It's Not a Race (Thing)

nm"We know that, historically, African American and Hispanic students do not score as well on the NAEP (National Assessment of Educational Progress) as their white and Asian peers."

What do preschool teachers know about race and poverty that educational researchers don't? How many times have we heard a statement like this as educators? For some reason - even though it may be true statistically - it doesn't seem to make sense because the statement itself obscures the real issues. It doesn't necessarily matter that many African American students score lower on these types of tests because not all of them do.

These types of statements contribute to bias against African American students because they connect the problem -poor achievement- to the descriptor instead of the cause. If a researcher said, "We know that, historically African American and Hispanic students do not score as well on the ______ test as their white and Asian peers because of the inherent racial bias of the _____ test," only then should a child's race be an important part of the discussion.

Often We have these types of discussions in my doctoral classes. This semester, I am taking a measurement in educational research course. It is meant to prepare us, the budding researchers, for our dissertation and future life as an educational "expert" ie Ph.D. This week the issue of race came up. Specifically, the professor mentioned the "achievement gap" in order to push the conversation into uncomfortable territory. I put "achievement gap" in quotes intentionally because it doesn't seem worth talking about outside of the context of test validity and reliability. Usually after a statement about testing and race someone will say, "That is is because tests are biased." Then another researcher might say, "But, there is more of a correlation between socio-economic status and test scores." Even though the statement is true, it doesn't contribute to a deeper understanding of the issue. It uses descriptors in an almost causal relationship to data. The "achievement gap" assumes that the tests are valid and the scores are reliable indicators of student achievement. In actuality, they may only be valid for students from the dominate white culture. By switching the argument to money you just change the context of the discrimination. By saying, "Well actually its about money not race." researchers are able to wash their hands of the issue because there is no way that schools can impact how rich or poor a kid is, when they know there is no reason a child's race should affect a test score either.

Many preschool teachers know that the only thing that matters in a child's educational trajectory is the frequency and types of language interactions and experiences children have in their early years. Hart & Risley pointed this out in Meaningful Differences. They found the strongest influence on a student's vocabulary was the types of talk children engaged in as an infant, toddler, and preschooler. The number of words that children knew at age 3 was found to be predictive of achievement at age 9. However, some children of professional parents, who didn't spend much time talking with their kids had similar average minutes of interaction as some welfare and working class families. Across the board the researchers found race was not found to influence the vocabulary of children at all. Hart & Risley "saw quality added to interactions when we saw parents talking to their children beyond what was necessary to manage or provide care." So yes, poverty is a factor in the development of children's vocabulary because the essence of poverty is a struggle to survive. Poverty requires parents to focus on the day-to-day survival of their family.


If we were to reframe the achievement as a "language development" gap we we might get closer to the "truth" about why some kids score higher and lower on standards based tests. Yesterday I heard a woman in a supermarket tell her 9 month old that her cookie looked "scrumptious". Does it matter what race the woman was? Of course not, I know that her child will likely hear many words like "scrumptious" in her early life. A more correct way of describing the achievement gap might be to discriminate between students with few language experiences (FLE), substantial language experiences (SLE), and many language experiences (MLE). Is this practical? No, but at least it would be true. By continuing to attach race and class to achievement we support the assertion that they are predictive when they aren't, they are just an easy way to sort data.