I analysed data using Amos software and in between one independent and dependant variable, I got negative correlation value and positive regression value. There is no multicollinearity problem between these variables. How can I interpret this kind of result?
It is spirited to illustrate the relationship between features to assign the pairs with high positive or negative correlation. hence, the features can be defined as the Pearson Coefficient into expressions; correlation values near to 1 appears robust and direct correlation between the two features. Otherwise, when the correlation values near to -1 appear robust yet inverse correlation between two features
When independent variables are correlated with each other, there is a presence of Multicollinearity. The results show there is no multicollinearity since a VIF less than 10 as show a high value of VIF as shown in Figure
Basically the correlation coeff. says that your two variables are inversely related. The positive regression value I assume is R^2 which is described here:
https://www.google.com/search?client=firefox-b-1-d&q=What+is+a+regression+R%5E2
Best wishes, David Booth
Without knowing anything about your model and how your variables are related (this would be helpful!), this could be some form of suppression effect, where the sign of the bivariate correlation and the regression coefficient changes. Please see Cohen, Cohen, West & Aiken - Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences.
- Badges
- Science topic
- Similar topics
- Philosophy
- Esthetics
- Interpretation