forum.smartpls.com

Hi, I think that you are right, but the explanation should be done more in terms of mediators than paths. For example: Model 1) IV --> (V1) --> DV [R² = 0.20] Model 2) IV --> (V1 and V2) --> DV [R² = 0.30] Model 3) IV --> (V1, V2 and V3) --> DV [R² = 0.35] Probably the total effect of IV is bigger i...

Hi, These results are uncommon, but if you are assessing 2nd order LV, the explanation is simple: SmartPLS uses the outer loadings or the repeated indicators instead the values between 2nd order LV and its 1st order LV (they are reported as path coefficients). See: http://forum.smartpls.com/viewtopi...

Hi,

Inter-item correlations must be computed in SPSS, R, etc.

SmartPLS just show us the LV correlations matrix.

Best regards,

Bido

Hi,
if you want to delete your account, just send an e-mail to Christian Ringle.
c.ringle@smartpls.de

Best regards,
Bido

Hi, You should look at outer loadings too. As the algorithm is “partial”, the measurement model is computed in one step and structural model in the next, until the convergence. For this reason, sometimes (rare) I found: - Positive path coefficient and all outer loadings of one LV negative - When I r...

Hi, CFA means that we do not have a structural model. All LV are connected on with each other, we use factor weighting scheme, and do not use the path coefficients, just the LV correlations and results of measurement model: Convergent validity: outer loading > 0.7 || AVE > 0.5 Discriminant validity:...

Hi, If HOCs are formative, it does not make sense compute AVE and composite reliability. We do not expect that their LOCs will be correlated, for this reason they are used as formative. If HOCs were reflective, we should compute their AVE and CC based on their relations with LOCs (they are presented...

Hi, the estimation of unstandardized LV in SmartPLS has a bug (see: http://forum.smartpls.com/viewtopic.php?t=812 ), for this reason, if you used these scores to compare SmartPLS with SPSS, the results are different. To have SPSS = SmartPLS: 1) Run your model in SmartPLS 2) Copy/paste the standardiz...

Hi,
just copy the "Key" in the right green corner -->

and paste it when is requested.

Best regards,
Bido

Hi, 1) yes PLS-PM is not a “silver bullet”, see: Marcoulides, G. A., & Saunders, C. (2006). Editor’s comments: PLS: a silver bullet? MIS Quarterly , 30(2), iii–ix. 2) Chin and Newsted (1999) suggest power analysis to sample size determination. Chin, W. W., & Newsted, P. R. (1999). Structural...

Hi,
the Sam's explanation is correct,
if we used the "search" function (green corner), there are other examples, usually with dummy variables (almost constant).

Best regards,
Bido

Hi,
see:
viewtopic.php?t=640
viewtopic.php?t=288

Best regards,
Bido

Hi,

It’s just regression, see one example here:

PPT = https://app.box.com/s/bd22560c4efd4a1eef50

XLS = https://app.box.com/s/6f69efce3761aebf884c

It is in Portuguese, sorry.

Best regards,

Bido

Hi, Inter-construct = latent variable correlations (see the Report/Quality Criteria/Latent…) Intra-construct = inter-items correlations = use SPSS, R or other statistical software to compute these correlations = item-to-total correlations = use SPSS / analyze / scale / reliability Best regards, Bido

Hi, If you want to compute the contribution of each independent variable.. Example: y = b1*x1 + b2*x2 +b3*x3 Contribution of x1 = ry1 * b1 Contribution of x2 = ry2 * b2 Contribution of x3 = ry3 * b3 Were: ry1 = correlation between x1 and y. b1 = path coefficient Sum of these contributions = R² Best ...