Home > Research > Glossary

Glossary

Benchmarks

Benchmarks are detailed descriptions, often accompanied by examples, of what students at a particular grade should be able to demonstrate in order to meet a defined learning standard. For example, a Mathematics standard would state a broad requirement such as "solves problems using number facts and operations". Benchmarks for this standard would include specific skills such as "multiplies whole numbers" and "estimates a solution to a problem," among many other skills.

[top]

Constructed Response

Constructed response, also called open response or performance assessment, is a type of test question. Constructed-response questions require students to produce their own answer to a question rather than choose an answer, as in a multiple-choice question. The most common type of a constructed-response is the essay question, where students create a written composition. Other types of constructed-response questions include short answer, fill-in-the-blank, open-ended math problems, oral presentations, lab experiments, exhibitions, and portfolios.

[top]

Criterion-Referenced Test

A criterion-referenced test, also known as a CRT, is designed to measure student performance against a defined set of learning requirements or expectations. Most criterion-referenced tests measure the knowledge, skills, and abilities as defined in learning standards or curricula developed by state educational agencies or school districts. The test results are reported in terms of what students know or are able to do as compared to the defined criterion.

[top]

Grade Equivalent

A grade equivalent (GE) is a score reported on norm-referenced tests that allows educators and parents to compare students based on the performance of other students relative to the school year. Based on a 9-month school year (typically September through May), the score represents a period during the school year, displayed as a number to show a grade and a month. The score is an estimate of the performance that an average student at a grade level is assumed to demonstrate on the test at a particular time in the school year. For example, a score of 5.8 represents a performance level typical of fifth-grade students in the eighth month (April) of the school year.

[top]

Individual Student Report

Most standardized testing programs provide individual student reports that state the test scores for individual students and help teachers, parents, and students interpret the results.

[top]

Item

An item is a specific test question or task to be completed in an examination.

[top]

Mastery

Mastery is a level of performance on a criterion-referenced test that shows that a student has demonstrated the knowledge, skills, and abilities for a unit of instruction or subject area as defined by a recognized standard.

[top]

Normal Distribution

Normal Distribution is a graphic representation, resembling the shape of a bell, of the frequency of scores from a test given to a population of students. Most of the scores are clustered in the middle of the score scale (or the tall arc of the bell), with a decline in the frequency of scores spread out evenly left and right from the center (or the lower curves of the bell).

[top]

Norms

Norms are the scores from large, representative groups of students. Typically, these groups are defined by common characteristics such as grade level, gender, national/regional/state/district level, public/private sector, and urban/rural location. Norms help educators compare performance of individual students to these larger groups.

[top]

Norm-Referenced Test

A norm-referenced test, also known as an NRT, is designed to compare student performance to that of other students. Typically, a student’s performance is compared to a general population of students, called a norm group.

[top]

Objectives

Objectives are statements of the expected outcomes of instruction that define what is to be learned.

[top]

Percentile Rank

Percentile rank identifies the percentage of a student’s peer group (e.g., grade level) that a student’s score surpassed. For example, a student receives a test score of 66 and a percentile rank of 83. This means that a score of 66 is higher than 83% of the comparison group.

[top]

Performance Assessment

Performance assessment, also called open response or constructed response, is a type of test question. Performance assessment requires students to produce their own answer to a question rather than choose an answer, as in a multiple-choice question. The most common type of a performance assessment is the essay question, where students create a written composition. Other types of performance assessment include short answer, fill-in-the-blank, open-ended math problems, oral presentations, lab experiments, exhibitions, and portfolios.

[top]

Range

The range refers to the spread of test scores among a group of students. It is calculated by subtracting the lowest score from the highest score. For example, if the lowest score is 54 and the highest score is 93, the range is 39.

[top]

Reliability

Reliability is a measure of the consistency and dependability of a test score's representation of a student’s knowledge or ability. Reliability is the analysis of scores over such factors as time, different administrations of the same test, different tasks or questions that measure the same skill, or different score raters of the same performance question.

[top]

Rubric

A rubric, or scoring guide, is a description of the attributes for each of the score levels for a specific question on a performance assessment. A rubric helps score raters apply consistent scores to students' work.

[top]

Scaled Score

A scaled score is a conversion of a student's raw score on a test or a version of a test to a common scale that allows a numerical comparison between students.

[top]

Standardization

Standardization is a systematic process of developing, administrating, and scoring tests. These procedures ensure that all students are tested under the same conditions, that they are all given equal opportunity to determine the correct answer, and that all scores are established and interpreted using appropriate criteria.

[top]

Standards of Learning

Standards of learning are statements of learning achievement expected of students at the conclusion of an instructional process.

[top]

Strands

Strands are categories within content areas that describe sets of knowledge and skills. For example, mathematics can include strands such as number sense, computation, and estimation, among others.

[top]

Validity

Validity refers to the degree that test scores appropriately reflect the level of knowledge and skills that a test is designed to measure.

[top]