Why Do the Achievement Gaps Exist?
A scholar argues that research on teacher bias can be confusing because research that supports one notion of bias can look unbiased from a different perspective. He offers three different understandings of "race neutrality" to help lessen the confusion in this field of research.
This reports some of the ideas and findings from the following source:
Ferguson, R. F. (1998). Teachers' perceptions and expectations and the black-white test score gap. In C. Jencks and M. Phillips (Eds.),The black-white test score gap(pp. 273-317). Washington, DC: Brookings Institution.
To see other reports that originated from this same citation, go to the bibliography.
Are teachers biased in their perceptions and expectations of black students?
This is not necessarily a question of teachers' intentions. Teachers, black, white, or otherwise, can have different expectations and perceptions of different groups of students. The question is, do they?
Researcher Ronald Ferguson argues that we cannot make sense of the question (much less find an answer) until we are clear about what we mean by "bias." He says that research on teacher bias can be confusing because research that supports one notion of bias can look unbiased from a different perspective.
In order to make sense of this confusion, Ferguson proposes three different understandings of "race neutrality":
Ferguson defines unconditional race neutrality as follows: "Unconditional race neutrality requires that teachers' perceptions, expectations, and behaviors be uncorrelated with students' race." (p. 276)
Ferguson says that research often finds that teachers have different expectations for black students and white students. So, using unconditional race neutrality as the standard, these teachers would be "racially biased."
However, the situation is not so simple, says Ferguson. In real life, where pervasive differences in black and white student performance is the norm, it is not unreasonable to predict that teachers will have differing expectations of students. He likens the situation to one in which the real-life experience of a group of people leads them to conclude that six out of ten times they flip a coin they will come up with heads. If you put these people in a situation where the real (but unknown) probability of flipping heads is 50:50 and ask them to predict the outcome, is their prediction biased? If unconditional neutrality is the benchmark, then biases should be expected, says Ferguson.
Ferguson says that experimental research confirms that teachers believe certain racial stereotypes and that they use these stereotypes in one-time experimental interactions. However, these findings of bias in experimental settings do not tell us:
Do teachers form accurate perceptions of their students' performance? Ferguson says that research shows that teacher estimates of black and white students' performance is roughly similar. This does not tell us, however, whether teachers' expectations are part of the explanation for student performance.
Ferguson says that race neutrality conditioned on past performance means:
"that teachers' perceptions and expectations are unbiased if they are based only on legitimate, observable predictors of performance, such as past grades, test scores, attitudes about school, and beliefs about personal abilities." (p. 280)
The problem is that if past performance is correlated with race, then the benchmark will also be correlated with race. That is, teachers will believe that they are forming expectations and perceptions based on race-neutral characteristics, but this will nonetheless lead them to have systematically lower expectations for one group compared to another.
Basing expectations on past performance is one thing. Basing expectations on an assessment of future potential is another. In this case, teachers would be racially biased if they consistently underestimated the unseen (and unseeable) potential of blacks more than whites.
Because teachers and parents set goals for children based on their estimation of the their potential, underestimating children's potential can lead teachers and parents to set their expectations too low. But, do teachers and parents underestimate the potential for black students more than white students?
Ferguson says that there is no way to test for this kind of bias because there is no reliable estimate of future human potential. However, he says that attitudes about blacks' and whites' "inborn ability to learn" have changed over the past few decades. In 1977, 27% of the respondents to the General Social Survey believed that blacks had less innate learning potential than whites. In 1996, this number fell to 10%. The perception in the general population, however, is somewhat different from the perceptions of experts. Ferguson reports that of 1,020 experts on intelligence surveyed in 1984, only 15% indicated that they did not believe that the black-white difference in intelligence is at least partially genetic.
Ferguson argues that we should not expect a consensus among researchers regarding the importance of racial bias until we have a common benchmark for what counts as "biased."
How do teacher perceptions, expectations, and behaviors relate to the differences in academic achievement between white and black students?
What can be done to break the link between teacher expectations and student performance for black students?
Ferguson essentially carries out a detailed literature review, drawing on a wide range of research studies. His goal is to provide an overview of what is currently known regarding teacher perceptions, expectations, and behaviors and their relationship to black student academic performance.
Copyright © 2007 Learning Point Associates. All rights reserved.