Most of the standardized testing conducted within our schools falls short of desired effectiveness with regard to the two most common means of assessing the quality of a standardized test—reliability and validity. Reliability means that an individual taking the test will have approximately the same score, each time she or he takes the test. Validity means that the items on the test measure what the test is designed to measure.
Within the present system, problems with reliability are related to the fact that the effort students put into taking a test may vary considerably from one testing session to the next and/or from one school year to another. In situations where standardized test scores do not affect students’ grades or chances for graduation, students have little or no incentive to put forth their best effort. When students have been taking tests for several days in a row, they may begin to put forth less effort. In some cases, students simply mark answers on the answer sheet without even looking at the questions. In other cases, they may take the time to answer the questions they know but put very little effort into questions or problems that may require more thought or effort. As a result of these factors, the test scores of some students vary significantly from one testing session to another and from year to year.
Problems with validity arise when sections of a standardized test cover materials, skills, or concepts that a student has not been taught. No matter how hard a school district works to align their course of study with the tests being given, there are frequently questions on a test related to classes a student has not yet taken, or which include material a teacher did not cover in a particular class.
The problems related to the reliability and validity of standardized tests can be effectively addressed by using standardized test scores as one of the factors in determining whether or not to award credit for required classes, or other classes where credit is issued, as well as in assigning grades. We should utilize course-specific tests that measure the level of mastery each student has achieved with regard to that subject. Pre-tests should be used to determine the knowledge of each student in a particular subject. If a student already understands some of the concepts, has already developed some of the skills, or has already acquired some of the knowledge, that will be covered within a particular course of study, she or he should be allowed to take an abbreviated version of the class, with the option of skipping sessions that cover what they have already mastered. Students should be allowed to test out of a class if they score high enough to receive a passing grade on the pre-test. Post-test data should be used to determine which objectives or standards have not been mastered by each student. Schools and teachers should utilize the test results on a student-by-student and item-by-item basis, to provide remedial assistance for those students who have not achieved the required level of mastery.
In situations where grades are being assigned, if a student is not satisfied with the grade they earn on the post-test, they should be allowed to repeat part or all of the course and then take the test again. Each test should cover only the objectives and standards that pertain to that particular course. Until a student is able to demonstrate a reasonable level of mastery, she or he should not be awarded credit for that subject. If a significant percentage of the students taking a class with a particular teacher seem to be having difficulty with the same standards, it would indicate the need to offer alternative forms of instruction or utilize different instructional materials.
It is worth mentioning again that not everything that can be tested is worth learning and not everything that is worth learning can be tested. Many standardized test items measure only the lowest level thinking skills: Recognition and recall. It is considerably more difficult to design items for a standardized test to effectively assess a student’s ability to analyze, synthesize, and/or evaluate information. Writing essays, book reports, and research papers, and completing other projects are important components of a well-rounded education. Some degree of subjectivity is unavoidable in assessing writing and research skills. It would be necessary to combine standardized test scores with more subjective assessments in classes that include these types of assignments. Including standardized test scores as part of the assessment process would provide a more objective means of assessment than teacher-assigned grades.