Criterion-Referenced Tests: Measuring Specific Academic Skills

Jerry Webster, M.Ed., has over twenty years of experience teaching in special education classrooms. He holds a post-baccalaureate certificate from Penn State's Educating Individuals with Autism program.

Criterion-referenced tests are designed to find out whether a child has a set of skills, rather than how a child compares to other children of the same age (normed tests.) The test designers analyze the component parts of specific academic skills, such as number understanding, and then write test items that will measure whether the child has all the component parts of the skill. The tested are normed, in terms of what skill level a child should have. Still, the tests are designed to measure a child's acquisition of specific skills.

A test of reading skills would seek to discover whether a child can identify the specific sounds consonants make before it would evaluate whether a student can answer comprehension questions. The questions in a criterion-referenced test seek to find if the student has the skills, not whether the student does as well as other third grade children. In other words, a criterion-referenced test will provide important information that a teacher can use to design specific instructional strategies to help those students succeed. It will identify skills that the students lack.

A criterion-referenced test for Mathematics should reflect the scope and sequence of state standards (such as the common core state standards.) It would reflect the skills needed at each age: for young mathematicians, understanding one to one correspondence, numeracy and at least addition as an operation. As a child grows, they are expected to gain new skills in a reasonable order that builds on earlier levels of skill acquisition.

State high stakes tests of achievement are criterion-referenced tests that are aligned with the state's standards, measuring whether children actually have mastered the skills that are prescribed for the students' particular grade level. Whether these tests are actually reliable or valid may or not be true: unless the test designer has actually compared the success of students (say in reading new texts, or succeeding in college) with their "scores" for the test, they may not actually be measuring what they claim to measure.

The ability to address specific needs that a student presents really helps a special educator maximize the effectiveness of the intervention he or she chooses. It also avoids "reinventing the wheel. For example, if a child has trouble hearing final consonant sounds in words while guessing at the word using the initial sound, it may merely call for some structured word blending as well as having the student listen for and name the final sounds will help them use their decoding skills more effectively. You don't actually need to go back to reteaching consonant sounds. You can identify which consonant blends or digraphs the student doesn't have in his or her skills set.

Examples

The Key Math Tests are criterion-referenced achievement tests that provide both diagnostic information and achievement scores in math.

Other criterion-referenced tests include the Peabody Individual Achievement Test (PIAT,) and the Woodcock Johnson Test of Individual Achievement.