For those struggling with the crippling effect that TikTok has had on their attention span (or those who are just better listeners than readers), here's a video overview of the project I completed analyzing the bias in standardized testing.
Otherwise, here we go! Have you ever taken a test and looked at a question and said, "who in the world would know this?" How about "I've never heard/seen that word in my life"? This is something all too familiar to many students in the American education system -- especially minorities and poor people.
Growing up in the United States, the concept of measuring progress, intelligence, and aptitude through means of standardized testing is far from a foreign one. Standardized tests are those “in which the questions, the scoring procedures, and the interpretation of results are consistent and which are administered and scored in a manner allowing comparisons to be made across individuals and groups” (Benjamin et al., 2012, as cited in Bazemore-Jones et al. 2017). Most Americans can recall taking standardized tests every year from elementary to high school. These tests come with many names and acronyms - IQ tests, SATs, ACTs, and Advanced Placement tests, just to name a few. These tests also come with high stakes attached to them. Test scores are used to quantify intelligence, measure learning and understanding, categorize people, and play gatekeepers to higher education.
The fact that these tests are called "standardized" implies that they are one-size-fits-all and would be accurate in getting generalized information based on scores. Unfortunately, this is not the case. Some test questions contain words, phrases, and concepts that are not common to every race, culture, or socioeconomic class. Blatantly ignoring the cultural differences that exist and including questions that do not appeal to everyone have detrimental effects on the development and success of future generations of minorities and less fortunate people. Also, minority students and students who come from families earning lower income do not have the same access to test prep and tutoring that more affluent students do. This negatively affects these students’ scores and, in turn, negatively affects their entire lives. If standardized tests are going to be a constant in the years to come, there must be a way to make them as fair as possible.
The Long (and Complicated) History of Standardized Testing
Since 1840, standardized testing has been a mainstay in the American educational system.
Prior to the Civil War, schools used mandated written exams to evaluate a student’s progress within the curriculum.
In 1890, as an alternative to having separate tests for each school, Harvard President Charles William Eliot presented a system of examinations that would be accepted by a wide range of colleges and universities throughout the U.S.
The 20th century gave rise to the concept of IQ testing as sectors such as the military and schools became interested in quantitatively measuring intelligence.
During World War I, standardized tests helped place 1.5 million soldiers in units segregated by race and test scores.
In 1925, a United States Bureau of Education survey showed that these intelligence and achievement tests were being used more frequently to classify students.
The No Child Left Behind Act of 2001 (NCLB) expanded state-mandated standardized testing in order to assess the performance of schools themselves.
The Every Student Succeeds Act of 2015 counteracted NCLB’s effects on testing by reducing standardized tests and separating testing and high-stakes decision-making.
Tests such as the PSAT, SAT, ACT, and Advanced Placement tests, are all a part of the onslaught of standardized testing that students endure during their educational careers and are still cornerstones of college admissions. The concept of racial, cultural, and socioeconomic bias seeping into the education system through standardized testing is in line with the racially, culturally, and socioeconomically divided history of the country. It is well documented that the concept of white supremacy has been a staple throughout and that these thoughts and ideas have shaped the minds of the general public and those in positions of power. The psychometric movement’s founding fathers believed that test scores provided evidence of the intellectual inferiority of dark-skinned people (Guthrie, 2003, p. 56). Princeton psychologist Carl Brigham wrote that African-Americans were at the end of the racial, ethnic, and cultural spectrum (Brigham, 1923). He was also influential in developing aptitude tests for the U.S. Army, the SAT, and the Advanced Placement test. These tests became a way for psychologists who believed in eugenics - the science of improving a human population by controlled breeding to increase the occurrence of desirable heritable characteristics - to peddle these ethnocentric beliefs under the guise of scientific experimentation. This way of thinking gave birth to the rise of Nazism in Germany. The same ideology that influenced Adolf Hitler and the Nazi Party is the same ideology that influenced the creation of tests that are still used today. Now let's take a second and imagine... The same ideas and thoughts that gave rise to the toothbrush moustache also played a part in the "harmless" questions kids have to answer every year in school.
Some people ignore the importance of cultural awareness and sensitivity in these tests and the education system as a whole. This viewpoint fails to recognize the necessity of understanding differences between cultures. However, these people and organizations still believe that standardized testing is essential and perceived bias is not a proper reason to cut it. When examining why standardized testing is viewed as essential regardless of the bias present, many people consider these tests as fair assessments simply because the same questions are being asked to a large number of people. According to Aaron Churchill of the Fordham Institute (2015), standardized testing presents a sense of objectivity that may not be reflected in normal grading practices due to “differences in teachers, favoritism toward certain students and non-achievement factors like classroom behavior, participation, or attendance.” Standardized tests are necessary because “[a pick-your-own-assessment policy] would undermine the comparability principle of statewide testing.”
Also, the standardized testing market is too lucrative to give up. The National Board on Educational Testing and Public Policy at Boston College reported that the value of the 5 standardized testing market was anywhere between $400 million and $700 million (Stauffer, 2017) and that the “testing marketplace is large and continuing to grow” (Clarke et al., 2001). In 2012, a report by the Brookings Institution found $669 million in direct annual spending on assessments in 45 states, or $27 per student. When you add administrative costs involved in tests, the total spent on testing rises to $1100 per student (Chingos, 2012). As of 2015, parents spent $13.1 billion on test prep, including preparation and tutoring (Strauss, 2015b).
