How do I determine whether a new measure I'm developing is "more valid" than existing measures of the same construct? I'm proposing to develop a new measure that is more inclusive of the experiences of marginalized groups, with the argument that existing measures don't capture their experiences. A reviewer of our proposal wanted to know how we will ultimately determine that our new measure is more valid. I can only find information on establishing validity and reliability, but none on testing whether one measure of a construct is more valid that another measure. Thoughts?
The criterion validity, which comprises concurrent and predictive validity, refers to the comparison of one instrument with another accepted or supposedly "gold standard" instrument that purports to measure the same construct. When assessing concurrent validity, both the instruments are administered at the same time to the same population and then correlated.
Please find attached the references for details.
Streiner, D. L., Norman, G. R., & Cairney, J. (2015). Health measurement scales: a practical guide to their development and use. Oxford University Press, USA.
Younas, A., & Porr, C. (2019). A step-by-step approach to developing scales for survey research. Nurse researcher, 27(1).
Hi,
Validity refers to the degree to which an instrument accurately measures what it intends to measure. Refer the Article Validity of Measures Is No Simple Matter
I understand what validity is. I know how to establish the reliability and validity of measures. I’ve been doing this for over 20 years. That’s not my question. Please read my question. I’ll try to restate it.
A grant reviewer wants to know how we will establish whether our newly developed measure is more valid than existing measures. We have a solid description for establishing its reliability and validity. We need to show that it’s better than the measures that already exist.
Please read my response again. I mentioned that a specific type of validity is used to compare one measure with another measure. You can administer both scales at one time and then compare the correlations. If you have prior literature review, you would have already identified the limitations and strengths of the existing instruments and you would know the instruments that are more widely used.
One way is to compare the psychometric properties of the existing instruments through systematic literature reviews and than compare the properties with your proposed instrument.
- Badges
- Science topic
- Similar topics
- Cognitive Science
- Thinking