1) Test-retest reliability is when a test is taken and after a period of time is retaken it should produce similar scores each time the test is taken. If the test takers scores are the same or very close then the test is considered reliable. Inter-rater reliability is the measurement agreement between two observers (Myers, A, & Hansen, C., 2012). The test-retest reliability is found with one person’s scores are used and the inter-rater reliability is found using multiple individuals.
2) Content validity is when one tries to see whether the measure is mostly representing the quality one is measuring. Predictive validity is when one wants to see if the procedure can yield information that allows one to predict a future behavior. Construct validity is when a test measures up to its claims.
3) Internal validity is the degree to which a researcher can state a causal relationship between an antecedent and the observed behavior. External validity is how the outcome applies in other settings (Myers, A & Hansen, C., 2012).
A test can be reliable without being valid. Reliability is about consistency and/or dependability and validity is about accuracy. I reliability measure is not always valid. Results might be able to be reproduced but they might not be correct. However, a test can not be valid if it is not reliable. Say you were conducting a personality test and you get the same results after testing a person more than once, than the test would be reliable.