Closing the Gap at North Carolina’s Historically Black Universities

Earlier this month, the Triangle Business Journal revealed that graduates from North Carolina’s Historically Black College and Universities (HBCUs) are lagging their peers in terms of median salary after graduation. As the state bolsters its efforts to attract more students to its public HBCUs, it’s especially important to discover the cause of such disparities and determine how to close the gap.

The Triangle Business Journal’s findings are a good start. Using data from the Department of Education’s College Scorecard tool, TBJ analyzed the median earnings of students who received federal financial aid ten years after starting their education at each of the ten HBCUs in the state. They discovered that students who attended North Carolina’s public and private HBCUs earn $4,055 less than the national median, and significantly less than most students who attended predominantly white UNC system schools.

Data from NC Tower, a website using government data to track the employment outcomes of public college and university graduates, mostly confirm TBJ’s findings. These data, which come only from students who have graduated by the time of data collection and are employed in North Carolina, show that five years after graduation, most HBCU graduates earn less than those who attended many of the predominantly white institutions (PWI) in the state.

But the disparity between HBCUs and PWIs isn’t cut and dried. To be sure, graduates of UNC-Chapel Hill earn, on average, $20,000 more than graduates of Elizabeth City State University five years after graduation. But these two extreme cases tell only part of the story. Graduates of UNC School of the Arts and UNC Greensboro, both PWIs, earn significantly less than the median for all UNC schools ($33,750). And students at NC A&T, a land-grant HBCU in Greensboro, falls nearly at the median. The income of graduates who stay in North Carolina after leaving school tells an even more complicated story, with graduates of UNC School of the Arts at the bottom and Winston Salem State University at the top. (It is unclear why data from NC Tower and the College Scorecard are so dissimilar for WSSU graduates.)

Looking more closely at NC Tower’s data reveals one possible reason for the difference in salaries: industry of employment. Few students from schools with below-average earnings work in high-skill industries such as natural resources and mining; trade, transportation, and utilities; manufacturing; information; and finance. These industries account for many of the high-paying jobs landed by graduates of UNC-Chapel Hill and NC State. At Elizabeth City State University, by contrast, many graduates are concentrated in public administration, education, and retail.

The high earnings variation between industries, and, by extension, college majors, is well documented. Data from the Hamilton Project, an economic policy initiative at the Brookings Institution, shows that “for the median bachelor’s graduate, cumulative lifetime earnings across majors range from just under $800,000 to just over $2 million.” It’s telling that NC A&T, whose graduates boast the highest earnings among UNC system HBCUs, has the highest proportion of STEM majors.

But students’ choice of majors across universities does not explain all the variance between salaries. UNC Pembroke graduates working in retail make nearly $7,000 less annually than ECU graduates working in the same industry. It is unlikely that their field of study is to blame.

Marybeth Gasman, director of the Center for Minority Serving Institutions at the University of Pennsylvania, posited in a comment in the Triangle Business Journal that disparities in earnings could be explained by racism on the part of employers—who view HBCUs as inferior institutions.

An alternative explanation is that employers are responding to a real difference in graduates’ skills—and hiring the ones who are able to solve problems, think critically, and adapt to their role in the workforce. On all of these measures, HBCUs, which often have low admission standards, perform poorly. (Predominantly white institutions with low admissions standards also perform poorly.)

In an expose published earlier this summer, the Wall Street Journal described college students’ learning outcomes as “discouraging” across the board. But results were particularly concerning at HBCUs. The article examined students’ performance on the Collegiate Learning Assessment (CLA+), a test of critical thinking skills that was administered at 200 universities nationwide.

“At more than half of schools, at least a third of seniors were unable to make a cohesive argument, assess the quality of evidence in a document or interpret data in a table,” WSJ reported.

