Should children be taught by teachers who "look like them"?
A problem of diversity and unexpected consequences.
Most teachers look like me, although probably not as tall and handsome*
In England, the teaching profession is disproportionately white British, with 84.4% of teachers in state schools belonging to this ethnic group. It's hard to know what to use as the base rate for comparisons, as many sources of demographic information include more than just England, but there are plenty of indicators suggesting that this is high. 74.4% of the population of England and Wales are white British. 62.6% of school pupils in England are white British. The reasons for the high proportion of white British teachers are not obvious, because the under-represented groups are so different, and the exceptions to under-representation are even less obvious.
A fun fact you probably didn’t know
In addition, the two ethnic minority groups that are better represented among teachers than among school pupils are white Irish and black Caribbean. Good luck finding an explanation for why these two groups, and these two groups alone, should be exceptions to the under-representation of ethnic minorities in teaching.
Reasons to worry
Our lack of understanding regarding the reasons behind the under-representation of ethnic minorities in teaching makes it challenging to assess how much of a concern it should be. It seems plausible that a teaching profession that is so skewed that it does not attract applicants from the whole of society will be missing out on talent. Likewise, it would be good for the profession to attract excellent teachers from all ethnic backgrounds. Moreover, pupils in English schools may suffer from being taught by a profession that overlooks potentially excellent teachers due to their ethnicity. However, much of the discussion about teachers and race has focused on the idea that ethnic minority pupils in particular are missing out because of a lack of ethnic minority teachers.
This argument was used in a recent Guardian article discussing a research paper:
Prof Stephen Gorard, the centre’s director, said the low numbers were partly the result of minority ethnic applicants to teacher training being more likely to be rejected, with a “considerable impact” for pupils.
“The evidence is quite clear that not being taught by someone who sounds and looks like them could affect pupils for things like suspensions and exclusions, the categorisations for special needs, their absence and their happiness, expectations and aspirations,” Gorard said.
A version of the research paper is now available online. It does indeed mention the possibility that having a teacher from a different ethnic group could harm a pupil. But, it also pinpoints the problem with the data on this issue:
...much of the worldwide evidence on this issue is from the US and concerns Black and Hispanic students, and almost all of the best evidence is based on patterns in existing large-scale datasets. There are also reports of in-depth studies, but no experimental studies (or equivalent) have been found that can provide more explicit causal evidence of the impact of disproportionality.
This is a recurring problem with education research and race: too much of it comes from the United States. The racial politics of the US are a source of fascination, perhaps even obsession, for English speakers around the world. However, attempts to map ideas from US race relations onto ethnic tensions in other countries always require an assumption that the details of history are relatively unimportant.
Why diversity alone will not give every pupil access to teachers of the same ethnicity
Assuming that pupils in England would benefit from having a teacher (or teachers) of the same ethnicity, achieving this would require more than just a representative teaching profession. As long as an ethnic minority is small, then the probability of any given pupil having a teacher from that ethnic minority will also be small, even if the teacher population is more representative. The 1.1% of teachers who are black Caribbean might "represent" (or over-represent) the 0.9% of pupils who are black Caribbean, but they are by no means sufficient to ensure that every black Caribbean pupil has a black Caribbean teacher. Ensuring that every ethnic minority is as well represented in the profession as in the pupil population (or perhaps, the entire population) is not going to ensure that every pupil has many teachers, or even a teacher, of the same ethnicity if their ethnic group is small.
We could widen the categories, perhaps saying that every black pupil should have access to black teachers. However, it is far from clear that race, in this broad sense, is the identity that matters to pupils. Will a boy whose great-grandparents were Jamaican think that a Muslim woman from Nigeria "looks and sounds like him" because they both happen to be black? The same complaint could, of course, be made about existing categories. Some of the census categories, such as "white - other", do not categorise people with much in common, but the limits are more obvious when the categories are broader. We either count hugely diverse groups as being the same, or we accept that pupils from small ethnic groups would be “represented” by only a tiny number of teachers, even if there were a perfectly representative teacher population.
Is it possible to provide every pupil with access to teachers of the same ethnicity?
In practice, ethnic minorities (as pupils or teachers) are not evenly distributed across all schools. Schools may recruit both teachers and pupils from the same local community. One would think pupils probably have a better-than-expected chance of being taught by someone from the same ethnic group. But this is another way of saying schools are not all equally diverse, and this is where the problem comes in. If the only concern were to maximise the chance of pupils being taught by somebody of the same ethnicity, the easiest solution would not be to recruit more teachers from underrepresented groups but to change the way ethnic groups are distributed across schools. It would be a reason to segregate schools by ethnicity, for both pupils and teachers. If pupils and teachers were allocated to schools according to their ethnicity it could be guaranteed that almost every pupil would be taught by someone from their ethnic group. The aim of ensuring everyone is taught by people "like them" is best achieved by grouping pupils of similar types in the same school, and employing teachers who are "like" the pupils.
This thought experiment should prompt us to reconsider the notion that ethnic minority pupils primarily need increased access to teachers of the same ethnicity. Greater segregation by race is not an obvious way to benefit ethnic minority pupils, and I would suspect most people reading this would oppose it to some degree. While there are good reasons for wanting the teaching profession to be diverse (in all sorts of ways, not just ethnicity), the desire to match up pupils and teachers of the same ethnicity might have undesirable consequences, and could in practice make schools less diverse.
*Actually, most teachers are women, but let’s not get intersectional about this.