Page 1 of 1


Posted: Wed Feb 08, 2017 3:01 pm
by soleimanileila24
I am studying about ERP system effect on business process.for doing that, i want to compare before and after ERP implementation and if there significant difference, means that ERP implementaion has significant difference on my model, i consider ERP system as independent variable and process as dependent variable. my model is first order reflective ,and second order formative.for process i use three latent variable and questionnaire's questions measure these three latent variables.for model defining, i allocate each question for each three latent variables,but i do not know what should i do for my dependent variable which is ERP systems before and after implementation and which indicator should i use for dependent variable which is consisting of three latent variables without any question.fro ERP system before and after implementation, should i use group?if i use group which indicator should i allocate for independent variable?what should i do for my dependent variable(process)?
best regards


Posted: Thu Feb 09, 2017 8:24 am
by jmbecker
For your second-order construct you can use the repeated-indicator approach or the two-stage approach. The following paper explains the differences and which approach should be used under which conditions:
Becker, J. M., Klein, K., & Wetzels, M. (2012). Hierarchical latent variable models in PLS-SEM: guidelines for using reflective-formative type models. Long Range Planning, 45(5), 359-394.

For your independent variable, I would use a dummy-coded variable: zero before and 1 after the implementation.