Hi all,
I have questions regarding the second order formative endogenous variables. My study model consist of both constructs i.e. reflective (exogenous)-formative (second order endogenous). I used repeated indicator approach for analysis in SEM PLS. Now I want to report the result of outer weights, outer loading, t and p statistics for formative construct. Should I report said results after first order analysis or should I perform second order analysis and then report the said results for formative analysis?
For exogenous variables, formative indicators are not creating troubles in interpretation, however, in case your endogenous variable is formative measurement model ( in your case it is second order construct), then it is difficult to ascertain the impact of the exogenous variable and the effect of those sub dimensions of the endogenous variable. Separation of these two effects would be problematic. Therefore, it is suggested that the paths should go from the Exogenous variables to the sub dimensions of the endogenous variables and then from the sub dimension to the exogenous variable (making it a mediation model). This mediation analysis would give you the true impact of the Exogenous variable on the Endogenous variables.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling