Dear All,
In my basic model I have 3 endogenous latent variables (DV_1, DV_2 and DV_3) and 4 exogenous latent variables each measured by 5 indicators (IV_1, IV_2, IV_3 and IV_4). Reflective measurement models have been chosen for the DV’s and formative for the IV’s.
IV_2 indicators are 5 multiplicative indices built through the multiplication of IV_1 indicator values and IV_5 indicators, both measured on a 7-point Likert scale.
The relationships between the exogenous and endogenous variables are modelled as follows:
IV_1 => DV_1
IV_2 => DV_1
IV_3 => DV_2
IV_4 => DV_2
DV_1 => DV_3
DV_2 => DV_3
Following Hair et al. (2014), I conducted collinearity diagnostics for 3 sets of variables:
Set 1: IV_1 and IV_2 as predictor for DV_1
Set 2: IV_3 and IV_4 as predictor for DV_2 and
Set 3: DV_1 and DV_1 as predictor for DV_3
For the first set VIF is 4.1, while for the other sets its close to one. May this cause problems in the estimation of the model?
Many thanks in advance for your support!
Collinearity Outer Model
-
- PLS User
- Posts: 19
- Joined: Sun Jul 31, 2011 3:49 pm
- Real name and title: