Book (2022)/Types of Tests

For a test to work as we envision it should, we must first clarify what we expect from it. Different types of tests are used in different situations. While all tests share some common traits, each test emphasizes certain qualities while overlooking others.

Depending on which phase of learning the test is a part of and what it is intended to aid in learning, we can divide tests into formative and summative. The purpose of formative assessment is primarily to provide feedback to students and teachers about the progress of the lesson. The test becomes part of the teaching dialogue, supports the active involvement of students in their learning, and contributes to their motivation. The student finds out to what extent his or her knowledge and skills correspond to the course’s requirements and tries to use them. The test will help the student identify their strengths as well as areas that still need work. The results of formative testing help the teacher to make teaching more efficient, as they show the teacher which areas need more attention and where, on the contrary, further attention is unnecessary. For both parties, formative testing should primarily be indicative. A formative test is not subject to very demanding procedural requirements. In some cases, the imperfection of a formative test can even help education by stimulating discussion and involvement of all those involved.

In contrast, the objective of summative assessment is to provide an overall picture of the learning outcome. The results of summative tests are often the basis for further steps in studies or career. This type of assessment is most often undertaken following the completion of some integral part of the learning or at the end of a course, or, conversely, it can be used to verify an applicant's ability to enroll in the course or to start performing a certain job.

In practice, purely formative and purely summative tests are the extremes on a continuous scale. We often come across the fact that even the results of a test that is primarily formative, are in some way included in the student's overall assessment, and further progress is conditional on the achievement of certain results. Conversely, even summative tests and exams should, in most cases, provide feedback to both the student and the teacher, helping to improve the quality of the course and develop study skills.

When preparing a test, it is also necessary to consider to what extent and to what depth the acquired knowledge and skills are to be evaluated.

In this regard, the most demanding are proficiency tests – tests and examinations of (professional) competence, which assess the overall ability to perform an activity, for example, to communicate in a foreign language, to perform certain tasks, etc. Professional competence tests usually require workplace-based assessment and separate written testing can only be used in specific cases or as a component of the aptitude test.

Achievement tests evaluate to what extent the student has mastered a part of the course or a certain section of study.

The aim of diagnostic tests is to describe in more detail the test subject’s strengths and weaknesses.

Finally, prognostic tests and aptitude tests are intended to estimate the extent to which the tested person will be able to successfully complete a certain course and acquire the target competencies in it. For example, the Modern Language Aptitude Test (MLAT) measures a student's potential to successfully master foreign languages, the Scholastic Aptitude Test (SAT) assesses academic ability and the potential to graduate from college or university.

Standardized test design and evaluation are intended to ensure provability, reproducibility, and long-term reliability of the results of the most important tests. With a standardized test, it must be guaranteed that the result depends primarily on the skills of the examinee, not on the specific test version, the environment in which the test is taken, the supervisor or the evaluators.

In psychometrics, the term “standardization” has a number of meanings, and these will be discussed in more detail later, in a separate chapter. Standardization of a test is required especially where the result of the test is a recognized certificate, or the outcome is important for the further career of the test taker. Standardized testing includes the collection of testing data and its statistical processing with the aim, among other things, of detecting “non-standard phenomena” (copying, item leaks, hinting...). One of the basic tools used when preparing standardized tests are calibrated test items, with psychometric characteristics combined so that the test as a whole has the desired properties. Comparability between individual test runs is checked using anchor items, which make it possible to compare the difficulty of tests given on different dates. Standardization also includes the objective setting of the threshold for passing the test and the consistent ensuring of comparable conditions for all those tested. Standardized tests must be prepared and performed in such a way that the objectivity of their results can be proven, even in court. The requirement for standardization means increased costs. It is therefore always necessary to consider where these costs are justified and where it would be sufficient to use common, non-standardized examinations and tests.

In a non-standardized test, the specific examiner or evaluator plays a more important role. They often focus more on the individuality of the examinee and can better assess his or her personal talents and achieved competences. However, this approach is not suitable for comparing examinees with each other.

In situations where standardization would be inexpedient or even unfeasible (for example, for a too small number of examinees), steps are often taken that lead to the impartiality of a non-standardized assessment and thus to the reduction of undesirable effects on the assessment, especially the subjectivity of the examiner.

In person tests have always been the primary form of testing, where students are in direct contact with the instructor. In recent years, due to the pandemic, remote tests and exams have become more prevalent, with development in forms of testing that do not require direct contact between teacher and student. Major progress has been made in proctored testing methods, and open book testing is also being developed.