Hi all,

I know it is wise to talk about the reliability of measures used in any study. What is the best way to discuss validity, and its various form, since it can't be computed on SPSS. Is there any other way aside from looking at previous research which may have used the same questionnaires?

Thanks!

There are many different forms of validity and reliability for questionnaires. Validity can be divided into face and content validity (both qualitative) and then criterion and construct (divergent, convergent and hypothesis testing) validity. Reliability can be divided into two main types: temporal and internal consistency. When looking for evidence for validity of questionnaires, there are usually validation studies available (remembering that psychometrics is sample dependent - classical test theory at least). If the questionnaire developer has been good, he/she will also have establish evidence for validation in the actual process of developing the questionnaire. Remember, a questionnaire is only as good as the evidence that proves that it measures what it is supposed to measure (validity) consistently (reliability).
Delta, which questionnaires are you talking about? (sprout)

I haven't yet written up my 'making a questionnaire' chapter - so my mind is pretty blank on this, but you can definitely get 'split half' reliability and 'test retest' reliability so you could test those.

======= Date Modified 14 Jun 2010 16:18:29 =======
Thanks Sneaks and Wally. You've both been very helpful. Wally, Ill send you a PM.