Testing is an institution which affects the lives of American people in a profound way. Early in their school years, children are classed and categorized based on standardized tests. In high school, such tests become even more life-determining as students largely base their decisions regarding higher education on the results of their test scores. Furthermore, decisions about graduate and professional schools are to a degree based on test results. Tests are also used to guide employment decisions and to determine professional advancement in certain careers. Indeed, testing is all-encompassing and powerful in its influence on modern American life. The problem, however, is that testing is also difficult to control and check even when its effects appears to be unfair and/or counterproductive to the welfare of students and teachers. In this light, this paper argues that, because of the unfairness in the tests themselves and in the use of tests and test scores, standardized testing contributes to the deterioration of the American education.
Basically, test bias exists when a test does not measure the same dimensions of achievement across different group (Green, 1975). These biases can be due to content factors, norms, and testing situation. The major issue in a discussion of content bias is the degree to which test items may be more relevant to one group of students than to another – for example whether it is relevant to middle socioeconomic status (SES) or to low SES. One may ask: Who writes the test items” What groups the items are tried out on? One may also point out the extent of dialect differences between the middle SES child and the lower SES child, and the existence of particular test items seen by many authorities to be difficult but irrelevant to the purported objective of the testing.
Thus it becomes clear that standardized testing is not the solution for all groups. Particularly, African-Americans, whose average tests scores are lower than those of other groups, consider standardized testing as a major obstacle to their admission. Some schools and universities that are eager to have a fair representation of minority students either set different requirements for standardized test scores or give more weight to other factors, such as overcoming adversity, in coming up with admissions decisions.
Another test bias concerns inappropriate test norms, in particular the so-called national norms. In using these norms, the performance of a student is compared with the norm arrived at through the standardization process. The concern here is how the samples of students comprising national norm are selected. While no norm will exactly fit any community, the perils of misinterpretation increase exponentially if the sample used in establishing the national norm does not include students from the major regional, racial, ethnic, community type and income groups.
Moreover, if the samples used in establishing the national norm misrepresents or leaves out some group, then the national norm is a misrepresentation of the population. While there may be occasions when school systems must make group comparisons for evaluation purposes, there should never be a time when it is necessary to compare individual students to national norms. In addition, there is also the bias caused by the testing situation. Variables in this bias include test-wiseness, time allowed for testing, item type, answer sheet format, examiner characteristics, achievement motivation, and perceived use of the test results.
Although standardized test can be relatively unbiased, it can be used unfairly. In addition to the evident problems of questionable predictive validity of aptitude tests, there are also technical problems with respect to their predictive validity. One problem is that course grades are unreliable measures of performance, since grades may fluctuate because of variables not related to how well a student actually performs. With criterion variables that may be such poor indices of actual performance, it is not clear that aptitude tests’ validity as predictors can be established. Another problem is that the correlations which establish the predictive validity of aptitude tests are attenuated because of the restriction of range.
There are many other factors that affect performance in standard testing. One of these factors is the anxiety (or the fear of failure) one faces in taking the test. Thus, test performance is motivated by fear of failure, which might lead to lower test performance. Another factor is the level of willingness of students to guess on multiple choice test items. Typically, risk takers gain an advantage by their willingness to guess, but this is because most guesses are informed guesses.
Standardized testing in the American education system has been criticized by many scholars. For example, Hilliard (2000) criticizes standardized testing as invalid, arguing that the misplaced focus on testing has allowed people to “ignore the well-documented importance of the quality of teaching as a key factor in student achievement” (p. 293). The author states that minorities and the poor have suffered from testing misuse, and he calls into question the predictive value of standardized tests without regard for access to high-quality teaching. Hillard (2000) reminds that teachers are the true agents of change in the education field.
In addition, Kohn (2000) states that no other nation in the world subjects its children to the quantity of testing that the United States does. According to him, the main problem with most standardized testing is that they fail to assess the skills and dispositions that matter most. Kohn (2000) enumerates the most menacing features of standardized testing: norm-referenced tests, time limits, multiple choice format, tests given to young children, and frequent testing. Possible costs of such tests include cheating, teachers turning against students, an atmosphere that fosters defensiveness and competition, experienced educators who feel compelled to leave the profession, and a narrowing of the conversation about education (Kohn, 2000).
According to Miner (2000), despite growing criticism from scholars, educators, and many parents, standardized testing is not likely to go away soon. The author blames politicians, corporate leaders, and policymakers for embracing reforms that heavily rely on standardized tests as a gauge of academic quality. Miner (2000) contends that this dependence has overwhelming costs: “It leads to a dumbed-down curriculum that values rote memorization over in-depth thinking, exacerbates inequalities for low-income students and students of color, and undermines true accountability among schools, parents, and community.” (p. 40)
As stated earlier, one argument against standardized testing is that standardization reduces the quality and quantity of what is taught and learned in schools. This direct negative effect of this kind of testing is the overwhelming evidence of schools where the imposition of standardized controls reduced the scope and quality of course content. Another immediate negative impact of standardized testing are the diminished the role of teachers and the distancing of students from active learning.
The long-term effects of standardized testing are even more damaging. In the long run, it creates inequities, widening the gap between the quality of education for poor and minority youth and that of more privileged students. The discriminatory effects of standardized testing are immediately evident in the reduction in both the quality and quantity of educational content for students who have historically scored low on standardized assessments. In due course, the longer standardized controls are in place, the wider the gap becomes as the system of testing and test preparation comes to substitute in minority schools for the curriculum available to more privileged students.
The educational costs of standardized testing are high. Standardization, when used as an instrument of accountability in a highly centralized, aligned system, masks and thereby perpetuates inequalities by its misleading indicators and aggregations and by directing resources toward maintenance of the accountability system rather than toward relieving the inequities. Standardized testing, which has proven to be so damaging to the social construction of knowledge in American classrooms, threatens to weaken the social contract by which education strengthens democratic governance. These criticisms of standardized testing might be less compelling if the problems of educational standardization are malfunctions of the system. However, this is not the case in many settings – the problems created by standardize testing (for example, the de-skilling of teachers, the creation of inequities, the restratification of access to education, and the de-democratization of American education) are not malfunctions in the system. They are the logical consequences of the system when it is working.
The American people need to ask about the potential costs of standardized testing to students and schools – the effect on the broader curriculum; the excessive classroom time spent on cramming for the test; and the risk that students perceived as low-achieving will be assigned to special education programs so they do not participate in standardized testing (Futrell and Brown, 2000). Furthermore, Americans must know the risks to students of using standardized test results as criteria for promotion and graduation and the fairness to building level administrators and teachers of their performance evaluation based on students’ standardized test scores (Futrell and Brown, 2000).
This paper has suggested that the use of’ standardized testing can be unfair and that tests themselves may have several sources of inherent bias. Moreover, standardized setting limits the role of the parents and teachers, limits the scope and quality of course content, and distances students from active learning. However, it is not the intention of this paper to suggest that all standardized tests should be abandoned altogether. Indeed, in the absence of standardized tests, selection and placement decisions might become completely subjective, which may be even more unfair. Therefore, it is enormously important that standardized testing practices involve less unfairness, and that the tests themselves have less inherent bias.