Exams and equity

How cultural bias in standardized testing holds students back

It’s difficult not to be familiar with the idea of standardized testing. Ontarians might recoil at the acronyms EQAO and OSSLT, and most North Americans will recognize the SATs, LSATs, and MCATs. Standardized tests are widely administered written evaluations, characterized by predetermined, consistent questions and scoring. What most people think of as a “test” is a standardized evaluation—classroom quizzes, university finals, IQ tests, and entrance exams are among the major categories.  

As tools for teaching, census-like means to assess average performance and guidelines for school curricula, standardized tests have many merits—but they’ve faced criticism from students and teachers alike. Sure, every high-schooler dreads final exams and college applications, and grading essay questions is certainly a source of grief for many educators, but the problem with standardized testing isn’t the stress. 

Standardization can mean something much more consequential to English language learners, students of colour, and students from underprivileged economic backgrounds. Racial and linguistic minorities, as well as those in lower income brackets, consistently underachieve compared to white and higher-income students—and it’s not a matter of coincidence or inherently poorer aptitude. 

Let’s look at some numbers from the SATs, one of the most widespread standardized college admission tests in the United States. In 2013, self-identified Black high school seniors scored, on average, 20 percent lower on the SATs than those who identified as white, while Hispanic and Latin Americans scored 14 percent lower. Students with families in the lowest income bracket scored the lowest, and students in the highest bracket predictably scored the highest, with a 23 percent gulf between them. 

From childhood to university, this gap in achievement tails those from disadvantaged backgrounds. And the problem, again, isn’t the abilities of these students; it’s not about their drive, their potential, or their willingness to learn—it’s about the tests themselves.  

It’s a simple and unpleasant truth that standardized tests, from high school evaluations to national college entrance exams, are written with a white, middle-class audience in mind. The cultural contexts in which most standardized tests are based often clash with those of minority students. The use of certain vocabulary, for example, or references to certain cultural events, might reflect little consideration for cultural and linguistic differences.  

Perhaps the most infamous example of cultural bias in North American standardized testing came in the 1990s, with an SAT question whose correct answer included the word regatta—referring to a type of yacht racing competition. More recently, an SAT essay question controversially asked respondents to analyze whether reality TV is harmful. This kind of culturally rooted question might be an intuitive concept to a middle-class demographic with the time, means, and social background to be familiar with reality TV, but the same can’t necessarily be said for those from different cultures or socioeconomic circumstances. 

Even stepping back from the SATs, it’s clear that standardized evaluations of all varieties—IQ tests, high school literacy tests, even word problems on math exams—can be and historically have been alienating to students without implicitly expected roots in popular culture, vernacular, and customs. Yachts and reality TV only scratch the surface of countless examples of this alienation: questions that require knowledge of law, medicine, sports, history, and so on all run the risk of penalizing children who might not be familiar with these fields from conversations at home. 

The problem expands when you consider unequal access to academic resources; wealthy students can afford tutors and online services, and they’re more likely to have well-educated family members to help them study and practice for tests.  

The fact of the matter is that testing is biased. But standardized tests don’t have to be part of the long list of ways in which white students with family backgrounds of academic success and economic stability have another leg up on everyone else. There are alternatives, and they don’t necessarily sacrifice the benefits of standardization.  

Demographic information can and should be collected and analyzed to find where gaps in achievement are coming from and focus on providing help to students who need it. Test questions should be written under a critical lens, with review from those experienced with racial, linguistic, or economic bias where possible. In some contexts, standardization could be partially or even entirely de-emphasized as a measure of learning. Instead, smaller samples of students could be tested on a random basis to determine a group’s strengths and weaknesses. 

There’s no denying that standardized tests are a useful tool for census, education, and evaluation, but it’s also hard to ignore the fact that they are often unfair to students in disadvantaged circumstances. Going forward, it should not only be a responsibility, but a gratification for educators and administrators to make standardized tests more equitable and useful for all students—not just the most privileged one