There are several types of testing questions. Some types are easier to create. However, it is important to choose the type of question based on the purpose and content the student is being tested on. The objective test items include multiple choice, true/false, matching, and completion questions. The subjective test items include short answer, and essay questions.
Subjective questions are more difficult to grade and require more preparation on the part of the student. Whereas objective test questions are easier to grade but can encourage guessing more often than the objective test questions.
Creating a test is demanding but can be made easier if there are other instructors that will proofread the test questions. Test-writing can be more manageable if the test-writer will follow the objectives of the course. Test reliability refers to the dependability or consistency of the test. Often, KR-20 will be used to measure reliability. A KR-20 that is closer to a value of "1" is more reliable. A KR-20 that is below 0.70 is not very reliable.
The validity of a test refers to how well the test measures what it intended to measure. The validity of a test is measured with a "P-Value". The "P-Value" determines how difficult the exam items were for top scorers. Point-biserial correlation is also a measure of the validity of an individual question on the exam. A positive point biserial indicated those that scored high on the exam answered the question correctly. A negative point-biserial indicates those students who scored low on the exam answered the question correctly.