Comparison of beta weights of different regression models possible?

[alexs96]

Dear all,

first of all. sorry for my english. I'm not a native speaker.

For my masters degree I am running several bivariate linear regressions (in addition to a multiple regression model) to measure the impact of biases on controlling processes. These models all are significant. But I would like to measure which cognitive bias (several items, each one represents one bias) has the greatest impact on controlling processes. Can I compare the beta weights of different bivariate regression models to find the bias with the biggest impact? My dependent variable always remains the same. Unfortunately, I cannot combine the individual variables in only one model.

I have searched for hours for information but unfortunately found contradictory information.

Thanks in advance!
Alexandra

 

[hlsmith]

What is the setting? If you simulated these data, you would know the truth and could sleuth out the biases' impact based on a quantification of the amount of bias in multiple settings.

 

[alexs96]

What is the setting? If you simulated these data, you would know the truth and could sleuth out the biases' impact based on a quantification of the amount of bias in multiple settings.

Hey,
I have a sample of 100 subjects. I measured a team's bias intensity reflectively on a seven-point Likert scale by combining five biases (each item representing the impact of a bias) into one variable. I measured the quality of three different processes. Now I would like to measure which bias has the greatest impact on which process.

If i run a multiple regression with this five items three of them are not significant.

 

[spunky]

You can use Vuong's likelihood-ratio test. It's particularly useful for non-nested models.