Test tasks

   An effective test can be called one that measures students' knowledge of an interesting level of preparedness better than other tests, with fewer tasks, better, faster, cheaper, and all this - if possible, at the same time.
    The problem of the effectiveness of pedagogical tests and test tasks is part of the general problem of the effectiveness of forms and methods of pedagogical activity. It is natural to ask the question - why testing is an effective form of organizing the control of knowledge of higher school students, and the test itself is also considered an effective and objective method of diagnosing the level and structure of knowledge? 

    The short answer to this question is that this test is being developed on a scientific basis, it is technologically advanced, not only easy to automate, but also, in fact, the main means of automating control; it is economical because it does not require those large expenditures of live labor of teachers. The test is objective to the extent that it is possible to limit the testing process from subjectivity, and sometimes from the arbitrariness of some teachers by providing the same time, the same conditions and evaluation rules for all subjects, without exception. Finally, the test is reflexive in the sense of the possibilities of evaluating the quality of test results: without evaluating the measurement error and the adequacy of the test data of the task, the results are not recognized as test, i.e. trustworthy.

Efficiency is defined as a concept derived from the result (effect). Since the effect belongs to the number of measurable concepts that have a clearly distinguished composition of empirical referents, then the effectiveness usually obtained from dividing the effect into costs, time, resources, etc., also belongs to the number of measurable concepts.
    An effective test can be called one that measures students' knowledge of an interesting level of preparedness better than other tests, with fewer tasks, better, faster, cheaper, and all this - if possible, at the same time. The concept of "efficiency" is also associated with the concept of "optimality", which is close to it in terms of content. The latter is interpreted as the best possible option, in terms of meeting several criteria taken alternately or together.
In determining the effectiveness of the test, attention is drawn to two key elements - the number of test tasks and the level of preparedness of students. If an optimal choice of a smaller number is made from a test with a large number of tasks, then a system can be formed that is not significantly inferior in its properties to a test with a relatively large number of tasks. In this case, a test with fewer tasks can be called comparatively more effective.
In addition, the effectiveness of the test can be assessed in terms of matching the level of difficulty to the level of preparedness of the students being tested at the moment. This assessment in the literature is often referred to as validity, bearing in mind the idea of the validity of the test, so to speak, by level. Since there have been many such private ideas that blur the concept of validity, there is a need to enrich them with justification methods and combine them with the concept of effectiveness.
    One of the main requirements for the objectivity of testing is to give all students the same set of tasks and the same time to complete. But at the same time, it is easy to understand the practical uselessness of giving weak students difficult assignments; most students are likely to fail to answer them correctly. The same is the case with an easy test: it is useless (ineffective) to give it to knowledgeable students, because there is a high probability of correct answers here, but now, and therefore almost all subjects will receive an equally high score on the test. In both cases, the students will not differ from each other. The measurement, therefore, will not take place due to the discrepancy between the level of difficulty of the test and the level of readiness of the student.
    From these considerations, it is easy to assume that the most effective test is a test that exactly corresponds to the difficulty of the tasks to the level of preparedness of the subjects. It is also easy to assume that no pre-made test is ever exactly appropriate to the level of preparedness of specific subjects. They are all different, and therefore tests of different difficulty levels are required for effective measurement.
    The correspondence of the level of difficulty of the test to the level of preparedness of students can be assessed by an indicator of the amount of information obtained during the measurement process. Since students differ markedly in their knowledge in real practice, the question of the effectiveness of testing each student with a certain set of tasks, depending on his level of preparedness, becomes one of the main issues in modern theory and practice of testing. That is why modern test systems simulate an adaptive test, the tasks of which are selected depending on the current responses of the subject.
    The effectiveness of the test can be assessed from the point of view of its differentiating ability; the latter is the higher, the better the differences between students' test scores are visible. Variance can be used as one of the possible indicators of the differentiating ability of the test. If, for example, there are two tests in the same academic discipline and one of them has a greater variance than the second (in the same group), then, all other things being equal, a test with a higher variance can be considered more effective than a test with a lower variance. The ratio of a larger variance to a smaller one with the same, for example, number of tasks, followed by multiplication by one hundred, can serve as one of the indicators of the comparative effectiveness of the test from the point of view of its differentiating ability.
    And finally, the effectiveness of the test can be assessed from the point of view of the differentiated accuracy of measuring the knowledge of students of different levels of preparedness. This is, in fact, the idea of a level-differentiated test reliability, which can be methodically implemented only in a new test theory. In the classical theory of tests, we were dealing with an average indicator of measurement accuracy. Meanwhile, even from the most general considerations, it can be said that the measurement error should depend on the correspondence of the level of difficulty of the test to the level of knowledge of the student. 

   The smaller the match, the greater the measurement error should be. In other words, for a qualitative measurement of a student's knowledge, a test of the appropriate difficulty level is needed.