We often use attitude scales developed by others after minor amendments. Since these scales are developed after an intensive research, is it sufficient to check mere validity and reliability of data after such changes?
I believe that psychometrically validated tools in the literature would first need to be piloted to check for cultural differences in interpreting questions, contexts etc...particularly if developed for studies in one country and are to be used in other countries. Additionally, some tools would need to be translated to another language obviously needing back-translation for validating the translated tool. After this, one would still need to check for reliability, and validity. If structural equation modeling is used, one would need to check construct reliability and average variance extracted.
The attitude scales are generally designed and validated. But applicability of these scales will differ from respondent's backgrounds. So if we modify them then at least we should conduct a pilot study first, check reliability and the go on with further research
Construct validity and reliability would be the issues to be addressed. But if you get impressive results on construct validity and reliability, then it will be a great contribution. The idea is worth exploring.
It is very common to adapt well established scales to fit a specific context. A rigorous way of adapting these scales is pretty similar to developing a new scale, as you will need to examine the validity and reliability of your "new" scales. Unfortunately, many articles using adapted measures do not provide specific information of the procedure that they employed for adaptation. For example, we often read articles whose authors say they replace word "A" with word "B" to fit their contexts, or delete some adjectives or phrases, or reframe some instructional statements.... It appears that when the validity and reliability are shown to be acceptable, these ways of adaptation would not prevent these studies from being published or called good research. We could also see a number of such articles published in top-tier journals. But I believe a pilot study used to test the adapted measures would be very helpful (particularly if the original measure was developed in a different cultural context).
- Badges
- Science topic
- Similar topics
- Psychology