In a survey, I used a measurement of uncertainty as a key dependent variable, and after implementing the measurement in 2 waves of my survey, I found a better measurement and this new measurement has better internal validity and fits better theoretically. I still have 2 waves of survey to be sent and I want to add the new measurement.

How to proceed with the analysis? I will likely have 2 waves of data with the old measures, and 2 waves with both the old and the new measures. The waves of data are just batches of sampling, but it seems all 4 waves are needed to have a good enough sampling size.


Changing measures in the middle of a survey seems a very dangerous thing to do. Are the advantages of the new measure so overwhelming that they justify all the difficulties this would create? You can of course still run the analysis with the original measure as planned and report the results obtained with the new one as exploratory findings.

In any case, this paper describes a more sophisticated approach to this sort of issues: Curran P.J., & Hussong A.M. (2009). Integrative data analysis: the simultaneous analysis of multiple data sets. Psychological Methods, 14 (2), 81-100.

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.