The test-retest reliability is when a person takes a test and after a period retakes the test. The reliable measures should reproduce similar scores each time the person is measured. If the person gets the same scores, the test is considered reliable. Interrater reliability is the measurement agreement between different observers. The test-retest reliability is found when one person’s scores are used to test the reliability and the interrater reliability is found using multiple individuals.
Content validity is when we try to assess whether our measure is fairly representing the quality we are measuring. Predictive validity is when we ask if our procedures yield information that enables us to predict a future behavior. Construct validity is easy to understand, this concept is when a theory moves into the realm of experimentation.
Internal validity is to which degree a researcher can state a causal relationship between the antecedent conditions and the subsequent observed behaviors. External validity is how well the outcome will apply to other settings. (Hansen and Myers, 2012a) and (Cuncic, 2021).
An experiment can be reliable without being valid. Reliability has to do with consistency and dependability. The experiment procedures can be solid but also those procedures can produce invalid results consistently, it really all depends on what is being tested. Valid results can also be gotten from unreliable procedures. The trick is producing evidence to highlight the validity (Hansen and Myers, 2012b).