Was DEI Hiring Good or Bad for America?

Things went too far at the FAA, yet hiring bias persists – BCB #137

When a passenger plane and an Army helicopter collided above Washington D.C. last week, Trump wasted little time blaming diversity initiatives at the FAA for the deaths of everyone onboard. Although this is pure political posturing, there is more than a grain of truth to the idea that the FAA skewed hiring practices for diversity reasons – as documents from a spate of lawsuits make clear.

Yet experimental evidence shows that hiring bias against minorities persists. For all its dysfunctions, DEI is an attempt to combat real problems.

There can be no blanket answer to the question of whether DEI-style hiring practices were constructive or destructive. As the following three case make clear, it’s important to be specific about where you’re looking and what you’re talking about.

Hiring bias isn’t a myth

Let’s start with the broadest view. There’s still plenty of evidence of discrimination in hiring, when examined on the whole. For example, last year a group of economists published a working paper which found that, on average, applicants with white-seeming names were 9.5% more likely to be contacted by employers than applicants who were presumed to be Black.

To conduct this experiment, the largest of its kind in the US, the researchers sent in resumés for 10,000 entry-level jobs at around 100 of the country’s largest companies. These fictional resumés featured nearly identical qualifications but different names that hinted at the applicant’s race.

Results varied dramatically by company and industry, though at least some bias against Black-seeming applicants was present across the board. For instance, sales and retail companies, especially in the auto sector, were most likely to prefer candidates who were presumed to be white.

(Source)

While many claims of bias are based on the demographics of who is hired, such statistics can’t distinguish between applicant quality and discrimination. Resumé studies like this one are important because they hold applicant quality fixed by changing only the name on a resume. And what they’ve found is that when everyone has the same qualifications, some people really do have an unfair advantage.

The FAA’s misguided diversity initiative

And yet, that doesn’t mean that DEI-style interventions in hiring have been effective – or even justifiable. One such instance can be seen, unfortunately, at the FAA. In a series of two articles, TracingWoodgrains dives into the byzantine story of efforts to diversify the air traffic control (ATC) workforce that adversely impacted staffing, and potentially aviation safety more broadly.

For many years, he explains, air traffic controllers were hired from the graduates of FAA partner schools (known as Collegiate Training Initiatives, or CTIs), after passing the accompanying professional aptitude test (called the AT-SAT). But over time the FAA came under mounting pressure to diversify. In response, the FAA made suddenly announced a series of major changes to its admissions process for air traffic controllers on New Years Eve of 2013:

First, every past aptitude test applicants had taken was voided…

Second, every applicant would be required to take and pass an unspecified “biographical questionnaire” to have a shot at entering the profession.

Third, existing CTI students were left with no advantage in the hiring process, which would be equally open to all off-the-street applicants—their degrees rendered useless for the one specialized job they had trained for.

The “biographical questionnaire” was a take-home assessment designed so that more than 90% of applicants would fail, the goal of which was to “maximiz[e] diversity.” The questions and correct answers were evidently bizarre, seemingly weighted with other criteria in mind than selecting the best possible candidates. For example, the correct answer to “the high school subject in which I received my lowest grade” was “science.” Applicants who got mostly D’s in high school were graded higher than those who got mostly C’s, and points were awarded for playing a greater number of different sports.

Grading scheme for the question “the high school subject for which I received my lowest grade was” (source)

What’s more, Shelton Snow, an FAA employee who was also president of the Washington Suburban chapter of the National Black Coalition of Federal Aviation Employees (NBCFAE), called up applicants who were members of the organization and gave them specific directions on how to answer it:

he sent voice-mail messages to NBCFAE applicants, advising them on the specific answers they needed to enter into the Biographical Assessment to avoid failing, stating that he was "about 99 point 99 percent sure that it is exactly how you need to answer each question."

The 2013 changes to the air traffic controller hiring process triggered a number of lawsuits, including the class action lawsuit that initially spurred TracingWoodgrains’ investigation. By 2016, Congress had passed a law that got rid of the biographical assessment as a first-line hiring tool and required the FAA to give individuals affected by these hiring changes the opportunity to reapply for a job as an air traffic controller.

But it seems likely, TracingWoodgrains writes, that this whole incident had a negative impact on aviation safety and air traffic controller shortages:

The 2014 change materially shifted applicant quality on net… Not only that, it shattered the pipeline the FAA had built with CTI schools, making the process towards becoming an air traffic controller less certain, undercutting many of the most passionate people working to train prospective controllers, and leading to a tense and unclear relationship between the FAA and feeder organizations.

Did anyone truly unqualified make it all the way through the pipeline? There's no reason to think so. Did average candidate quality decrease? There's every reason to think so. Would that lead to staffing issues? Unambiguously yes.

Don’t assume bias, check for it

Predominant narratives in the academic world would suggest that bias against women persists in the research sciences. But a 2023 synthesis of scholarly literature on the subject shows that better answers are available, if we look for them.

This is the result of an adversarial collaboration between three researchers with different views on the subject. They looked at six different areas where sexism is often thought to be most pervasive: tenure-track hiring, grant funding, teaching ratings, journal acceptances, salaries, and recommendation letters. They found that tenure-track men and women were at parity when it came to grant funding, journal acceptances, and recommendation letters. When it came to teaching rates and salaries there was evidence of some bias against women, though the salary gap was smaller than is often thought to be the case.

When it came to hiring, however, they found clear evidence that women who apply for tenure-track jobs receive offers at an equal or higher rate than men do. Perhaps the most striking study they drew from, which was published in 2015, found that faculty members in charge of hiring for tenure track positions in biology, engineering, economics, and psychology preferred female applicants 2:1 over males with all of the same credentials.

The authors behind the 2023 study do offer the caveat that there are a host of systemic reasons that lead to women being less likely than men to go after these jobs in the first place, including “difficult work schedules and inflexible timing imposed by the tenure-track system during the decade when most people are building families.” But their review of existing literature, records, and experiments clearly indicates that the reason there aren’t more women in tenure-track STEM positions is not because of a discriminatory hiring process.

All of this is evidently complicated and highly particular. But what this study makes clear is that gender hiring bias isn’t universal across industries or contexts. Never assume you know the answer! And in the meantime, it’s worth asking whether a 2:1 advantage is really best for anyone in the long term.

What now?

In order to have a better conversation about DEI, we’re going to need to stop making sweeping statements. There’s no universal answer to the question of whether a particular company or institution is discriminatory vs. woke.

Taking a look at these three cases, it seems clear that some DEI programs may have been excessive, perhaps even to the point of impacting safety. But it’s also clear that very real hiring biases in race and gender persist. Somewhere in all of this, there are the kernels of a balanced and effective hiring strategy. As TracingWoodgrains argues in his investigation,

Fundamentally, people should not be forced to choose between “burn it all down” and “sweep it under the rug.” People should not be forced to choose between lowering the bar and rejecting all outreach to struggling communities. Given the recent election, Democrats have years to build a vision of diversity that involves providing more opportunities for everyone to reach a high bar, not lowering the bar—one that involves making institutions work, not obfuscating institutional failure until it can no longer be hidden.

Quote of the Week

do not allow the worst 1% of your political enemies define your world view

this is the single biggest hook that the mind virus uses to draw more people into its maw of derangement and vitriol

- @MarkovMagnifico