These are precisely the skills that employers want. A 2016 report of workforce skills preparedness by Payscale revealed there exists a significant skills gap between what employers want and the skills new college graduates possess. Sixty percent of managers identified critical thinking and problem-solving as the skills that most recent graduates lack.

Only four UNC schools were included in the report. But this snapshot of learning in the UNC system is telling. At East Carolina and Western Carolina, very few seniors left school with “below basic” skills. At Fayetteville State and UNC Pembroke (a minority-serving institution though not technically an HBCU), more than a quarter of seniors were considered “below basic” in their abilities to analyze, evaluate, and synthesize information.

Further data reported by schools to the Voluntary System of Accountability confirms the trend. Seniors at Appalachian State, East Carolina, and Western Carolina outperformed seniors at Fayetteville State, UNC Pembroke, and Winston-Salem State University on every almost measure. Among the schools who participated (see below), the scores of WSSU students came closest to those of students at predominantly white institutions.

Serious reforms are needed in order for graduates of HBCUs to compete with their peers from predominantly white institutions. In North Carolina, NC A&T is a model. It boasts higher admissions standards than other HBCUs in the state and a large number of students who major in in-demand fields such as nursing and engineering.

Reform starts with increasing minimum admissions standards. This will ensure that students arrive on campus prepared to learn at a college level. North Carolina Central University, Elizabeth City State University, and Fayetteville State University—three schools that lowered their standards for SAT scores to 750 combined math and critical reading—should return to 800 as the minimum score they will accept.

Once students are on campus, faculty and administrators should encourage them to take more challenging and in-demand majors. Course content in all courses and majors should be geared towards increasing problem-solving, critical thinking, and writing skills. A strong general education program, including strict requirements for math and composition, can help students learn the habits of mind that will prepare them for the workplace.

Competition in the modern workforce is fierce. North Carolina’s HBCUs must improve their performance in order to adequately prepare graduates for the future.

  • Guest

    Question: Did the three schools lower their SAT admission for all students or just lower for a limited trial (less than 100 students for each institutions for 3 years) with GPAs higher than the minimum admission requirement to do research as to whether the leading indicator of success is SAT score or High School GPA? Your article would lead readers to believe it’s an across the board reduction in standards.

  • thinkingprof2

    Increased admissions standards will not guarantee that students arrive on campus “ready to learn.” Increased admissions will only guarantee that students with scores showing they are “ready to learn” will be admitted. Are students in some of these colleges ill-prepared for college because the K-12 schools they attend lack resources? Some students avoid STEM areas that they are least prepared to study. Do the students with lower SATs do significantly worse at college than those with higher SATs? How do the earnings of graduates from the “least successful” schools compare the earnings of people who don’t graduate from college at all? In the current culture of higher education, students are concerned with grades, less so with “learning.” We can continue to blame universities, but only so much more assessment and attempts to force students to think critically if they won’t can be piled on faculty. And now universities are being pushed to focus on retention and future jobs. University isn’t only about jobs, But is it possible that some of the jobs students get after graduation are under valued and under compensated–for example public administrators and teachers? Isn’t that a bit sad?

  • bdavi52

    The market is chock-full of disparities. That is the nature of markets.

    Goods & services (including people & their talents) are offered …in response to demand… and through market operation over time, equilibrium prices are established at which exchange occurs. It’s really very simple.

    It should not be at all surprising that graduates from sub-average programs earn sub-average wages. Why would they earn more? It should not be surprising that graduates from Harvard or MIT earn more than graduates from UNC…and that graduates from UNC earn more than graduates from Elizabeth City State and that graduates from Elizabeth City State earn more than graduates from various Community College programs. That is how this whole thing works.

    We pay more for Heineken than for PBR; we pay more for Cheerios than for Generic Grocery O’s. That is what branding does.

    And when brand/reputation is backed-up by actual on-the-job performance — demonstrating that these ‘branded’ graduates really do possess a superior & productive skill-set (a skill set that sub-program grads tend not to have), the outcome is obvious & entirely predictable. We’ll pay more for Superior Candidate “X”, and less for Not-So-Superior Candidate “Y”.

