Types of Standardized Tests & Scores
Schools use standardized testing to identify academic and learning strengths and deficits and to promote accountability, but these tests are virtually useless unless teachers and parents know how to correctly interpret the scores. Most tests provide several different scores, with each score representing a specific factor. When parents and teachers clearly understand the various types of standardized test scores, they can work together to create better, child-centered education plans and lessons.
A standardized test is, by definition, standardized. A standardized test includes the same format, same types of questions and the same content no matter when or where the test is administered or who is taking the test. Questions may be multiple choice, true/false or short answer and it is administered either paper and pencil or on a computer. Examples of standardized tests at the undergraduate level include the SAT, ACT and PARCC. These tests typically report scores as a range and as percentile ranks.
Standardized tests often used for acceptance into graduate school include the LSAT, MCAT and GRE. No matter what type of standardized test you take, it is important to understand the different types of scores associated with each.
What are Raw Scores?
On the score sheet of any standardized test, the raw score is most likely to be the first one listed. The raw score is exactly what it sounds like: the raw data which is the number of items a student answered correctly on a given test section. Without a basis of comparison raw scores are useless. For example, if a test indicates a student answered 50 questions correctly but doesn't say how many total questions were given, the student's performance can't be measured.
What are Standard Scores?
Standard scores represent how close to the average, or mean, the student's scores fall. Generally represented by a single number, standard scores can be interpreted by knowing which number represents the mean. Any number higher than the mean is considered above average and any number lower than the mean is considered below average. A common reporting method is the Standard Nine (Stanine), which gives a number score of 1 though 9, with the numbers 4, 5 and 6 being the mean. Standard scores can also be reported as Normal Curve Equivalents (NCE), which have a mean of 50, and Standard Age Scores (SAS) that have a mean of 100.
What are Norm-Referenced Scores?
Norm-referenced tests measure and rank test takers to each other. A test taker’s score is compared to the ‘norm’ of similar test takers and may be expressed as a percentile, grade equivalent, or stanine. Common norm-referenced tests like the Woodcock-Johnson, the California Achievement Test and the Stanford Achievement Test, use data to compare a student's performance to that of his or her peers. Scores can be reported in percentile ranks and age and grade-equivalencies. Percentile rank shows what percentage of students perform higher or lower than the student's score. For example, a student with a score of 95th percentile, this means that the student scored better than 95 percent of all other test takers.
Age and grade-equivalencies are the age or grade level at which the student performs. For example, a test may indicate the student reads at a fifth-grade level. Many researchers warn that grade and age-equivalency scores are unreliable because children acquire skills more rapidly when they are younger. So, a six month delay affects young children more than adolescents or adults. However, students can make gains over time, therefore a one-time grade equivalent score should not be used to determine a student's full potential.
What are Criterion-Referenced Scores?
Criterion-referenced tests measure the number of correct responses based on a specific ‘criterion’ of what is expected to pass the exam, or acceptable achievement. A criterion-referenced test score may be expressed in a percentage correct out of the total. Criterion-referenced scores may be used to determine if a student has mastered specific skills or concepts in specific areas of study.
Most curriculum-based assessments, in-class quizzes and exit exams, like the California High School Exit Examination are criterion-referenced tests. These are sometimes referred to as "achievement tests" because they score a student's performance on a predetermined set of skills or standards.
Generally, raw scores from criterion-referenced tests are converted into a percentage, and performance is valued relative to 100. For example, if a student answered 40 out of 50 questions correctly on a criterion-referenced test, the score would be converted to 80 percent correct. Student progress can be tracked by how much improvement is made to this percentage over time.