Standardized Test Scores Common Representations for Parents and Students

Types of Standardized Tests  Norm Referenced Test: Normed using large groups of test takers. Compares one taker to another. Measure achievement, predicts future performance.  Criterion Referenced Test: Measure a student against a specific set of knowledge (criterion).

Criterion-Referenced Tests  To determine whether each student has achieved specific skills or concepts.  To find out how much students know before instruction begins and after it has finished.  Measures specific skills which make up a designated curriculum.  These skills are identified by teachers and curriculum experts.

 Each skill is expressed as an instructional objective.  Each individual is compared with a preset standard for acceptable achievement.  The performance of other examinees is irrelevant.  Each skill is tested by at least four items in order to obtain an adequate sample of student performance and to minimize the effect of guessing.  The items which test any given skill are parallel in difficulty.

Norm-Referenced Tests  To rank each student with respect to the achievement of others in broad areas of knowledge.  To discriminate between high and low achievers.  Measures broad skill areas sampled from a variety of textbooks, syllabi, and the judgments of curriculum experts.

 Each skill is, usually, tested by less than four items.  Items vary in difficulty.  Items are selected that discriminate between high and low achievers.  Each individual is compared with other examinees and assigned a score--usually expressed as a percentile, a grade equivalent score, or a stanine.  Student achievement is reported for broad skill areas, although some norm-referenced tests do report student achievement in specific sub-areas.

The Bell Curve Standard Divations 2% 14% 34% 34% 14% 2%

Raw Score (RS)  A raw score is the number of points earned from correct answers on a properly scored test. The RS should not be used directly in interpretation.

Percentile Rank  Not to be confused with percentages, percentiles rank individuals within a group. Percentiles, defined on a scale of 1 to 99 with 50 being average (mean). This shows the percentage of scores in the group that are at or below a specific student's score.

Grade Equivalency Scores  The first digit represents the year of the grade level and the second represents the month of that grade level. It is a misinterpretation of the GE to interpret it as an estimate of the grade in which a student should be placed.  “If Mary, a second grader, made a GE of 4.7, her score is the same as the average score made by the students in the seventh month of the fourth grade on the same second-grade test that Mary took.”

Standard Nine (Stanine) Scores  Show a comparison of student scores ranging from a low of 1 to a high of 9. Scores of are considered average.

Reliability  The degree to which the test yields consistent results.

Validity  Degree to which the test measures what it is supposed to measure

Types of Validity  1. Content Validity: Does the test reflect the area to be tested? (If it is a comprehension test, is there a vocabulary portion?)  2. Criterion Validity: a) Predictive - how well does the test predict future performance in the area tested?; b) Concurrent - How close do the results match tests that measure similar competencies?  3. Construct Validity: How closely do the items on the test match what you believe about the area tested? Does it match your philosophy of reading?  4. NEW!!!!!!! Consequential Validity: To what use will the test results be put and is it a fair use of these results?

Standard Error of Measurement  Variability of the score if given to the same person a second time? (If the SEM is + 5 on a 50 point test then there is a 2/3 chance that your score the second time would fall between 45 and 50

Comparison – Norm / Criterion Referenced Tests