A glossary of testing terms
By the Notebook on Mar 13, 2003 12:00 AM
A set of fixed questions given to large numbers of students under the same conditions (such as a time limit) and scored in the same way. Standardized tests can be given on a citywide, statewide or national basis. The format of test items typically is heavily multiple-choice, though open-ended items such as problem solving in math and questions that require a written response can be included.
The TerraNova and PSSA are standardized tests, and both include multiple-choice and open-ended questions.
Tests designed to evaluate whether a student has mastered certain knowledge, concepts, and skills in particular subject areas as judged against standard criteria. Students are not ranked against each other but against pre-set levels of content mastery or performance -- advanced, proficient, basic, below basic. In theory, 100 percent of students can reach an advanced level.
The PSSA is a criterion-referenced test designed to assess student performance in light of Pennsylvania academic standards in reading, math, writing, and science. Individual scores can also be reported as percentiles -- indicating the percentage of students taking the test who scored above or below each student test taker.
Tests designed to compare a student's score against the scores of a statewide or national sample of test takers. Norm-referenced tests compare or rank student scores on a bell curve, or normal distribution. The test is designed so that a few test-takers score very high, a few score very low and most score in the middle range. Scores are usually reported as percentile ranks, ranging from l percent to 99 percent. Content of the test reflects material covered in textbooks used nationally and may or may not be well aligned with particular state standards and district curricular frameworks.
The TerraNova is a norm-referenced test on which scores can also be reported by performance levels. (The SAT-9 used in the Philadelphia School District from l995-2002, was also a norm-referenced test.)
Standards define what students are expected to know and be able to do in specific subject areas at various levels of schooling. In recent years, standards have been developed in all states, typically with input from educators, parents, representatives from government and business, and various national professional organizations such as the National Council for Teaching Mathematics. Philadelphia developed its own standards in a similar process in l996, and teachers received curricular frameworks that provided guidance for implementation of District standards in l998.
Academic or content standards are the concepts, skills, and facts students are supposed to learn in a particular grade or level of schooling.
Performance standards are criteria used to determine how well students are able to demonstrate knowledge of academic standards. For instance, by the end of third grade, students will be expected to know how to multiply numbers.
An assessment approach that uses the results of standardized tests to make educational decisions about individual students, such as whether a student will advance to the next grade, be able to enter a preferred school, or graduate from high school. Tests can also be considered to be high stakes assessments if they are used to evaluate teachers, principals, or schools, with rewards or punishments based wholly or primarily on standardized test scores.
MULTIPLE-CHOICE TEST QUESTIONS
A test question format that requires students to choose an answer from a selection of pre-written choices or alternative answers. At least one of the answers, in addition to the correct answer, should be a plausible alternative.
OPEN-ENDED TEST QUESTIONS
A test question format that requires students to show their work and/or write out their answers. For instance, an open-ended test item may require students to explain how they solved a math problem or to write a short answer. Open-ended questions require students to construct their response, rather than to simply choose a pre-written response.
Assessments that go beyond multiple-choice or short answer test formats and are used to evaluate a student's progress in real world activities. Performance assessments can include writing an essay, conducting a science experiment, making an oral presentation, explaining how the student arrived at an answer to a math problem, designing a model, or creating a piece of art. Performance assessments are scored using rubrics or descriptions of what "good" work looks like in each type of activity. Sometimes performance assessments are referred to as "alternative" or "authentic" assessments.
NO CHILD LEFT BEHIND ACT (NCLB)
This legislation, passed by the United States Congress in 2002, mandates certain accountability measures that states must put in place in order to receive federal education funding.
By 2005-06, every state must administer standardized tests annually to students in grades three through eight in math and reading, as well as once in grades 10-12. (By 2007, students will also be tested in science.) Schools are required to show "adequate yearly progress" on students' achievement scores (see definition below), or "school improvement" measures will be implemented.
Student performance on state tests must be reported to the public and include the following information at the state, district, and individual school levels:
- Student academic achievement on statewide tests that must be disaggregated by subgroup (including by race/ethnicity, income levels, and other designated groups such as special education).
- Percent of students achieving at below basic, basic, proficient and advanced levels. (Performance levels are determined by individual states.)
- High school graduation rates.
- Number and names of schools identified for "improvement."
- Professional qualifications of teachers: by 2005-06 every child must have a "highly qualified" teacher.
- The percentage of students not tested.
Individual school reports, which must be made widely available to the public, will also include whether the school has been identified for school improvement because of inadequate progress and how its students performed on the state test compared to other students in the school district and in the state as a whole.
ADEQUATE YEARLY PROGRESS (AYP)
Adequate yearly progress (AYP) is a key component of the accountability system required by the No Child Left Behind Act. It is an individual state's measure of yearly progress toward achieving proficiency on state academic standards as demonstrated by student performance on the standardized tests used by the state.
AYP is the minimum level of academic improvement that states, school districts, and individual schools must achieve each year in order to reach proficiency levels for all students by the end of 12 years in 2013-14. Achievement data must be reported in disaggregated or separated form for students from various racial/ethnic or low achieving groups who are most in danger of being "left behind," rather than reporting only the average achievement of students at a school.
Individual states must each define and implement a plan for monitoring and achieving adequate yearly progress. Schools that do not meet the AYP goals for two consecutive years will be identified as "needing school improvement" and options including technical assistance, supplemental educational services, and giving parents the choice to transfer their child to a better public school will be offered.
Pennsylvania is in the process of developing how it will specifically measure and implement the AYP requirement of the No Child Left Behind Act.