Hi everybody,
I am at the end of my analysis, but stucked with the interpretation.
Could you give me a hint on how to interpret very low, but significant path coefficients (0.11) in relation to dependent variables with R squares that still reach the minimum level of 0.19?
--> Why are the path coefficients so low? Shouldn't a 0.19 R² reflect that the variance in that construct is explained at least to a little extent by the relation between the independent and dependent variable?
(I included two independet variables as control variables for the dependent variables, since the R squares were actually first below 0.19. This caused the R squares to increase and to reach the 0.19 level, however the path coefficients dropped to below 0.2 as mentioned above (before they were 0.21-0.22). Should I still leave the control variables in the model?)
Thank you very mich in advance for any help.
Best regards
Michelle
low path coefficients with acceptable R²?
-
- PLS Junior User
- Posts: 5
- Joined: Tue Aug 03, 2010 9:53 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 Michelle,
Remembering that:
R2 = b1*r1y + b2 * r2y + .... + bn*rny
Where:
b1 = path coefficient from x1 to y
r1y = correlations between x1 and y
The path coefficient goes down when we have multicollinearity (in your model, correlation between independent variable and the control variables)
Best regards,
Bido
Remembering that:
R2 = b1*r1y + b2 * r2y + .... + bn*rny
Where:
b1 = path coefficient from x1 to y
r1y = correlations between x1 and y
The path coefficient goes down when we have multicollinearity (in your model, correlation between independent variable and the control variables)
Best regards,
Bido
-
- PLS Junior User
- Posts: 5
- Joined: Tue Aug 03, 2010 9:53 pm
- Real name and title: