Torsten
03-03-2009, 09:22 AM
Hey,
I have a problem concerning multiple regression analyses and the beta coefficient.
I conducted a multiple regression analysis with two predictor variables.
The predictors are not correlated. I found a significant interaction between the two predictors and continued to analyze the interaction further by using the simple slope method (Aiken & West, 1991).
For 1SD below the mean for the moderator, I have a significant slope for the other predictor, predicting the dependent variable.
However, the problem is that the standardized beta is above 1 (1.008).
I only know that something is wrong if that happens, but I don’t know what it is.
Does someone know what the problem is (is it a sign that some assumptions are violated) ?
And if yes, is there a way to solve the problem?
I would appreciate any help. Don’t hesitate to ask me if you need further details about the procedure etc.
Thanks,
Torsten
I have a problem concerning multiple regression analyses and the beta coefficient.
I conducted a multiple regression analysis with two predictor variables.
The predictors are not correlated. I found a significant interaction between the two predictors and continued to analyze the interaction further by using the simple slope method (Aiken & West, 1991).
For 1SD below the mean for the moderator, I have a significant slope for the other predictor, predicting the dependent variable.
However, the problem is that the standardized beta is above 1 (1.008).
I only know that something is wrong if that happens, but I don’t know what it is.
Does someone know what the problem is (is it a sign that some assumptions are violated) ?
And if yes, is there a way to solve the problem?
I would appreciate any help. Don’t hesitate to ask me if you need further details about the procedure etc.
Thanks,
Torsten