In what situation, a questionnaire could be invalid but its resultant scores could be valid? I received a comment from a journal article reviewer that says:
In keeping with the preferred terminology, the authors are reminded that the [Questionnaire] itself is not reliable and valid, but its resultant scores are. Rephrasing of this is recommended in several places in the manuscript.
What does this preferred terminology refer to? Does it refer to a general preference of attribution of reliability and validity to the resultant scores or perhaps my questionnaire poses a problem? It's good to mention that I removed 9 poor items (with low factor loadings) from 80 items of my questionnaire and the resultant scores are obtained from the remaining 71 items.