A r c h i v e d  I n f o r m a t i o n

Examinations

Traditional In-Class Tests

Tests signal students as to what teachers consider important to learn, thereby shaping learning. Because it is time-consuming and difficult to construct tests that assess thorough understanding, teachers often devise tests that are geared at a relatively low cognitive level (Linn 1990).

Performance-Based Assessment

To complement traditional testing programs, many school districts and states are turning to performance-based forms of assessment. This type of testing requires students to create answers or products that demonstrate what they know by presenting them with problems they have not previously met but that call on the skills and information they should have acquired in class. Performance-based assessments may include portfolios of student work, exhibitions, simulations, science experiments, oral interviews, and student performances. They are based on the premise that testing should be closely related to the kinds of tasks students are trying to learn.

State and local districts have recently begun adopting some form of performance-based assessment for a range of grade levels. For example, 36 states now use direct writing samples in testing students (U.S. Congress 1992).

Other assessment reforms are slowly being implemented around the nation. For example, the New Standards Project is creating standards of assessment based on the use of real-life tasks that students are asked to perform. Thus far, the New Standards Project has begun the task of setting student performance standards by examining those in other industrialized countries. The purpose of the project is to create standards for assessment that will promote better student performance (Learning Research and Development Center 1994).

Performance-based assessments raise many issues as well. Issues of reliability, expense, and time must be grappled with in promoting such types of assessment. For example, one estimate revealed that scoring a writing assessment is 5 to 10 times more expensive than the electronic scoring of multiple-choice tests (U.S. Congress 1992).

Standardized Tests

Students in American schools are subjected to a large variety of standardized tests from kindergarten through graduate school, with the process often becoming an annual practice in many parts of the country (Romberg 1992). Standardized achievement tests are used for:

• guiding classroom instruction,

• monitoring school achievement results, and

• the selection, placement, and certification of students.

Standardized tests are used most widely in elementary schools. Even in states that do not mandate such assessments, they are very common. For example, a survey of Pennsylvania school districts found that 91 percent use some form of standardized test and that nearly 70 percent of eighth-grade students take a mandated test at either the district or state level (U.S. Congress 1992).

Reliance on standardized tests has increased in the past few decades. As a result of their frequent exposure to these types of tests, students may respond to them differently as they grow older (Paris, Lawton, Turner, and Roth 1991). Accordingly, older students tend to report decreasing motivation to excel on the tests, anxiety about the tests, worrying about doing poorly, cheating, and not putting forth their full effort. Such a pattern is especially the case among low achievers (Paris et al. 1991).

Although these tests have positive features—they are easy to develop, inexpensive, and convenient to administer—they are also frequently criticized (Romberg 1992). Since teachers know both the form and style of the tests, they may modify their teaching to conform to the tests. One frequently heard criticism of these tests is that they emphasize procedural knowledge rather than understanding. Romberg (1992) has shown this to be the case in the six most common tests used in eighth grade (see table 3).

Table 3—Number of items in six commonly used eighth-grade tests that tap conceptual and procedural knowledge

The most frequent criticisms lodged against the tests are that most are norm referenced, rely exclusively on multiple-choice questions, produce scores that are not direct measurements of the questions, contain questions that do not represent the local curriculums, derive scores that are not direct measurements of the traits, and are often used as a basis for decisions they were not designed to address (U.S. Congress 1992).

One new trend in standardized testing is computer-based testing (CBT), which has its own set of advantages and disadvantages (U.S. Congress 1992). Although tests done via the computer are readily scored, provide fast feedback, and reduce grading errors, they also place students who lack familiarity with computers at a disadvantage. In one form of computerized testing, computerized adaptive testing (CAT), the computer chooses items to administer based on the examinee's responses and previous test questions. Thus, not all examinees receive the same set of test items.

The Curriculum

Proposed Changes in Curriculum of Math and Science

Mathematics and science education have been particularly affected by changing perspectives on the nature of mathematics, the need for mathematics, and mechanisms for learning mathematics. An additional influence has been the changing roles of computers and calculators in the practice of mathematics (National Academy of Sciences [NAS] 1990). Table 4 presents data from the 1990 NAEP which reveals a decline in the use of computers and a concurrent increase in the use of calculators as students grow older (USED 1993a).

