From Sacramento, California, we got news this week of a strange new practice. Several high schools and middle schools have organized race-based “heritage” assemblies not to celebrate Black History Month, or Cesar Chavez day, but to promote improved achievement on state tests.
According to this report in the Sacramento Bee, one high school held several separate assemblies at the same time; students had to choose between African American, Pacific Islander, and Latino-themed rallies. “Last year we scored the highest percentage increase of any group,” the African American students were told.
The school justifies this segregation by citing the need to motivate students to do well on the tests, which otherwise have no effect on their individual status. Many schools are conducting other activities designed to motivate students, including recognition and rewards for improved scores. But the race-based rallies are a new, and disturbing twist, especially in light of some other news this week.
A study was released from Stanford University which examined the impact of the California high school exit exam (CAHSEE). Researchers, led by Sean Reardon, compared students who took the exam in 2005, for whom the results were inconsequential, to those who took it later, when it determined if they would receive a diploma.
They found some fascinating things. First of all, contrary to fears, the exit exam does not seem to have substantially increased the number of students who drop out prior to completing high school. However, the exam did result in a large number of low-achieving students failing to graduate, as a result of their diplomas being withheld.
In the cohort of students not subject to the requirement, a typical student in the bottom quartile of 10th grade achievement had a roughly 45% probability of graduating from his or her original district; in the cohorts subject to the requirement, the same student had a 30% probability of graduation. Our difference‐in‐difference estimates indicate that the CAHSEE requirement caused at least 11 percentage points of the decline in graduation rates (and as much as 15 percentage points). These figures imply that roughly 3.6%‐4.5% of California high school students (roughly 18,000‐22,500 students per year) do not graduate as a result of the high school exit exam policy.
The most interesting finding was that certain groups were affected more than others.
Minority students perform less well on the CAHSEE exam than do white students with the same level of prior and current academic achievement (as measured by 8th, 9th, and 10th grade performance on the California Standards Tests). Female students perform less well on the math CAHSEE test than do boys with the same level of academic achievement. This pattern of results is consistent with ‘stereotype threat’ explanations of test performance.
Stereotype threat is the phenomenon whereby the fear that if one performs poorly on a high‐stakes test it will confirm a negative societal stereotype about one’s group leads to increased test anxiety among negatively stereotyped student groups—minority students and girls, for example—which in turn leads such students to underperform on such tests relative to similarly skilled non‐stereotyped students. The evidence suggests that the high‐stakes nature of the CAHSEE test induces minority students and girls to underperform on the CAHSEE, relative to what their prior academic performance (on low‐stakes tests) would predict. Failure to pass the CAHSEE for these groups leads to lower graduation rates. Thus, it is not that the CAHSEE test is biased, but that the high stakes nature of the exit exam testing situation combined with the presence of negative societal stereotypes about the academic skill of minority students and girls (in math) leads to underperformance by minority students and girls.
These findings have me wondering about the paradoxical affect of one of the key elements of No Child Left Behind – the highlighting of racial subgroups. This “spotlight” on the achievement gap is supposed to force schools to do a better job with historically underserved groups. What happens when this spotlight is shined on the gap in the context of race-based motivational rallies? It seems to me that this might have the effect of strengthening stereotype threat, by reminding students of the poor standing of their subgroup. I also wonder about the nature of the achievement gap itself, and how much of it might be attributed to stereotype threat.
Lastly, I can only assume that there are few, if any, white students at these schools. A rally to celebrate white test score supremacy is simply too weird to imagine.
Update: An interesting resource for anyone interested in exploring the concept of stereotype threat: http://www.reducingstereotypethreat.org/
What do you think?
Is anyone else experiencing racially based rallies for test motivation?
How do you think this sort of emphasis on the different performances of racial subgroups will affect our students?
Mightn’t this sort of thing intensify stereotype threat, given that African American and Latino performance is usually lower than that of other groups?
image: Creative Commons: thesituationist.wordpress.com/2007/12/