Paul Spella / Shutterstock / The Atlantic

In the coming decades, millions of aging Americans will find themselves needing medical services more than ever before—right in the midst of a projected massive physician shortage. The seas will rise, as will the need for scientists to further develop carbon-free energy and engineers to build the infrastructure to protect people and ecosystems from extreme weather events. And as artificial intelligence, cybersecurity, and data science advance and become more enmeshed in everyday life, the expertise to harness them in an ethical and effective manner will depend on those with advanced training in these fields.

And yet America is nowhere near drawing on all of the American people’s talents to solve these problems. Our future depends on a robust scientific workforce. But racial and ethnic minorities are underrepresented in these fields, and millions of people who should be making important breakthroughs are instead—whether because of inadequate public education where they live, a lack of resources and support for college and graduate school, discrimination as they try to get their first job, or a culture of science that weeds out rather than encourages undergraduates—doing other work. While African Americans make up 13 percent of the U.S. population and Latinos 18 percent, they were just 4 and 5 percent, respectively, of new natural-sciences and engineering (NSE) doctorates in 2016. If more people of color were able to join the ranks of scientific researchers across America, our society would be much stronger for it.

This isn’t a new problem. For decades, leaders in higher education have lamented the lack of diversity in these fields. But for the most part, they have acted as though nothing could be done about it, at least not yet. They were guilty of a lack of imagination at best or a lack of commitment at worst. But at the University of Maryland at Baltimore County (UMBC), we imagined that we could support the success of scientists and engineers of all backgrounds. And for three decades now, we have been doing it.

In the late 1980s, during a period of racial unrest on our campus, we sought to understand what was underlying student frustration. What we found opened our minds. Most students of all backgrounds were not succeeding academically. Black students were faring worse than other racial groups, and black men worst of all. Also, students of all groups in NSE were performing worse than those in other fields and, again, black students were faring worse academically than others.

Some would have seen these results and developed a “remedial” program that would have focused on the weaknesses of students. In contrast, we decided to focus not on the shortcomings of our students, but on those of our institution. We developed a program that emphasized student strengths, changing the perception of who could succeed in higher education—and specifically in science and engineering. We created a program that would recruit annual cohorts of high-achieving students and provide them with the support they needed to succeed as undergraduates and then go on to earn doctorates.

We began with the group that was performing at the lowest levels: black men. Thinking that we should learn from others, we looked for institutions that had been successful in supporting substantial numbers of African American undergraduates who went on to earn doctorates. At the time, there were none other than historically black colleges and universities . UMBC was about 85 percent white. What would it take for a school that was predominantly white to get its black students to thrive in STEM fields?

In 1988, we launched the Meyerhoff Scholars Program, named for the philanthropist Robert Meyerhoff, a real-estate developer from Baltimore who was especially interested in improving the fortunes of black men. Originally, the program focused exclusively on African Americans. The first cohort, in the fall of 1989, was composed of 19 black men, and the second included black women. However, white parents brought a lawsuit against another university in Maryland in the 1990s arguing that a minority program there could not be based on race and the courts found in the parents’ favor. Consequently, we decided to open up our program to anyone who demonstrated a strong commitment to supporting diversity and inclusion in NSE. This academic year, the program has 245 students, three-quarters of whom are from underrepresented racial and ethnic groups. (The program today receives financial support from the Meyerhoff family, the National Institutes of Health, and other agencies.)

We built the program on four pillars. First, we set high expectations for students, staff, faculty, and the institution. To participate in the program, students must aspire to careers in research. To support these aspirations, the program provides these students with financial support so that they do not need to work off-campus and can focus on studying. It also provides personal advising, involves family members who reinforce program goals, and engages students in community service.

Second, a strong sense of community lies at the heart of the program, because people persist and thrive when they feel as though they belong to something bigger than themselves. Students enter as a cohort of about 60 students. They live together on campus. We strongly encourage group work, so that students learn from one another and thrive together. We speak of the “Meyerhoff family.”

Third, we provide academic support for rigorous coursework that makes high-achieving students even stronger. The summer before freshman year, students participate in a “bridge program” that introduces them to one another and college life. The program encourages study groups and provides tutoring. Our faculty pull students into their research and we provide extensive research internships. This is a central program component: Engaging in scientific research enhances learning, promotes identification with the field, enables networking, and supports professional development.

Fourth, we built evaluation into the program from the very beginning to ensure that what we set out to do was actually happening. We collect data—both on the experiences of students in the program and on how their outcomes compare with those of similar groups of students—to help make the case to our campus and to the higher-education community more generally that the program has succeeded.

UMBC is now first among predominantly white institutions and second overall in the number of African Americans it educates who go on to earn NSE doctorates. We have educated more African American undergraduates who have gone on to earn the M.D.-Ph.D. than any other institution ever. Meyerhoff Scholars are more than five times as likely to have graduated from or be currently enrolled in a doctorate or M.D.-Ph.D. program than students who were invited to join the program but declined. More than 1,100 students have graduated from the program; nearly 500 have earned doctorates, M.D.s, or M.D.-Ph.D.s.; an additional 300 have earned master’s degrees; and another 300 are currently enrolled in graduate- or professional-degree programs.

In March 2018, the unimaginable happened when UMBC upset the University of Virginia in the first round of the 2018 NCAA men’s basketball tournament. Something else, once almost as unimaginable, could be observed at the tournament. With us in the crowd that weekend were four Duke University School of Medicine faculty members—all black men who are UMBC alumni. They all engage in cutting-edge research. One, for example, is working on the development of a pacemaker for the brain that may someday be used to treat conditions such as depression, autism, and schizophrenia.

The Meyerhoff Program was not a marginal effort, but one that shaped UMBC’s culture, defined our institution, and provided insights into how to support academic success for all students. The importance of institutional culture change for program success cannot be overstated.

Change requires honesty. Our approach began by looking in the mirror and being honest about strengths and weaknesses, opportunities and challenges. We gather data to understand problems and assess potential solutions. We have deep, sometimes difficult conversations about what it would take to fix the problems. Change requires courage because it rests on a willingness to see things differently and act in the face of uncertainty. When we launched the Meyerhoff Program, we represented it as a scientific experiment. Faculty who were initially skeptics eventually became allies, and later program champions. And that’s important, because change also requires what we call “grit.” Successful programs do not just happen. They require buy-in, commitment, and hard work from people who value the effort. Our university leadership made this program an institutional priority. Staff and faculty put in the hours to make the program successful.

Even after we demonstrated success, many leaders in higher education argued that the program could not be replicated elsewhere. This criticism has been put to the test. Since 2014, the Howard Hughes Medical Institute (HHMI) has invested $8 million to replicate our program at Penn State and the University of North Carolina at Chapel Hill. The successful results of this effort have recently been published in Science. HHMI is now investing in replicating the program at another six universities, and the Chan Zuckerberg Initiative is doing likewise at UC Berkeley and the University of California at San Diego.

Today, African American and Latino researchers remain deeply underrepresented in the professoriate and our scientific enterprise. Among full-time faculty, 77 percent are white, 10 percent Asian/Pacific Islander, 6 percent black, 4 percent Hispanic, and 3 percent other or mixed race. Just 2 percent of scientists at the National Institutes of Health are black, and 4 percent Latino. No institution is producing more than 12 African American undergraduates a year who go on to earn STEM doctorates.

The work to build the pipeline is urgent and must be widely shared. We have demonstrated that students of all backgrounds can succeed in science and engineering. But in the end, this is not just about underrepresented groups. The problem of diversity in science is the problem of helping all Americans in science, which is in turn the problem of helping America.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.