Standardized testing is often viewed as a necessary evil in a less-than-perfect world. In recent years, there has been a significant increase in the number of people calling for the end of standardized testing due to the amount of bias and issues that can and have been introduced surrounding the tests themselves. According to Harold Berlak (2020), “standardized testing perpetuates institutionalized racism and contributes to the achievement gap between whites and minorities.” Valerie Strauss (2015a) gives a lengthy list of problems with standardized testing, including that they provide little to no useful feedback for teachers, penalize non-standard ways of thinking, and make time a significant factor in scoring. In a report to Congress, the National Academy of Sciences (2011) said that the use of standardized tests “has not increased student achievement.” The questions on some standardized tests can either be answered differently or be deemed unanswerable due to cultural and language differences that exist naturally in the world. Every child does not have the same upbringing or access to information. There are certain words, phrases, and concepts that are not a part of every cultural experience. This does not give everyone the same opportunity to succeed on these tests.
Minority students (including those in which English is not their primary language) and students that belong to lower-income families are at a disadvantage. According to College Board (2016), in the SATs taken during the 2014-15 6 school year (the last year family income appeared on the report), the lowest average scores were those with less than $20,000 in family income (total average score - 1314), and the highest averages were those with more than $200,000 in income (total average score - 1717). In 2019, combined SAT scores for Asian and White students averaged over 1100, while all other groups averaged below 1000 (College Board, 2019).
Less fortunate students, minority students, and their families are tired of getting the short end of the stick from the educational system. There is growing frustration with unfair tests being recognized as the final verdict on intelligence levels. The profound effects of the overt racism and classism embedded within the figures associated with the development and furthering of standardized testing have transformed into more subtle hints of racial, cultural, and socioeconomic bias within the tests themselves. Thus, racism, classism, ethnocentrism, and implicit bias are barriers to intercultural communication between the stakeholders. Also, anxiety is a massive barrier to communication because, with the high stakes associated with test scores, students have been known to experience anxiety. There are immense ramifications that stem from the constant biases encountered by students throughout the history of the American education system. Therefore, it is imperative that a solution arises that would eliminate or limit the amount of bias present within the tests that every child has to take. The concept of standardized testing has always been oxymoronic in a sense, and the educational system and society itself would benefit significantly from taking away a one-size-fits-all approach that does not, in fact, fit all.
The use of standardized testing has been a constant in the United States’ education system. In conjunction with this, implicit bias has been an ongoing problem in standardized testing. Questions on these tests often come with the assumption that every student has lived the same life and has been exposed to the same things. This method of thinking ignores cultural, racial, and socioeconomic differences between people in America. The solutions proposed below will lessen the amount of division felt within the country and give a more accurate measurement of the metrics assessed in testing (IQ, learning and retention ability, etc.).
Possible Solutions
One solution to prevent bias in standardized testing is to eliminate the high stakes customarily attached to the test scores. Test scores are often viewed as the most important thing in school, above the actual learning, understanding, and retention of information. This mindset is drilled into the heads of students, parents, teachers, and school officials. Test scores can make or break a student’s educational career because they are commonly used as gatekeepers to the next level of education. An extreme focus on testing creates disengaged students, putting many at risk of joining the “school-to-prison pipeline” (FairTest, 2010). As of May 2019, eleven states still require students to pass a standardized test to graduate (FairTest, 2019). Many factors contribute to some students not having a fair opportunity to learn the material for the tests, including attending poorly-funded schools with large class sizes and using inadequate books, libraries, laboratories, computers, and other facilities. Since some of these students come from low-income families, other personal issues may arise - including problems with housing, nutrition, and healthcare - that would negatively affect these students’ test scores. High-stakes tests tend to penalize students for things beyond their control, so the level of importance attached to these raw scores should be lessened. One way to eliminate these high stakes would be to make all colleges and universities test-optional. Due to the COVID-19 pandemic, many schools have ceased requiring admissions tests. This shows that making them test-optional as a whole is a possibility, and it should be explored.
Another solution is to ensure that the questions that end up on standardized tests only contain words, phrases, and concepts familiar to all cultures, races, and income levels. Suppose standardized testing continues to be how the American education system measures a student’s intelligence and performance. In that case, it is only right that each student has an equal opportunity to perform well on them. The education system must be able to adapt to the ever-changing society. Society, as a whole, has come a long way in trying to be more understanding when it comes to cultural differences. Statistical measures (differential item functioning) are already employed to show the extent to which a test might measure different abilities for members of different subgroups, but the collection of data could be extended. During quality testing, the United States Department of Education could send a yearly survey to the homes of students from all races, cultures, and income levels to gauge their understanding of potentially unfamiliar words, phrases, and concepts that may show up in future tests. If a large group of people (or even multiple groups of people) show a lesser understanding of said words, phrases, or concepts, then the question is scrapped. This way, having implicit bias within the test questions is lowered because there was communication with different types of people during the question formulation process.
A third solution surrounds simply ensuring that a diverse group of people formulates the standardized test questions and puts them together. There must be different types of people involved in the test development process. If the people responsible for coming up with test questions are from all different backgrounds, this will help limit the amount of bias that would make its way into these standardized tests.
The last solution would be to remove standardized testing from the curriculum altogether. Taking out standardized testing would alleviate a lot of stress that might come with the department’s aim to please everyone. Some countries ranked ahead of the United States in aspects of education do not have standardized tests. Finland, for example, is ranked ahead of the U.S. in reading, math, and science (Education Rankings By Country 2021, 2021). They do not have standardized tests, but rather teachers are trained to assess students using independent tests they create themselves (Partanen, 2017). This could easily continue to be the norm for the American education system, but - if there must be a test - there are plenty of performance-based alternatives that accurately measure learning, understanding, and ability to apply learned concepts.
Commenti