Improvements for the Future: Assessing Achievement and Revising Standards

---

Improvements in Assessing Achievement

---

The United States has relied upon standardized tests to evaluate learning because, in part, these tests are relatively inexpensive, easy to administer, and efficient in determining both individual and aggregate scores.(Click here for footnote 20.) The most commonly used tests include the California Achievement Test, the Comprehensive Test of Basic Skills, the Iowa Test of Basic Skills, the Survey of Basic Skills of Science Research Associates, the Stanford Achievement Test, and the Metropolitan Achievement Test.

These tests, however, have met with skepticism and questions about their validity and comprehensiveness. Concerns raised about standardized tests include their

• emphasis on low-level thinking;

• inability to test process or method;

• inability to test depth of knowledge;

• inability to capture various levels of thinking (e.g., to award partial credit for a correct approach but a wrong final answer); and

• tendency to lead teachers to "teach to the test" by emphasizing less advanced forms of learning in the curriculum.

This latter practice is particularly egregious when practiced by teachers of minority students. (See "Standardized Tests and Minority Students.")

According to one recent study funded by the National Science Foundation and conducted by the Center for the Study of Testing, Evaluation, and Educational Policy, teachers are dissatisfied with the standardized tests. Over 60 percent of 2,229 mathematics and science teachers in grades 4 through 12 surveyed felt that standardized tests negatively affected student learning. "The mandated testing caused narrowing and fragmenting of the curriculum, limited the nature of thinking, or forced them to rush too much for students to learn well" (Madaus et al. 1992, p. 16).

The study also found that the content covered by mathematics and science standardized tests was not well-balanced (Madaus et al. 1992, p. 16). The math tests emphasize number systems and theory, and minimize probability, algebraic thinking, measurement, and geometry. Similarly, the science tests emphasize life sciences and minimize physical sciences. The standardized mathematics tests ask questions demanding higher order thinking skills only 3 to 5 percent of the time (Madaus et al. 1992, p. 12). Only 8 percent of the standardized science test questions ask students to apply procedural skills toward problems and experiments; most do not stress application of knowledge.

Recently, there has been a good deal of activity among some organizations and in some states to design new assessment instruments. These new assessment tools are being designed to (1) track progress over time, (2) show how individuals learn, (3) assess educational programs, (4) indicate curriculum or teaching changes needed for improvement, and (5) inform policymakers about educational progress (Arter and Spandel 1991). These new tools will have to grapple with many of the problems discussed above.

Although there are some promising new approaches, test directors and researchers are concerned about quick implementation without sufficient investigation of the new tests' effects. Also, while the intention of the new assessment tools is to have them closely aligned to new, more demanding curriculum standards and better instruction practices, assessments are also expected to motivate students. Many are concerned that the same instruments cannot accomplish so many diverse tasks. Still, a number of new assessment approaches warrant continuing development.

Alternative forms of assessment to test students on higher order thinking skills and concept application rather than on rote memorization are now being developed. Some of these are discussed below.

---

Constructed Response Items

Performance-Based Assessments

---

Footnote 20:

This reliance has grown over the years. "Revenues from sales of tests used in elementary and secondary schools more than doubled (in constant dollars) between 1960 and 1989, a period during which student enrollments grew by only 15 percent" (OTA 1992, p. 3).

---

ContentsSearchContinue

---