Table 4—Percent of students who used a computer or calculator in mathematics class, 1990

There is a new priority on reducing routine skills in favor of promoting higher-order "thinking skills." As the Mathematical Sciences Education Board describes, "By reducing emphasis on manual skills, it will be easier to develop a curriculum that will allow all students some level of mathematical accomplishment while retaining the interest and enthusiasm of the more able students" (NAS 1990, p. 20).

In accord with this, the Board expects that by the year 2000, all students will have hand-held, grade-appropriate calculators, all mathematics classes will have permanent computers, and students will have access to other facilities, such as portable computers (NAS 1990).

The National Research Council (1994) proposed three critical issues in curricular reform for science and mathematics:

• All students need to develop higher-order thinking skills,

• students should learn by actively constructing their own knowledge, and

• students should learn a smaller quantity of information in greater depth if they are to increase their understanding of science and mathematics.

These general themes recur throughout the curricular reforms currently underway in the United States, such as Project 2061 of the American Association for the Advancement of Science (AAAS), and the Scope, Sequence, and Coordination Project of the National Science Teachers Association (NSTA).

Agents of Development and Implementation

School programs often evolve in response to pressure from many different directions: the public, professional advisory groups, different levels of government, parents, and teachers. In fact, curriculum planning may vary even within a school.

The curriculum is also subject to the influence of factors that lie within a school, such as the quality of the teachers, the textbooks used, the classroom dynamics, and the amount of time teachers have to plan effective lessons. Consequently, the curriculum as it actually is implemented is often quite different from the goals intended for the curriculum (Robitaille, et al. 1993).

Teachers often do not have the time or resources to plan the curriculum since they are busy teaching 5 days a week, and many have families or other jobs (Walker and Soltis 1992). Who then, does the bulk of curriculum planning? The responsibility for specifying a curriculum has fallen by default to authors of textbooks. In the absence of a national curriculum, textbooks define what students should know and be able to do in a given subject. In fact, students spend as much as 90-95 percent of class time involved in one way or another with textbooks (Young and Reigeluth 1988).

American textbooks generally are thick volumes of many hundreds of pages that are loaned to the students for the year by their school. The textbooks often include colorful illustrations and interesting stories related—but often not central—to the lesson. The "spiral" curriculum is followed in the textbooks, whereby concepts are introduced at one grade level and discussed at successively more advanced levels in subsequent grades. Ultimately, it is teachers who must decide what aspects of the textbooks to cover, and they seldom have time to cover all the topics presented in them (Stevenson 1992).

There is no national consensus on selection of textbooks; rather, each school has autonomy in making decisions about which textbooks to use, although individual districts usually provide some recommendations. Typically, school districts will have a curriculum committee or a person specifically hired to screen and select textbooks that are appropriate to fill various needs. Publishers, then, devise textbooks that appeal to the broadest market (Venezky 1992).

The Influence of Testing on Curriculum

One way that standardized tests have affected the curriculum is through the creation of new programs. For example, Measurement Driven Instruction (MDI) programs have been created so that the content of high-stakes achievement tests "drives" instruction (Airaisian 1988). In MDI, a heavy emphasis is placed on the test, which provides direction for instructional emphases. Mostly, it has been the case of tests informing the curriculum, rather than curriculum shaping the tests.

Course Choice Patterns

When students reach high school, they generally have some options as to which courses they will take within their general "track." The courses that students are directed toward and eventually choose give some indication of how the curriculum tapers off in later years. For example, of the yearly cohort of 4 million students, 500,000 are studying mathematics 12 years later (NAS 1990). Moreover, one-third of America's 21,000 secondary schools do not offer a sufficient number of mathematics courses to qualify graduates for admission to accredited university engineering departments (Holmes and McLean 1989).

The statistics are similar for science courses. Although most science courses are offered by a majority of schools in the United States, they are not offered at all schools (NSF 1993). In addition, students do not tend to take the advanced courses. For example, as shown in table 5, only 21.5 percent of 1990 high school graduates had taken physics. A severe drop in enrollment in both science and mathematics courses occurs as courses are perceived to be more difficult (NSF 1993). Table 5 shows which mathematics and science courses are most frequently taken by high school students and the changes that have occurred between 1982 and 1990 (USED 1993a).

