(“Free” Topic this week…… even harder :S )
Validity – A process of which to ensure your findings.
- having some foundation; based on truth; legally acceptable; logic
- having legal efficacy or force; especially : executed with the proper legal authority and formalities <a valid contract>
- well-grounded or justifiable : being at once relevant and meaningful <a valid theory> logically correct <a valid argument> <valid inference>
It is very important for any study or research to have validity. A test is valid when it actually measures what it intended to measure. Validity is the key to interpret the result, because without validity, the result would not be able to generalize and not representable.
Validity is usually assessed with reliability, however they are different. If the result of a test is reliable, it means the result is consistent and the test can be replicated and will obtain similar results.
For a finding to be valid, it must be accurate and appropriate. [Using the appropriate statistical test to analyze the data]. Results can be reliable, but surprisingly, it might not be valid, this is due to factor A might not truly effected by factor B for example, but it just happens that either the it is a positive skewed or negative skewed graph, but yet, if results are not reliable, results are definitely Not valid, and of course, if results are not reliable, validity does not exist in this case.
Once, if your results are valid (tested it over and over again), it is reliable, representable enough to generalize the population, then you can be confident to draw our your findings as conclusion.