This week, the Albert Shanker Institute released a new research brief, authored by myself and Klarissa Cervantes. It summarizes what we found when we contacted all 51 state education agencies (including the District of Columbia) and asked whether data on teacher race and ethnicity was being collected, and whether and how it was made available to the public. This survey was begun in late 2017 and completed in early 2018.

The primary reason behind this project is the growing body of research to suggest that all students, and especially students of color, benefit from a teaching force that reflects the diverse society in which they must learn to live, work and prosper. ASI’s previous work has also documented that a great many districts should turn their attention to recruiting and retaining more teachers of color (see our 2015 report). Data are a basic requirement for achieving this goal – without data, states and districts are unable to gauge the extent of their diversity problem, target support and intervention to address that problem, and monitor the effects of those efforts. Unfortunately, the federal government does not require that states collect teacher race and ethnicity data, which means the responsbility falls to individual states. Moreover, statewide data are often insufficient – teacher diversity can vary widely within and between districts. Policymakers, administrators, and the public need detailed data (at least district-by-district and preferably school-by-school), which should be collected annually and be made easily available.

The results of our survey are generally encouraging. The vast majority of state education agencies (SEAs), 45 out of 51, report that they collect at least district-by-district data on teacher race and ethnicity (and all but two of these 45 collect school-by-school data). This is good news (and, frankly, better results than we anticipated). There are, however, areas of serious concern.

Most troublingly, six states – Alabama, Delaware, Maine, Vermont, Virginia, and West Virginia – report that they do not collect any data at all. This is not acceptable. In these states, there is no way to measure even district-level teacher diversity levels and trends.

In addition, the overall situation is a bit less positive when it comes to whether (and how) SEAs make the data available to the public. Specifically, of the 45 states that collect at least district-by-district teacher diversity data, four do not make these data available to the public. This means that ten states – roughly one in five – either do not collect or do not make public teacher diversity data. Again, it is important to acknowledge the positive side here – i.e., four in five states do collect the data and make it public. In our opinion, however, ten states is ten too many.

Moreover, among the 41 SEAs that report (to their credit) that they make at least district-level data available, only 21 do so on their websites. The remaining 20 SEAs require that the data be requested. We made all these requests. Most SEAs filled the requests relatively quickly and free of charge, but some took a while to respond and a few charged a small processing fee (all of the datasets we received are available for download on the report's webpage). Alabama, for example, took several months (and repeated inquiries in writing and over the phone) to report that they do not collect the data. 

Even if the requests are filled quickly and without charge, however, the requirement represents an unnecessary obstacle to access for policymakers and especially the public.

To conclude our report, we offer a few basic recommendations. First, the six states that do not collect any teacher diversity data should begin doing so immediately. Second, the four states that collect only district-by-district data should begin collect the more detailed school-by-school data. Third, all states should make their teacher diversity datasets available on their websites. And, fourth and finally, the U.S. Department of Education should begin collecting teacher diversity data as part of its Civil Rights Data Collection.

Education is a field awash in data, including data on teacher characteristics (experience, certification, etc.). That is generally a good thing. Teacher diversity should be no exception. It matters for student outcomes (and I like to think that students being exposed to teachers from different backgrounds is desirable in itself, regardless of whether this benefit shows up in observable outcomes such as test scores). 

Moreover, the U.S. public school student population is quickly becoming more diverse, and the teaching force is becoming increasingly unrepresentative of the students it serves. Collecting and promulgating data will not by itself solve this problem, but it is certainly a basic necessity for understanding and addressing it effectively.