A Social Psychological Perspective on the Achievement Gap in Standardized Test Performance between White and Minority Students: Implications for Assessment

Abstract

The National Study Group for the Development of Affirmative Academic Ability (Bennett, Bridglall et al., 2007) examined national trends in standardized tests for White, Black and Hispanic students in the United States. The data selected for examination was broadly inclusive by design, and included middle- and high-school samples, a range of years, and multiple academic subjects. More specifically, the group examined representative National Assessment of Education Progress (NAEP) scores for eighth- and twelfth-grade students between the years 1996 and 2002 in science, math, and reading. The group also examined SAT scores for Black and White students between the years of 1996 and 2003.For NAEP scores, the data revealed that across years, subjects and grades-in other words, in every single comparison conducted-Black and Hispanic students scored significantly lower than their White counterparts. For SAT scores, Black-White differences across eight years ranged from 92 to 108 points. The consistency of these achievement differences is as striking as it is sobering, and made even more so by the fact that they replicate decades of data demonstrating achievement gaps in standardized tests among these groups (e.g., College Board, 2010). These achievement gaps are the central focus of this article.Scholars have debated the source and meaning of these academic achievement gaps intensely. While psychologists tend to interpret these achievement gaps as examples of test bias and unfairness due to the fact that one should not, a priori, have reason to expect factors such as race to influence test scores (e.g., Helms, 2006, 2009; Walton & Spencer, 2009), the testing field rejects the existence of such differences as evidence of bias per se (Sackett, Bomeman, & Connelly, 2008,2009). Sackett and colleagues (2008) put forward the proposition that, in fact, such differences may reflect constructrelevant variance: ethnic group membership, for example, may be associated with undervaluing academic achievement and/or spending less time studying. In the words of Sackett and associates (2008),It is precisely because of these potential alternative explanations that the dominant view in the testing field rejects the position that a finding of a relationship between race... and test scores can be directly interpreted as signaling bias or unfairness, (p. 223)This position reflects the strong assumption that if a student is unmotivated or spends less time studying than his or her peers, there is a lack of personal responsibility and, as such, is fairly reflected on the standardized test. Sackett and others (2009) expanded on this notion of personal responsibility, arguing[We] find it hard to imagine any instructor responding positively to a student who says, 'I chose not to read the assigned material covered on the test, so it's unfair that you penalize me for my incorrect responses.' (p. 286)This author reviews research suggesting that, even when it came to processes that reflect personal choice-such as time spent studying or becoming invested in academics-the interpersonal milieu surrounding the student has a profound influence on the development of these processes. The aim is to shed light on how certain classes of students, in particular ethnic minority students, are systematically exposed to toxic messages about their abilities that force many to minimize the importance of academics (Major et al., 1998; Ogbu, 1991; Steele, 1997). This, in turn, suggests that academic achievement gaps, rather than reflecting test bias, may be more accurately described as reflecting societal bias. The author concludes with some possible directions that fair testing may take in the context of a cultural context that does not support all students to the same degree.This article explores the issue of academic motivation and performance from a social psychological perspective. It focuses specifically on the academic achievement gap between White and minority students, not only because it represents one of the widest and most persistent gaps in testing, but also because social-psychological research in this area has, to date, concentrated most heavily on this dimension. …