My model consisted of one independent variable, two moderating variables, and one dependent variable. Which technique is better to use in Hierarchical Linear Regression, SEM or Process Macro and Why? Please share with me any paper or link which highlighted the merits and demerits of each technique.
Thanks in advance.
Structural Equation Modeling requires that you have a measurement model, so there is no reason to use that for the analyses you are proposing.
Process Macro is just a pre-built set of commands that will put together the appropriate regressions for you to test moderation. Its results will be identical to a properly specified set of hierarchically ordered linear regressions. (Note there are two different things that sometimes get confused under the label "Hierarchical Linear Regression," and what you proposing is not related to what is better known as "Multi-Level Modeling.")
David L Morgan If I were proposing a theoretical explanatory model wouldn't SEM analysis be more important?
SEM only matters if you have items that represent "unmeasured variables," i.e., there are empirical indicators that are tied to theoretical concepts. In simple terms, you can think of SEM as incorporating factor analysis into a regression, so that the multiple indicators of each concept are included in the overall model.
This is not connected to whether you have a theory that makes predications about how your observed variables will be related to each other.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling