Most of us had a favourite teacher.
Perhaps they made an extra effort in explaining difficult concepts, gave you undeserved praise and acknowledgement, or were kind and supportive during a difficult time.
Perhaps they could see something special in the young you that you had not yet discovered yourself: they expected more of you than what was warranted by your performance.
New research now shows that we should be thankful to those teachers.
Three US economists, in a National Bureau of Economic Research (NBER) working paper, show that teacher expectations matter in explaining students’ educational outcomes.
The idea that teachers have a powerful influence over their students is, of course, not new.
The overt or subtle messages that teachers give to students about their potential should, theoretically, have large and long-lasting effects; a student that must frequently hear how terrible she is at math is likely to perform much poorer on a math test compared to a student of the same ability who frequently receives positive encouragement.
It was always difficult to prove this link between teacher expectations and student outcomes causally.
Perhaps the teacher could identify math potential quite early and was only signalling to the two students their respective advantages and disadvantages.
The three authors, however, find a novel way to show that teacher expectations do matter.
They use a nationally representative sample of teacher expectations to show, first, that teachers are often overly optimistic of their students’ education potential.
They then show that this optimism pays off.
Because two teachers report the expectations for each student in their dataset, the authors can use the variation in the teachers’ evaluation as a way to causally measure the effect of teacher expectations.
The students that receive a higher rating from teachers, controlling for other observables and the other teacher rating, also seem to achieve higher educational outcomes: they are more likely to go on to college, for example.
Teacher expectations, the authors argue, become a self-fulfilling prophecy.
A too optimistic rating, given the student’s innate ability, may push a student beyond the level he was ‘supposed’ to perform at.There is also a dark side to this.
More realistic expectations of students’ educational prospects may actually be counterproductive.
A student whose teacher predicts that they are unlikely to finish high school is less likely to finish high school than a student with similar characteristics, but with a more optimistic prediction.
This may result in discrimination, as male teachers may have less optimistic predictions for female students, and white teachers may have less optimistic predictions for black students.
How to solve these (often implicit) biases is a difficult question.
One answer may be to focus on appointments from a more diverse teacher pool.
Appointing teachers from the same socio-economic or ethnic background may have other positive consequences too.
A second NBER working paper recently published by a team of five authors, shows that black students in the Tennessee school system who were randomly assigned in Grade 3 to a black teacher are 7% more likely to graduate from high school, and 13% more likely to enrol in college than classmates who had a white teacher.
They ascribe this to the ‘role model effect’. Black teachers, they argue, “provide a crucial signal that leads black students to update their beliefs about the returns on effort”.
In other words, by the mere fact of being in the classroom, black teachers show black students what is possible.
In a series of tweets, Trevon Logan, professor of economics at Ohio State University, questions this ‘role model’ interpretation.
The findings “implicitly denies any agency to the teacher or the student”. It presumes that standing in front of a class with a degree is sufficient to alter a student’s behaviour.
“Black teachers bring skills and pedagogical approaches to their work. Removing teacher behaviour stops us from looking at what teachers do to and for students. It also denies the extensive literature in black education studies about unique strategies employed by black teachers. And it suggests that white teachers need not improve with regards to black students.”
Logan is correct that there may be multiple reasons why teachers that reflect the same socio-economic or ethnic background are more likely to get better results from their students.
This may be due to the role model effect, but it’s probably more likely a combination of more relevant teaching methods and greater investments in time and attention.
It may, as the first paper shows, also simply be that these teachers have greater (perhaps even unrealistic) expectations for these students.These results are likely to have implications beyond the school classroom:
for those of us who teach at universities, work as human resource professionals, or lead large teams, the results of these studies suggest that our students or junior colleagues are likely to gain from our optimistic expectations about their prospects, rather than a more realistic or even pessimistic outlook.
And perhaps it is time to pick up the phone and thank that primary school teacher who had, most likely unknowingly, overly-optimistic expectations about your future career prospects.
Johan Fourie is associate professor in economics at Stellenbosch University.