” lawyer received an average score of 4.1 out of 5 and was judged a “generally good writer.” The “Black” lawyer got a 3.2 and was deemed “average at best.”

Even when we think we’re being objective, biases can creep in. So how can we be more consistent and fair when we evaluate candidates and co-workers?

Mackenzie offered some ideas in “The Myths and Rituals of Inclusion,” a talk she gave last spring. A starting point, she says, is to be aware of how we shift our criteria for people based on irrelevant assumptions. “If you start by thinking carefully about how you’re going to evaluate someone before you do, you’re less likely to shift.”

Those shifts may be subtle, but can skew outcomes.

“Sadly, this happened to me once at work,” Mackenzie recalls. A strong applicant for a position was penalized for misspellings in their cover letter. “I didn’t say, ‘Did you equally check for spelling mistakes in all the candidates?’ Because if we had, I’m sure we would have found a similar number.”

An easy way to hold everyone to the same standard is to use a written framework or rubric for assessment. “If you in your work are making decisions about people without some sort of scorecard, likely you are making these shifts,” Mackenzie says. “While bias thrives in ambiguity, consistency has a chance in blocking biases in decision making—and that is good for everyone.”

First published Oct. 4, 2022, on Insights by Stanford Graduate School of Business.


About The Author

Dave Gilson’s picture

Dave Gilson

Dave Gilson is a senior editor at Stanford Graduate School of Business, where he edits coverage of faculty research, profiles of alums and professors, and Stanford Business magazine. Previously, he was the deputy editor at Mother Jones and taught an introductory reporting class at the University of California-Berkeley Graduate School of Journalism.


Trying to rate people instead of the system?

Didn't Deming teach us that rating people rather than the system they work within is destructive?

He said again and again that the annual evaluation should be abolished but here is a quality publication pushing better methods to do something that is a bad idea.  Why?

We should focus on improving the system.

Trying to rate people instead of the system?

People represent the system. It's not what you measure in the system, but what you are going to do with the measurements.