1.1. TEST-RETEST OR STABILITY IS WHETHER THE TEST IS GIVEN TWICE AND A CORRELATION BETWEEN THE FIRST SET OF SCORES AND THE SECOND ARE DETERMINED.
1.2. ALTERNATE FORMS OR EQUIVALENCE IS WHERE IF THERE IS TWO EQUIVALENT TESTS THESE FORMS CAN BE USED TO OBTAINED AN ESTIMATE OF THE RELIABILTY OF THE SCORES FROM THE TEST. BOTH FORMS OF THE TEST ARE GIVEN TO A GROUP OF STUDENTS AND THEN THE CORRELATION BETWEEN THE TWO ARE DETERMINED.
1.3. INTERNAL CONSISTENCY MEANS THAT IF A QUESTIN IN A TEST IS DESIGNED TO MEASURE A SINGLE CONCEPT WHICH MEANS IF A STUDENT GETS A QUESTION RIGHT AND THERE IS A SIMALAR QUESTION ON THE TEST IT IS SAFE TO PRESUME THEY WILL GET THE SIMILAR QUESTION CORRECT.
2. CONSTRUCT VALIDITY EVIDENCE IS WHERE THE TEST IS CONSTRUCTED IN A WAY THAT ITS RELATIONSHIP TO OTHER INFORMATION CORRESPONDS WELL WITH SOME THEORY. A THEORY IS SIMPLY A LOGICAL EXPLANATION THAT CAN ACCOUNT FOR THE RELATIONSHIP BETWEEN TWO VARIBALES.
3. CRITERION-RELATED VALIDITY EVIDENCE IS THE SECOND FORM OF VALIDITY EVIDENCE IN WHICH SCORES FROM A TEST ARE CORRELATED WITH AN EXTERNAL CRITERION. THIS COMES IN TWO FORMS: CONCURRENT AND PREDICTIVE. (CONCURRENT: MEASURES THAT CAN BE ADMINISTERED AT THE SAME TIME.) (PREDICTIVE: HOW WELL A TEST PREDICTS THE FUTURE BEHAVIOR OF THE EXAMINEES.)
4. VALIDITY EVIDENCE
4.1. CONTENT VALIDITY EVIDENCE IS ESTABLISHED BY INSPECTING TEST QUESTIONS TO SEE WHETHER THEY CORRESPEOND TO WHAT THE USER DECIDES SHOULD BE COVERED BY THE TEST.
