Hi everybody,
I am testing a model where Value Proposition, together with Cost are antecedents of Overall Value. The latter, together with Perceived Quality affect Satisfaction.
Thus, I obtain two R squared: the first is equal to 37,5 and the second is 49,0. Can I consider such values robust? Is there any rule of thumb for regression with PLS to determine a minimum level of R suqared.
Thanks in advance.
Fabio
R squared threshold
-
fabio.colacchio
- PLS Junior User
- Posts: 7
- Joined: Thu Mar 22, 2007 4:47 pm
- Real name and title:
-
Diogenes
- PLS Super-Expert
- Posts: 899
- Joined: Sat Oct 15, 2005 5:13 pm
- Real name and title:
- Location: São Paulo - BRAZIL
- Contact:
Hi Fabio,
bigger R2 are better, but depends on the strenght of the relations and how many exogenous variables are in the model.
Each area of the knowledge must have practical tresholds, for example, I have colected some information about Costumer Satisfaction Index:
Year 1992) R2 = 0,44 a 0,52 with 2 Exogenous LV
Year 1996) R2 = 0,70 a 0,80 with 3 Exogenous LV
Year 2000) R2 = 0,79 with 4 Exogenous LV --> R2 was pre-specified to be bigger than 0,65 in this study
Year 2004/5?) R2 = 0,60 with 4 Exogenous LV
Year 2005) R2 = 0,67 with 4 Exogenous LV
Your study has 2 antecedents like Fornell (1992), there he said that with only 2 variables, a R2 like this was Ok.
Best regards.
Bido
FORNELL, C. (1992) A national customer satistaction barometer.. Journal of Marketing, v.56, n.1
FORNELL, C. et al (1996) The American Customer Satisfaction Index...Journal of Marketing, v.60, n.4
KRISTENSEN, K. et al. (2000) Customer satisfaction measurement at post Denmark... Total Quality Management, v.11, n.7
LAURO, C. VINZI, V. E. (2004/5?) Some contributions to PLS Path Modeling... (google)
TENENHAUS, M. et al. (2005) PLS path modeling. Computational Statistics & Data Analysis. v.48.
bigger R2 are better, but depends on the strenght of the relations and how many exogenous variables are in the model.
Each area of the knowledge must have practical tresholds, for example, I have colected some information about Costumer Satisfaction Index:
Year 1992) R2 = 0,44 a 0,52 with 2 Exogenous LV
Year 1996) R2 = 0,70 a 0,80 with 3 Exogenous LV
Year 2000) R2 = 0,79 with 4 Exogenous LV --> R2 was pre-specified to be bigger than 0,65 in this study
Year 2004/5?) R2 = 0,60 with 4 Exogenous LV
Year 2005) R2 = 0,67 with 4 Exogenous LV
Your study has 2 antecedents like Fornell (1992), there he said that with only 2 variables, a R2 like this was Ok.
Best regards.
Bido
FORNELL, C. (1992) A national customer satistaction barometer.. Journal of Marketing, v.56, n.1
FORNELL, C. et al (1996) The American Customer Satisfaction Index...Journal of Marketing, v.60, n.4
KRISTENSEN, K. et al. (2000) Customer satisfaction measurement at post Denmark... Total Quality Management, v.11, n.7
LAURO, C. VINZI, V. E. (2004/5?) Some contributions to PLS Path Modeling... (google)
TENENHAUS, M. et al. (2005) PLS path modeling. Computational Statistics & Data Analysis. v.48.