Hi everyone,
I am conducting a study testing the UTAUT2 model in a smartwatch context. I am using SmartPLS3, which allows you to use age as a continuous moderator.
Is this a valid approach to testing a moderator?
It allows me to retain my entire sample size of 330 participants. Otherwise, I would only be able to test the moderating effect across 2 age groups.
What is everybody's opinion on this?
Thank you!
Yes - no reason why you can't use age as a moderator (and this is better than using age categories in terms of statistical power and for other reasons). It may be helpful to centre the age before adding it to the model as this can aid interpretation (as interpreting coefficients when age = 0 is rarely intuitive without centrering or similar).
Hi Thom,
Thanks so much for getting back to me. Do you have any literature that would help me justify my use of a continuous moderator for the reasons you suggested?
Is the normal process to test as a continuous moderator and then if you find significant results divide the data into groups?
Thank you!
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size