The literature regarding the shortcomings of categorization of moderating variables and also the use of 1 SD above and below the mean is quite understood and clear. My concern, however, is that is there an alternative to the above two options, giving that using the full range of the continuous moderating variable, to me, will produce an infinite number of range along which X relation to Y varies. Will it be possible to graph the effect without some form of categorization of the continuous moderating variable?
Yes and It is the region of significance. Take a look here: Bauer, D. J., & Curran, P. J. (2005). Probing Interactions in Fixed and Multilevel Regression: Inferential and Graphical Techniques. Multivariate Behavioral Research, 40(3), 373-400.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Graphs