How to assess Reliability and Validity of experimental tests

What is reliability and validity?

Reliability=  Refers to the consistency of a piece of research and how likely it would for someone to obtain the same results if they repeated your method.

Validity =  Falls into two branches;

Internal and external

Internal is the extent to which an experiment/test measures what it set out to measure.

External is the generalisability of the findings to others.

1. Test re-test; This is where you administer the same test twice to similar or the same participants over a period of time.You would then correlate the two scores to evaluate the test for stability over time.

2. Inter-rater reliability; Assesses the degree to which different experimenters agree in their scoring decisions. This is particularly useful when the data is subjective such as scoring a piece of art compared to data which isnt subjective such as scores in a maths test.

1. Content validity; Ask experts in that area of research to check the content of your questionnaire or interview Q's for example to make sure they measure what they intend to measure.

2. Face validity; This method is quicker as it is far less rigorous and simply involves looking at your questionnaires or whatever to see if they genuinely measure what they are supposed to.