Book/Determining Borderline Test Scores

Finding the threshold to pass a test is often referred to as “standardization”. The objective is to find a line between those whose performance is considered adequate for the purpose for which the exam is intended, and should therefore pass the test, and those whose performance is considered insufficient from this point of view. Determining a cutoff score, like any human activity, will contain some margin of error, which can lead to false positive and false negative decisions. The objective of standardization is to minimize these errors.

The method for determining this threshold should simultaneously be:

• Defensible
• Credible
• Supported by evidence in the literature
• Easy to do
• Acceptable to stakeholders

It is usually not sufficient to merely give points on a written test. For most tests, it is also necessary to say which students passed the test and which did not, or grades must be assigned to individual point gains. Determining the cut score (passing grade) is sometimes an underestimated, yet extremely important step. When compiling a test, it is rather difficult for its writer or writers to estimate how difficult individual items will be for students, and it is even more difficult to “hit” a certain value for the overall difficulty of the created test. Nevertheless, cut-off scores are often determined tentatively, based on the estimation of one or a few teachers. If testing is of greater importance, such an approach is questionable – the test results can be contested by claiming that the evaluation was unreasonably strict, or, on the contrary, that the test was too benevolent and allowed even students who have no business there to further study or practice. Therefore, in the case of standardized testing, the cut-off scores should also be set in a standardized way. As a result, the established cut-off score is substantiated, justified and much more reliable. There are several ways to find the cut-off score using a standardized procedure. The individual approaches differ depending on the purpose of the test, and there are also significant differences in their complexity and demands on qualified experts and their time.

Relative, absolute and compromise methods

Student evaluation can be based on comparing students' performance with each other. We call such an assessment relative. Or the evaluation can be based on the fulfillment of some absolute (independent of the performance of others) criteria. We call such an assessment absolute. Alternatively, it can combine elements of both, and in this case we are talking about the compromise method.

The method of relative assessment is based on the assumption that in large groups there is always a (approximately the same) part of the test takers that are prepared to pass the test. There is a certain optimism in this, because if all the test takers were poorly prepared, the method will still select some part of them as satisfactory. That is why this method is especially suitable where we are not focused on the specific competence of the applicants, but about selecting the best from the given group. A typical use of this method is, for example, acceptance tests.

Absolute assessment, on the other hand, requires test takers to demonstrate specific knowledge and skills that entitle them to pass the test or to perform some activity. An example of this kind of test evaluation is the exit test from a driver’s training course, state exams, certifications, etc.

In theoretical considerations about the assessment and classification of students, we can view these two different concepts of assessment as a manifestation of two different views on the purpose of higher education.

In the first case, we can view education as a perennial intelligence test that sorts individuals according to their intellectual skills and work habits. This approach reflects the interest of potential employers to select the most suitable candidates for a limited number of prestigious positions and helps to ensure that the most capable are selected for key positions. In the course of studies, this approach pits students against each other, letting them compete with each other. The evaluation method in this case will be relative evaluation.

The second view is different. It assumes that the purpose of education is to enlighten, strengthen and socialize citizens. According to this view, the educator should not focus so much on sorting students according to skill, but on helping them to find the right view of the world and themselves, with the objective of equipping them with the knowledge, tools and habits that will make them useful and culturally literate members of society. Assessment of students within this concept is based on the fulfillment of absolute criteria and is therefore an absolute assessment.