Table 5—Percent of high school graduates who had taken selected mathematics and science courses: 1982 and 1990

There has recently been a move to increase course requirements in mathematics and science. In 1990, about 50 percent of the states required 1.5 to 2 years of mathematics; those requiring 2.5 to 3 years increased to approximately 20 percent. Nevertheless, more than 15 percent of the states did not require any mathematics for high school graduation (NSF 1993).

Several factors influenced the available courses and eventually the ultimate enrollment (NSF 1993).

• Income level of parents: children from low-income families tended to take lower-level courses.

• Ethnic background: more Asian Americans took science and mathematics courses than did students from other racial-ethnic groups. For example, in 1987, the percentage of high school graduates who had taken biology, chemistry, and physics included 42 percent of Asian Americans, 18 percent of whites, 9 percent of blacks, and 8 percent of Hispanic Americans. Differences in the percentages enrolled in upper-level mathematics classes were even greater: 15 percent of Asian Americans, 2 percent of whites and Hispanic Americans, and 1 percent of blacks had taken a second semester of algebra, geometry, trigonometry, and calculus.

• The school's surrounding population and wealth: the percentage of students taking mathematics and science was generally lowest for rural schools, followed by inner-city schools. The highest percentage of students enrolled in advanced courses was from advantaged urban schools. For example, in 1990, more than 90 percent of advantaged urban communities offered calculus, compared to 55 percent of schools in the rural areas. The same pattern was evident for physics courses: nearly 60 percent of schools in advantaged urban communities offered physics, compared to less than 10 percent of schools in disadvantaged urban and rural communities.

• The size of the school: students at midsize schools were more likely to take mathematics or science courses than those in larger or smaller schools.

Preparation for Postsecondary Education

College Entrance Examinations

The SAT (I and II). The Scholastic Achievement Test (SAT) of the Educational Testing Service is taken by nearly 1.5 million people a year and used by more than 1,500 colleges and universities as part of the process of making decisions about admission (Crouse and Trusheim 1988). The test is constructed to measure general ability as it has developed over the full range of experiences in a student's life, but it is only minimally related to the main curricula of schools (Owen 1985). The test, which is composed of seven subsections, requires a total of 3 hours, with each subsection being timed separately. Among these seven subsections, three are devoted to mathematics and three to verbal items; one experimental section tests new questions.

The entire cost for taking the SAT is borne by the students. Some students incur additional costs by requesting special services or materials from test makers or private publishers. For example, there are many books on preparing for the SAT, some created by the Educational Testing Service itself. Courses and private tutors are also available to help students prepare for the tests, although their expense limits their use to students from middle- and upper-income families.

Recently, the SAT has undergone important changes in content and format (College Board 1993). The verbal sections place greater emphasis on reading comprehension by including longer passages and appropriating more testing time for fewer questions. Changes have also been introduced to the SAT mathematics test. Some of the multiple-choice questions have been discarded in favor of student-produced answers—there are now 50 multiple-choice questions and 10 questions requiring responses produced by the students. Additionally, calculator use is permitted and recommended on this test; hence, the test presumes experience in the use of scientific calculators.

The SAT II (subject tests) measures students' knowledge or skills in one of the following general areas: English, history and social studies, mathematics, sciences, or foreign languages. Each such test takes 1 hour and consists entirely of multiple-choice questions, except for the writing test. According to the bulletin Taking the SAT II Subject Tests, "Scores on the Subject tests can help in assessing how well prepared you are for different programs of college study" (Educational Testing Service [ETS] 1994, p. 4). Some institutions use the subject tests for placement and guidance; others use them for decisions about admissions.

The ACT. A second major college entrance examination is the ACT (American College Test). It is designed to measure the skills necessary for college coursework by assessing English, mathematics, reading, and science reasoning via multiple-choice items (American College Testing Program 1993). Much like the SAT, the test is timed so that students have to pace themselves quickly in order to finish all items. Traditionally, the ACT has been regarded as an achievement test—emphasizing the content and processes of students' knowledge that are amenable to change with further learning. Although the SAT has been considered to be an aptitude test, distinctions between aptitude and achievement tests have been rejected (Owen 1985).

