Student Achievement

Shrinking the Achievement Gap One Word at a Time

By Holly Kurtz — May 27, 2014 5 min read

An African-American 7th grader is sitting in her homeroom or English class in Madison, Wis. It is the fall of 2011. Her teacher passes out a worksheet listing 11 things that people value. She picks one of these things, maybe “being with friends and family” or perhaps “enjoying sports,” and writes about why it is important to her. Something similar happens again that autumn, and then again, two or three more times in spring.

At the end of the school year, the girl’s GPA goes up. In fact, the GPA gap between whites or Asians and blacks or Hispanics shrinks by 12.5 percent at her school, which happens to be among seven schools in her district with the smallest percentage of minority students and the biggest disparities between the performance of whites or Asians and blacks or Hispanics on the state’s standardized exams.

It may sound like magic but it’s not. It is a summary of an intervention described in an article in the current issue of the peer-reviewed journal, Sociology of Education. It’s also one of the latest results in a line of research that has found that simple, brief, and inexpensive interventions can sometimes increase achievement.

The intervention described in the Sociology of Education article is based on the work of Stanford University professor Geoffrey L.Cohen and his colleagues, whose research on similar interventions has found that they can narrow the black-white achievement gap by 40 percent. The Madison study replicates and scales up the interventions described by Cohen and his colleagues. These interventions fall into the category of so-called “self-affirmation exercises” that aim to “bolster the perceived integrity of the self, one’s overall self-image as competent, effective, and able to control important outcomes.’'

“For instance,” write Paul Hanselman and his co-authors in the Sociology of Education article, “a potentially susceptible student who has reflected on the value of her sports participation may be less likely to interpret negative teacher feedback as a threat to her sense of belonging in the academic domain, therefore heading off a negative performance-evaluation feedback loop.”

Self-affirmation exercises are not designed to change people’s beliefs about themselves, said Hanselman, who is a doctoral candidate in sociology at the University of Wisconsin-Madison. Instead, they are supposed to “remind” students that they are more than just students.

“The idea is that an individual is not just a student in school, but also a sister, an athlete, and a friend, for instance,” said Hanselman. “Social psychologists point out that diversity in self identity helps individuals deal with stressful situations when they are subject to negative stereotypes. In middle school, this may be especially true at the beginning of the school year and during periods of academic testing, which were the two times throughout the year we targeted.”

What’s especially interesting and novel about the article by Hanselman and his co-authors is that it considers the outcomes of an individualized, psychological intervention (that writing exercise) in the context of the broader social environment of schools. It does so by examining how the effects of the intervention vary depending on the overall racial composition of the school and the size of its achievement gap.

To get at these issues, researchers sorted 11 Madison middle schools into two main groups: The first group contained four schools with higher percentages of minority students (defined as Hispanics and blacks) and smaller achievement gaps on the state’s standardized exams.The second group contained seven schools with fewer minority students and bigger achievement gaps. Based on past research, the study’s authors theorized that students at the second group of schools would be more likely to encounter “social identity threats” that occur when a person’s self-image is challenged because he or she is a member of a group with a negative academic reputation. Because social identity threats often operate on a subconscious level, students may be affected by them even if they do not personally buy into the negative stereotypes that are their basis.

A total of 910 7th graders participated in the intervention. Half were randomly selected to do the writing exercise described in the first paragraph of this piece. The rest were assigned a different writing exercise that was not designed to be self-affirming. Regular classroom teachers administered the exercises. They did not tell the students that it was a research study. Nor were they told who got the “treatment” and who did not.

As they predicted, researchers found that the self-affirmation exercises had a more dramatic impact on the grade-based achievement gaps in the seven high-threat schools, where it fell by 12.5 percent. The grade-based achievement gap did not really change in the four low-threat schools. Nor did the intervention affect the GPAs of white and Asian students in either setting.

The results were more mixed when researchers turned to test scores in language usage, reading, and math. As with grades, the exercises did not affect the test scores of white and Asian students. But they were associated with smaller achievement gaps in high-threat schools in all three subjects. The math achievement gap also shrank in low-threat schools. However, unlike with GPAs, the differences between high- and low-threat schools were only statistically significant in language usage.

“One possible explanation for the differences across outcomes is that social identity threat and self-affirmation may work differently across different academic domains,” Hanselman and his co-authors proposed. “For instance, social identity threats related to English language may be more sensitive to social context. We find additional suggestive evidence for this possibility in supplemental analyses of subject-specific grades: The school context by affirmation interaction among black and Hispanic students is strongest in English classes and not statistically significant in mathematics classes alone.”

The findings are promising, according to University of Wisconsin education professor Geoffrey Borman. Borman co-wrote the article. He also directs the Madison Writing and Achievement Project, which conducted the underlying research.

“Mind-set interventions of this type have been shown, in a select number of classrooms, to have powerful impacts on reducing inequalities in student performance,” Borman said. “Our work, which scales up these efforts across an entire school district, shows how and under what circumstances, the interventions can make a difference. Our work suggests that mind-set interventions targeted at affirming student values may not work in all school contexts, but in the right contexts, these brief and inexpensive strategies can help narrow existing gaps more substantially than many other proven educational interventions costing far more in terms of both time and resources.”

A version of this news article first appeared in the Inside School Research blog.