Students should know exactly what will be on their tests. Ideally, everything learned in the class should be on the test. However, tests cannot really be expected to be truly comprehensive both because of the amount of time it would take students to complete them and because of the unacceptable amount of time teachers would need to grade them. Naturally, only a sampling of the material taught would appear on the test. Therefore, tests must be constructed in such a way that forces student to study all the material presented in the course, yet tests them on only a fraction of that material.

Rather than telling students that everything included in the course is liable for inclusion on tests, students should receive, at the beginning of the course, a comprehensive study guide containing a comprehensive listing of terms and questions to study and for which to gather answers. If these terms and questions are constructed correctly, their answers would effectively represent all of the material taught in the course. The students would then be told that this list includes all of the information and questions that will be on their tests, but because of time limitations, only a random sampling of these questions and terms would actually appear. For example, at the beginning of a course students would be given a list of perhaps 200 questions, terms, and concepts they must know and that, when studied for properly, would represent a comprehensive knowledge of the course. But the student would also be told that 15 questions and 5 terms (for example) from this list would be chosen randomly just before the start of the final exam.

The list of questions that is given to students at the beginning of the course should be arranged in the order in which they are planned to be covered in class. This would help students focus on developing comprehensive answers to these questions. If these questions are worded properly, a correct answer would require that students have gained a comprehensive understanding of the material.

There could be many benefits to using this kind of system. Students would be guided in their studies by knowing exactly what to study for (basically everything in the course), they would feel more confident and enthusiastic as a result of the reduction of the ‘unknown’ factor so often found in testing, they could spend less time taking notes in class (because they would know exactly what information they would need in order to fill in gaps in their knowledge rather than just write everything down), they could spend more time listening intently to the lecturer, and they may be able to ask more intelligent questions that would help build a greater comprehension of the subject. Students would also be able to study ahead with the confidence that they are not wasting time studying things that will not be on the test because nobody knows what will be on the test until the random drawing is performed on the day of the test. The use of well-written lecture notes by the instructor would further enable students to concentrate on filling in gaps in their knowledge.