Tony wrote  a thoughtful comment on the previous post about the New York Times article I had tagged on del.icio.us.

He says,

After reading the article “Education Is All in Your Mind,” I have questions for any reader and I have set a goal.

Questions: The article mentioned the “Obama effect.” What exactly is the “Obama effect?” Has anyone really seen or experienced it? Is it really possible that an ethnic group can change it’s mental image of itself simply because a man from that group entered an office? What’s the official name for this effect?

After reading the article, I’ve set a goal to structure my tests such that the student does not have to declare anything about themselves which might be psychologically hindering until after the test is taken, if at all. Steele and Aronsan found that African-American’s did significantly BETTER on exams when they didn’t have to state their race. I recall taking an exam where all the gender, age, ethnic background, etc. was asked at the END of the exam, and I recall the relief it was to answer such simple questions after the difficult ones where completed. Maybe my relief was more psychological about not having to state my age or ethnic group BEFORE I had to answer all the tough questions. Regardless, if I have to administer such an exam, I’ll remember this lesson and do my best to have the stereotyped questions either eliminated or completed after the exam.

Here is some background on the studies mentioned in the article. Two decades ago Stanford Psychologists Claude Steele, PhD, Joshua Aronson, PhD, and Steven Spencer, PhD did a series of experiments that identified a phenomenon they called the “Stereotype Threat Effect.” The study was significant because it refuted  conventional assumptions that it was genetics or cultural differences that led to some students performing poorly on standardized academic tests. Instead, it appeared that negative stereotypes raise inhibiting doubts and high-pressure anxieties in a test-taker’s mind, resulting in poor test scores.  Here is a description of the study:

Steele and Aronson gave Black and White college students a half-hour test using difficult items from the verbal Graduate Record Exam (GRE). In the stereotype-threat condition, they told students the test diagnosed intellectual ability, thus potentially eliciting the stereotype that Blacks are less intelligent than Whites. In the no-stereotype-threat condition, the researchers told students that the test was a problem-solving lab task that said nothing about ability, presumably rendering stereotypes irrelevant. In the stereotype threat condition, Blacks – who were matched with Whites in their group by SAT scores — did less well than Whites. In the no stereotype- threat condition—in which the exact same test was described as a lab task that did not indicate ability—Blacks’ performance rose to match that of equally skilled Whites.

Additional experiments that minimized the stereotype threat endemic to standardized tests also resulted in equal performance. One study found that when students merely recorded their race (presumably making the stereotype salient), and were not told the test was diagnostic of their ability, Blacks still performed worse than Whites.

The stereotype threat effect has been reproduced and the results confirmed numerous times since.  The effect also occurs among women when given math tests. They scored significantly worse than men when they were told the test showed gender differences. There was no difference in scores when the women were told nothing about the test.

All this is background to the  study just completed by Ray Friedman, a professor at Vanderbilt University.

Documenting what Friedman and his co-authors call the “Obama Effect,” the study found the performance gap between black and white Americans in a series of online tests was dramatically reduced during key moments of the 2008 presidential campaign, when Obama’s accomplishments garnered the most national attention.

“Our results document compelling evidence of the power that real-world, in-group role models like Obama can have on members of their racial or ethnic community,” said Friedman.

In the study, tests were administered to a total of 472 participants using questions drawn from Graduate Record Exams (GREs) to assess reading comprehension, analogies and sentence completion. The tests took place at four distinct points over three months during the campaign: two when Obama’s success was less prominent (prior to his acceptance of the nomination and the mid-point between the convention and election day) and two when it garnered the most attention (immediately after his nomination speech and his win of the presidency in November).

The nationwide testing sample of 84 black Americans and 388 white Americans – a proportion equivalent to representation in the overall population – matched for age and education level. It revealed that white participants scored higher than their black peers at the two points in the campaign where Obama’s achievements were least visible. However, during the height of the Obama media frenzy, the performance gap between black and white Americans was effectively eliminated. In addition, researchers pinpointed that black Americans who did not watch Obama’s nomination acceptance speech continued to lag behind their white peers, while those who did view the speech successfully closed the gap.

If you would like to watch a video of Friedman talking about his study, go here

We will talk more about this issue in class at the end of the term when we discuss motivation, but please comment on this post if you want to continue the discussion Tony started.

Reminder: Please remember to take the survey on the previous post before class on Wednesday.