Mechanisms for Entering College

Highly selective colleges. Colleges usually have multiple objectives in admitting students and are willing to make trade-offs among these objectives. Highly to moderately selective colleges consider many factors in making their admissions decisions: high school record, test scores, campus interviews, teacher recommendations, personal statements, essays, patterns of courses taken, special talents, ethnic background, and residence. The weight allocated to each of these variables, however, is not clear-cut; colleges do not usually discuss how they choose to admit or reject students. Thus, prospective applicants are not always sure just what it takes to get into certain schools. Various books are available to guide students in deciding where to apply, informing them of the mean test scores and grade point averages of admitted students. Such guides often indicate that nonacademic factors may not be of great importance for admission to highly selective colleges.

The majority of colleges. The college admissions system can be viewed as self-selecting in that students generally choose to apply to colleges at a level appropriate to their abilities rather than risk rejection by applying to more-stringent schools. As a result, most prospective students are admitted to at least one of the schools to which they have applied. Admissions requirements for two-year colleges are less stringent than those of four-year institutions. In fact, most two-year colleges will accept all prospective students who apply or who meet minimal standards (Ravitch 1985). For example, in 1985, 90 percent of two-year public colleges and 15 percent of four-year colleges did not review their applicants' qualifications, admitting all high school graduates (Crouse and Trusheim 1988). Consequently, "American high school students who plan to go on to college do not need to work hard and get good grades in order to achieve their goal . . . no matter how poor their grades, they will be able to find a college that will accept them" (ESCPC 1993, p. 52). Once a student has been admitted to a college, the possibility of transferring to a better school remains open, and it is not uncommon for those who maintain a strong academic record to do so.

Vocational options. A growing need has become evident for more attention to be focused on the transition from school to work. According to one report, "The U.S. does the poorest job in the advanced industrial world of facilitating students' transition from school to work . . . and, when students graduate from high-quality vocational education programs, they are rarely hired for good jobs until they are well into their twenties" (ESCPC 1993). High school graduates typically have to wait until they are 21 to 24 years old to be considered for a decent job. "In the meantime, they float in the churning sea of a youth labor market that is mostly made up of poorly paid, high-turnover jobs . . . often in the retail sector" (ESCPC 1993). Students who desire vocational jobs may also go to career academies, technical preparation programs, or special vocational programs. A small percentage enter apprenticeships.

Summary

Student achievement has been evaluated primarily through state testing programs. The absence of clear expectations for students has shaped the recent call for establishing voluntary national standards for education. These standards would set guidelines for what students should know at different levels of schooling. In addition to funding the development of standards for academic achievement in specific subjects, the United States government has also devised national goals for education. The fourth goal specifically targets achievement in mathematics and science, proposing that, by the year 2000, U.S. students will be the first in the world in mathematics and science.

At a more informal level, standards for student achievement can be inferred from examinations and the curriculum. Traditional in-class tests, although still widely used, are being supplemented in some schools by performance-based assessment. Portfolios of student work, exhibitions, simulations, oral interviews, and student performances, all performance-based types of testing, attempt to close the gap between what students are learning and how they are tested.

The curriculum in U.S. schools is highly fluid and varied. Each state delegates to local school districts the responsibility for establishing its curriculum, the subjects and topics that will be covered, and the requirements for each course. Textbooks, instead of supporting the chosen curriculum, often determine which route the curriculum will take. Reform efforts are being made to include the use of calculators and computers in the mathematics and science curriculum. By upgrading requirements for graduation, states are attempting to increase the number of students taking upper level courses. Those students who wish to continue on to four-year postsecondary institutions must take one of two college entrance examinations, namely the ACT or the SAT. In accord with changes in school curricula, the SAT is moving toward student-produced answers, as well as allowing the use of a calculator for its mathematics sections. These tests will open the gateway to college for many students, but only a select number of students will gain entrance to the top colleges and universities. Other students may choose to continue with vocational education.

-###-

[Components of National Education Standards in the United States - Part 1]

[Components of National Education Standards in the United States - References]