Thread by @lavaredmonds

🚨 Working Paper Alert! 🚨

I’ve basically been soft launching this for a while now, but I’m ready to share my first Big Kid working paper. Join me for a 🧵 on HBCUs, teachers, and Black student achievement! Comments and suggestions are much appreciated!

lavaredmonds.com/uploads/1/4/2/…

Tl;wr Black elem school students in North Carolina score higher on standardized math tests/are less likely to be suspended when assigned to teachers from Historically Black Colleges. Effects are independent of teacher race: students score well w/ Black AND White HBCU teachers

There’s no way I could do them justice in a few tweets, but I’ll try: HBCUs comprise roughly 100 institutions founded before 1964 with the purpose of educating Black Americans locked out from higher education

They have a rich history of educating Black teachers, many founded or existing as normal schools (institutions designed to train teachers). This has continued to today: in 2019-2020, they graduated about 8% of Black college grads, but just over 20% of Black grads with ed degrees

I also argue they’re the successors to 20th century Black educator pedagogy, an edifying period with curriculum developed by Black teachers and principals across the country explicitly geared towards educating Black students in all-Black K-12 schools

During this period, many K-12 principals and schools worked directly with HBCUs to develop a pipeline for sending their students to HBCUs for training, ultimately helping to facilitate a labor market for Black teachers to return to those K-12s

Moreover, with culturally-fluent instructional practices key to teacher education at HBCUs, it’s clear they have historical and contemporary grounding in preparing teachers to excel in the classroom especially with Black students.

Can we quantify any effects? We can!

An overview of the empirical strategy: I specify a TWFE regression, using within-student variation in teacher assignment to estimate differences in end-of-grade math exam scores for Black 3rd-5th graders in North Carolina, 2010-2018

What are the effects on test scores for Black students assigned to HBCU-trained teachers?

Positive! Black 3rd-5th grade students in NC score ~0.033 SD higher on math exams when assigned to HBCU-trained teachers, equivalent to ~5% of the estimated Black-White test score gap

(And don’t worry: I don’t spend this much time on EconTwitter just to post a TWFE paper without doing some homework. I try out lots of different estimators, all of which point to the same conclusions. If anything, TWFE may be underestimating the “true” effect of HBCU teachers)

What about effects on other students? Do non-Black students do worse with HBCU-trained teachers?

Nope!

Is it about some sort of learning on the job or teacher peer effects?

Probably not! Even when restricting to novice teachers (i.e., those in their first three years of teaching), HBCU-trained teachers appear more effective with Black students

How does this fit w/ same-race teacher effects? Isn’t this all just driven by teacher race?

I don’t think so! Black students score higher paired w/ Black HBCU teachers over other Black teachers, but they also score higher with *White* HBCU teachers over other White teachers

Moreover, in this setting, I find weak evidence of a same-race teacher effect at all: there’s basically zero difference in Black student test scores with Black teachers. In fact, I only note a same-race teacher effect for Black students when that teacher *went to an HBCU*

What about outcomes other than test scores?

Incidentally, I also look at suspensions! Evidence is more suggestive here, but I find Black student assignment to an HBCU-trained teacher is negatively associated with their likelihood of being suspended (particularly for Black boys)

Once again, the HBCU effect seems to move independently from race: Black boys are *more* likely to be suspended with White teachers in general, but *less* likely with HBCU-trained White teachers (compared against other White teachers)

What do we know about teachers at HBCUs?

Tricky! W/o pre-teaching data, it’s hard to rule out a selection story (e.g., of course White students at HBCUs get there non randomly). I offer conjectures, but I hope some enterprising scholars continue after me and expand on this point

That said, one of the things I find most curious in the data: by conventional measures, these teachers should be “weaker” academically. They have much lower licensure exam scores (almost 0.75 SD gap in the sample), and HBCUs accept students with much lower SAT scores

Yet they’re clearly outperforming more “qualified” teachers. At a minimum, this raises the question of what we’re measuring if exams purported to capture teacher preparation are under-predicting what these HBCU teachers are bringing to the classroom for Black students

Wrapping up: HBCUs very much still matter in today’s higher ed landscape and excel at producing exemplary teachers, seemingly independent of teacher race, to the benefit of Black student academic and social-emotional outcomes

Racial achievement gaps stem from historical/present-day deficits in resources allotted to Black students and the schools they attend, which can include access to high-quality teachers committed to both their academic success and mental and emotional well-being

In addition to extending same-race match lit to consider teacher training/background as a new channel, this paper offers evidence supporting a simple, targeted, and effective approach for raising student achievement, one largely omitted from prior teacher labor market discussions

With declining enrollment shares and financial concerns plaguing many HBCUs in recent years, policymakers should grapple with the considerable value to education potentially lost in the absence of these institutions

Post-script: I got the hunch most might be less interested in the “How does this fit into X literature?” parts, so this is at the bottom for those still reading!

Researchers often consider ?s about racial match — what are the effects of having a same-race [judge/physician/boss/etc.]? In ed, this comes up w/ same-race teachers, many showing + effects (on test scores, disciplinary outcomes, HS completion) for racial/ethnic minority students

Mechanisms remain unclear, however. Some non mutually exclusive options have been raised — role model effects, stereotype threat, culturally-fluent pedagogy — each one with rich theoretical/empirical research. Still, in this particular setting, I find them somewhat less fitting

To say Black students perform better with Black teachers b/c the teachers serve as role models implies a lot about Black student backgrounds and teacher agency that doesn’t seem entirely empirically grounded

Stereotype threat can be a powerful phenomenon, but I’m not sure how much weight to give that if students are unaware of the negative stereotypes about their group that could influence their behavior

Culturally-fluent pedagogy can be invaluable in the classroom, but Black teachers aren’t monolithic, and sociocultural competency isn’t innate, so I’m somewhat wary of linking this to such a diverse group of teachers

Instead, I put forth that variation in teacher training is a vital piece of teacher effectiveness, one that has a role to play here in Black student achievement. One setting for testing this hypothesis? Teachers from Historically Black Colleges and Universities!

There, I think that about ties everything together. Thanks for reading this far! I hope this has been at least somewhat insightful. Feel free to reach out with any feedback; I’m very much still learning as I go, so, agree or disagree, I’d love to talk with you! 😁