4 July, 2016
Cut-Off Score is the the lowest possible score on an exam, standardized test, high-stakes test, or other form of assessmentthat a student must earn to either “pass” or be considered “proficient.” In some cases, tests may have multiple cut-off scores representing tiered levels of proficiency, such as basic, proficient, or advanced.
In education, cut-off scores may also be applied in certification and licensing exams that are used to determine whether educators and other school staff are professionally “qualified.”
Whether cut-off scores are set by individual teachers on course exams or determined by groups of experts using sophisticated psychometric methods on large-scale standardized tests, all cut-off scores are informed judgments based on either individual or collective opinion, and it is impossible to demonstrate that any particular cut score is unequivocally “correct.” In other words, cut-off scores are professional judgments that fall somewhere on a continuum between art and science, subjective and objective, and arbitrary and reasoned.
For standardized tests developed by testing companies and administered to large populations of students by states and national organizations—such as the SAT, ACT, or National Assessment of Educational Progress (NAEP), for example—cut-off scores are determined through a process generally called standard setting (for criterion-referenced tests) or norming (for norm-referenced tests).
In a typical standard-setting process, a test developer will form a standard-setting panel by recruiting a group of experts, such as psychometricians (specialists in the science of educational measurement) or teachers from a relevant content area. The panel will then use one or more research-based methods, developed by psychometricians and academics, for setting testing standards and determining cut-off scores. The process typically entails reviewing test items (questions, problems, tasks), determining levels of difficulty for each item, and using a statistical process, based on collective opinion, to establish a cut-off score or a set of cut scores corresponding to different levels of “proficiency” (e.g., basic, proficient, or advanced). While the complexities and nuances of standard-setting are beyond the scope of this resource, many readily available papers, reports, and resources are dedicated to explaining the intricacies of the process, including A Primer on Setting Cut Scores on Tests of Educational Achievement by Educational Testing Service.
When teachers create and grade tests or other assignments, cut-off scores necessarily rely more heavily on individual professional judgment. In addition, the criteria used to determine cut-off scores can vary widely. For example, historical precedent is often used to establish cut scores for course exams and assignments—e.g., a score of 70 has long been considered a “passing” score in many schools, regardless of the content of the test, how it was designed, or what the score represents in terms of academic achievement. The same passing score of 70 may also be applied to different types of assessments that are evaluated in different ways. For example, a score of 70 on a multiple-choice test may be determined using a simple mathematical formula—70 percent of the questions were answered correctly and 30 percent incorrectly. Yet a grade of 70 on science project or written essay may require a far more nuanced judgment about the content and quality of the work, which might be based on a single teacher’s professional opinion, for example, or on clearly defined criteria described in a rubric or scoring guide that several teachers use to evaluate student work in a more consistent manner from student to student or course to course.
Cut-off scores may become an extension of ongoing debates related to high-stakes testing—i.e., the use of tests to make important decisions about students, educators, schools, or districts. For example, “high stakes” test scores may be used to determine punishments (such as sanctions, penalties, funding reductions, negative publicity for districts and schools), accolades (awards, public celebration, positive publicity), advancement (grade promotion or graduation decisions for students), or compensation (salary increases or bonuses for administrators and teachers).
High-stakes testing is one of the most controversial and contentious issues in education today, and because the consequences or benefits tied to cut-off scores on these tests are—at their origin—based on judgments made by a relatively small number of people, cut-off scores can become the object of debate, particularly if they are perceived to be incomplete, flawed, or unfair.
In addition, cut-off scores—and the bar for proficiency they represent—can diverge significantly from system to system, state to state, test to test, school to school, and course to course, or from year to year when changes are made to learning standards and accompanying tests. All proficiency levels change in direct relation to the methods used to determine cut-off scores. It is therefore possible, for example, to alter the perception of student “proficiency” by raising or lowering cut-off scores on tests, or by changing the process or criteria used to determine them. In fact, some states have been accused of manipulating the perception of student proficiency by lowering cut-off scores or developing tests that are based on low standards—e.g., an eleventh-grade test that evaluates student performance based on content knowledge and skills they should have learned in eighth grade. For these reasons, proficiency determinations based on cut-off scores may become a source of confusion, debate, controversy, and even deception.
Psychometricians, researchers, and other specialists specific may also debate the specific standards-setting processes used to determine cut-off scores on standardized tests, but these debates rarely extend beyond academia