As Florent points out, there are several different ways that different people define validity. One of the simplest (conceptually) is discussed in the linked publication. Testing it could be done by conducting an experiment in which you manipulate what you are trying to measure.
As Florent points out, there are several different ways that different people define validity. One of the simplest (conceptually) is discussed in the linked publication. Testing it could be done by conducting an experiment in which you manipulate what you are trying to measure.
With regard to reliability, the most commonly used statistic is Cronbach's alpha, and there have been so many questions about that topic that I have complied a set of resources: