This article provides a general overview of standardized achievement tests. Beginning with a working definition of achievement tests and providing comparisons to Criterion Referenced Tests, the article seeks to provide a foundation from which to build an understanding of the different perspectives on achievement tests. The article explores, in depth, perspectives related to the actual design elements of standardized achievement tests and the effects on curriculum and instructional methodologies as related to "teaching to the test." It further provides an analysis of socioeconomic status as a contributor to the achievement gap and highlights a common accommodation provided for students with learning differences when taking standardized achievement tests.
Keywords Achievement Gap; Achievement Tests; Criterion Referenced Tests (CRT); High Stakes Tests; No Child Left Behind Act of 2001 (NCLB); Norm Referenced Tests (NRT); Standardized Achievement Test
Definition of Achievement Tests
Achievement tests are used to measure student achievement. These tests are often referred to as norm referenced tests (NRT). Achievement tests are also referred to as standardized tests and therefore have a specific set of criteria that must be followed in order to classify as such a test. Ediger (2003) indicates that in order for a test to qualify as a standardized achievement test, it must enforce time limits, must ensure that the subject matter tested is the same for all students taking the test, and it must provide the same directions for all students.
Questions on standardized achievement tests are written by groups of professional educators and writers. Ediger (2003) indicates that pilot tests are given to specific groups of students in order to determine if test items are too easy or too difficult. For example, if students in a pilot study all answer the same question correctly, test writers then know the test item is too easy and needs to be rewritten. Ediger (2003) further indicates that numerous pilot studies are conducted to ensure that test items are appropriate and that student scores spread on a continuum from low to high.
Marchant (2007) highlights popular national standardized achievement tests, including the Terra Nova and the Standford-9. Although some people believe the SAT is an achievement test, it is actually considered an aptitude test because it measures the likelihood that a student will achieve in college. Marchant (2007) further highlights the fact that many states have turned to developing their own standardized achievement tests in order to better align test questions with state standards. As achievement tests continue to become more high stakes in nature, states aim to ensure that tests reflect the curriculum taught in the classroom as closely as possible.
Marchant (2007) indicates that most standardized achievement tests are considered norm referenced tests because students are compared as individuals to a large group of test takers. Test takers define a cut-off point at which students either pass or fail a standardized test. The cut-off point is used to provide a means to compare between individuals and groups of students. Much careful consideration is given to assessing the placement of the cut-off point.
Difference between Standardized Tests
Popham (2007) points out that many people believe all academic achievement tests are interchangeable; one test is the same as another. However, this is definitely not the case as a quite a variety of tests with different objectives exist. Since standardized achievement tests are designed to spread scores out among participating students in order to differentiate between students, Criterion Referenced Tests (CRT) were developed to provide another means to measure student achievement. Criterion Referenced Tests differ from Standardized Tests in that CRTs are designed to assess how well an individual student meets a defined standard or objective. Ediger (2003) indicates that CRTs provide objectives for teachers to use as guidelines to assist with instruction. Additionally, Ediger (2003) highlights that some states use CRTs as a means to determine which students are passed on to the next grade and which students are held back.
Among some of the differences highlighted by Ediger (2003), the author indicates that CRTs do not spread students across a continuum of achievement. Therefore, as many students who can achieve the highest percentile possible, can actually do so without being compared to others. Furthermore, Ediger (2003) indicates that many test items on CRTs tend to be more open-ended as compared to standardized tests.
Standardized Achievement Tests Transform into High Stakes Tests
Tests originally designed to measure student achievement and to help diagnose areas where curriculum improvements can be made are transforming into high stakes tests used to evaluate success of students, teachers, schools, and school districts. Marchant (2007) indicates this trend is occurring because states are being pressured to implement accountability measures, especially with pressures felt from the implementation of the No Child Left Behind Act of 2001 (NCLB).
Marchant (2007) defines high stakes tests as any test that carries a serious consequence for students or educators. He clarifies that such consequences can include grade retention for students and rewards or punishments for schools or school districts that either meet, exceed, or fail to meet set objectives. Furthermore, Marchant highlights that such tests can determine whether or not a student is eligible for a particular program, whether not a student can graduate from high school and may even decide to which college a student is admitted.
Alternative Perspectives on Standardized Tests
Popham (1999) indicates many standardized achievement tests aim to differentiate students according to the number of questions they answer correctly and the number they do not answer correctly. Therefore, many achievement tests, by nature, do not ask a large number of basic skills questions. To some educators and researchers, this fact alone poses a significant issue simply because achievement tests cannot be used to determine how many children have mastered specific basic skills. Furthermore, Popham (1999) asserts given the limitations of time on standardized achievement tests, the number of questions asked to assess a particular skills or concept may, in fact, be too few to accurately determine if students have achieved mastery. Marchant (2007) further highlights the fact that many achievement tests use multiple-choice questions as a primary form of assessment. Given time limitations and the range and scope of questions, many educators and researchers believe achievement tests provide a narrow perspective on student learning and therefore are limited. If questions on achievement tests were more closely aligned with what is actually taught in the classroom and more reflective of the skills and knowledge required for basic mastery instead of primarily focused on differentiating students, some educators believe achievement tests would provide a more sound analysis of student achievement.
Marchant (2007) discusses the idea that standardized tests do little to improve student knowledge, skills, and learning outcomes. In particular, Marchant highlights the fact that immediate feedback is critical to provide necessary scaffolding for optimal learning to occur. In the case of standardized tests, feedback if often provided weeks, if not months, after a student completes the final question on the test. Learning, therefore, does not take place via standardized tests. Rather, standardized tests provide one snapshot of one particular moment in a student's learning journey.
Increased Test Anxiety
Due to the high stakes nature of many standardized achievement tests, Paris (2000) asserts that rather than viewing such tests as a learning tool, students are actually beginning to fear standardized tests and therefore are beginning to shift their focus from actually learning to determining whether or not they need to know a particular concept for a specific achievement test. Paris (2000) highlights that anxiety and fear escalate as students become older. Naturally, young students do not place as much emphasis on standardized tests and scores simply because they have not internalized the high stakes associated with such tests. Marchant (2007) discusses a study conducted by Hill (1984) in which researchers concluded that as many as 10 million elementary and secondary students performed below their ability level due to anxiety. Marchant (2007) asserts it is likely this number has increased over the years due to the...
(The entire section is 3887 words.)