    Is this discrimination? Of course it is. The market discriminates according to quality & proven performance. Is it racial discrimination? Nah, not at all….especially not in a world in which significant weight is given to having enough URM hires to avoid lawsuits.

    As with most things, as Occam noted, the simplest & most direct explanation is almost always the most likely.

    • AtlanticReader5

      No any gap is bad and we must fight to close it.

      • bdavi52

        Why on earth would you think that?

        A gap is an indication of difference and the world is full of difference, good & bad. If I’m taller than you, better looking than you, smarter than you, if I work harder than you, if I accomplish more than you — then there will be a series of differences between the two of us, rightfully so.

        Harvard is a better school than UNC. UNC is a better school than ECState. The better schools are better because they are able to hire and retain better teachers, from better programs. They’re better because they have better resources and more resources. They’re better because they select only high-quality students who themselves then perform better than lower-quality students. If given a choice, all other things being equal, who among us would choose ECState over Harvard or UNC, when the latter two choices are — by every measure — superior??

        We differentiate constantly. Differentiation is a fact of life.

        And absolutely, I’m sure you’d agree, that if you’re faster than me you should win more races than i do. If you’re smarter than me, then you should solve problems more effectively than i do. If you’re more qualified than I am, then you should get a better job than the one I get

        • AtlanticReader5

          Obviously if you’re taller than me or smarter than me and society rewards you thats ableism and you should be punished and feel guilty for being tall. We need lots of government programs to pay people who are short and dumb.

          Harvard is a racist institution because it doesn’t admit students in exact proportion to their share of US demographics. Any argument you try to counter this means you are a racist.

          You should only be allowed to win more races if you are a bigger victim of systemic oppression. This will counter the history of injustice you have faced.

          /sarcasm…

          • bdavi52

            Sorry…I should have known.

            Sadly, we have reached (as others have noted) the Onion Singularity….that point at which it’s impossible to distinguish sarcasm & sarcastic exaggeration from actual news & actual commentary.

  • Glen_S_McGhee_FHEAP

    At a more basic level, I am concerned about the reliability of the data in Table One.
    Plopping the info into a spreadsheet, the variances between the TBJ/Scorecard and NC Tower data seem almost random. Shouldn’t they track each other?

    TBJ NC Tower Variance
    UNC School of the Arts $31,200 $15,611 99.86%
    Elizabeth City State $29,500 $27,300 8.06%
    UNC Asheville $34,500 $27,303 26.36%
    NC Central $31,600 $31,740 -0.44%
    Fayetteville State $30,500 $31,792 -4.06%
    NC A&T $33,000 $31,882 3.51%
    UNC Wilmgton $39,400 $33,490 17.65%
    UNCP $32,300 $33,535 -3.68%
    Western Carolina $35,800 $33,642 6.41%
    Appalachian State $36,400 $33,951 7.21%
    UNC Greensboro $32,300 $34,409 -6.13%
    UNC System $33,750 $35,303 -4.40%
    UNC-Chapel Hill $51,000 $36,670 39.08%
    East Carolina $39,000 $37,601 3.72%
    UNC Charlotte $41,500 $38,376 8.14%
    NC State $47,600 $40,562 17.35%
    WSSU $32,300 $43,678 -26.05%
    Average $611,650 $566,845 +7.90%

    • AtlanticReader5

      The variances should be random. If both the TBJ and NC Tower estimates are modeled as (True Value + Independent Random Variable) then the differences between these values will be a random variable.

      TBJ is biased upwards but theres a lot of noise.

      • Glen_S_McGhee_FHEAP

        If the variances are random, then so are the underlying independent variables, right?

        Thank you for trying to clear this up, but what is being measured? Are TBJ and NC Tower methodologies trying to measure the same thing, or not?