Obtain beta-weights for omnibus tests in linear mixed models

[st_kilian]

Hi all,

Issue:
I want to obtain beta-weights as effect size measures for fixed effects and/or fixed effects parameter estimates omnibus tests in GAMLj linear mixed models.

Questions:
1. Are beta-weights the appropriate effect size measure in this regard? If not, which are?
2. I assume "estimates" are unstandardized b-values. Can you confirm this?
3. Is there a way to obtain standardized beta-weights within the module or another module? (If not, I would strongly suggest to add it.Thank you)

Kind regards
Kilian

(FYI: This is my first thread, feel free to leave a feedback on my style)

[mcfanda@gmail.com]

Hi
beta coefficients are not uniquely defined in the mixed model, so there's no general rule to compute them that applies to any possible design. If you have a simple design (one clustering variable) you can standardize your independent variables and dependent variable. However, depending on the specific application, you may need to standardize within cluster.

[earlhdges]

Hello.
1. Beta-weights can be used as effect size measures in linear mixed models, including GAMLj. They represent the standardized regression coefficients, indicating the change in the outcome variable (dependent variable) associated with a one-unit change in the predictor variable (independent variable), while controlling for other variables in the model. Beta-weights are commonly used as effect size measures in regression-type analyses. territorial io

2. In GAMLj, the "estimates" typically refer to the unstandardized regression coefficients (b-values). These estimates provide information about the magnitude and direction of the relationship between the predictors and the outcome variable, without taking into account the scale of the variables. These unstandardized estimates can be useful for interpreting the specific effects of the predictors.

3. GAMLj does not directly provide standardized beta-weights within the module. However, you can calculate standardized beta-weights by dividing the unstandardized estimates (b-values) by the standard deviation of the respective predictor variable. This process standardizes the beta-weights and allows for comparisons of the relative strength of the predictors. Alternatively, you can standardize your predictor variables before running the GAMLj analysis, which would result in standardized beta-weights in the output.

[mcfanda@gmail.com]

Great!