6 b ii). Analyze the characteristics of a good test.
- Characteristics of a good test
There are basic characteristics to consider in test construction. These are validity, reliability, objectivity, usability and discrimination.
It refers to the degree to which the test measures what it is supposed to measure. In other words, it should adequately sample the content and the objectives of the areas the teacher has taught. In test construction, the teacher should be guided by the following key questions concerning the validity:
- Does the test adequately sample the content area?
- Does the test involve the skills that are important in the area e.g. cognitive, affective, and psycho-motor.
- Does the test relate to all behavioral objectives for the unit?
If the test cannot provide this essential information, then it is of little value and therefore should not be used. The validity depends on the purpose of the test, the procedure and the interpretation of the results.
This is the accuracy and consistency with which a test measures what it measures. In other words, the test should generate quite similar scores when it is administered to a similar group of students on two different occasions. If the test gives nearly the same results on the two successful occasions, then it is concluded that it possesses high degree of consistency and reliability. On the other hand, if it cannot produce the results that are consistent, then the results cannot be trusted or relied upon.
The reliability of a test depends on:
- The conditions under which it is administered.
- The nature of the attributes being measured,
- The characteristics of the group of students being assessed.
- The way in which the lest is scored,
It refers to freedom from subjective judgment, The test must be as free as possible from bias, A measuring instrument has objectivity when two or more competent observers can independently agree whether or not a learner’s test performance meets the standard criteria stated in the learning objective.
Objectivity of a test will be affected by: type of tests, mode of scoring and state. of ritmd and condition of the scorer.
- Usability (Economy)
This refers to the practical aspects of the time and resources required for the test compared to the value of the information obtained. Sometimes, the time and resources involved in preparing a test compared to the value of the information obtained may render the test useless.
M refers to the ability of a test to separate students on how well they perform in the test. The questions set must be representative of the subject matter taught and should be sufficiently difficult so that those who, blow the subject can be separated from those who do not know it.