[R] Calculating effectsize or standarized coefficents for gls models in R

Ganjeh, Parisa p@r|@@@g@njeh @end|ng |rom med@un|-goett|ngen@de
Wed Aug 12 11:11:46 CEST 2020


Hi,

I am new in R and I would appreciate if you guide me how I can estimate effect size or standardized coefficients for a gls model (generalized least square) in R. if I can find the way for estimating standardized coefficients is better, because I used effect size package for other models (glm) in my study and got a standardized coefficients as effect size. Unfortunately this package could not give a standardized coefficients or another effect size estimator for gls model. As far as I have searched I could not find a formula in R for calculating effect size or standardized coefficients for my gls model.
My model :
Model1<-gls(Ehyp1~Sex1+SESc1+Ebmi1+Age1+PA1,weights=varIdent(form =~1|PA1), data =X6_17_years_Wave1, na.action = na.exclude)
PA1 is  independent variable and categorical and has 3 levels.
Sex1+SESc1+Ebmi1+Age1: consider as covariates
Sex1 is nominal and has 2 groups (girl and boy).
It is the result of R for my gls model:
Generalized least squares fit by REML
  Model: Ehyp1 ~ Sex1 + SESc1 + Ebmi1 + Age1 + PA1
  Data: X6_17_years_Wave1
       AIC      BIC    logLik
  27156.95 27224.53 -13568.48

Variance function:
Structure: Different standard deviations per stratum
Formula: ~1 | PA1
Parameter estimates:
        3         1         2
1.0000000 0.9268285 0.9310299

Coefficients:
                Value  Std.Error    t-value p-value
(Intercept)  5.957760 0.19569967  30.443382  0.0000
Sex12       -0.832327 0.05159467 -16.132034  0.0000
SESc1       -0.099512 0.00727373 -13.681054  0.0000
Ebmi1        0.021064 0.00837363   2.515553  0.0119
Age1        -0.134280 0.00983979 -13.646654  0.0000
PA12        -0.192089 0.06553337  -2.931163  0.0034
PA13        -0.137575 0.07748981  -1.775395  0.0759

Correlation:
      (Intr) Sex12  SESc1  Ebmi1  Age1   PA12
Sex12 -0.203
SESc1 -0.572 -0.017
Ebmi1 -0.578  0.053  0.145
Age1  -0.230  0.001 -0.001 -0.504
PA12  -0.267  0.116 -0.089  0.005  0.086
PA13  -0.339  0.138 -0.038  0.028  0.179  0.621

Standardized residuals:
        Min          Q1         Med          Q3         Max
-2.34330971 -0.76897681 -0.07360691  0.62658220  3.13138556

Residual standard error: 2.146151
Degrees of freedom: 6363 total; 6356 residual


	[[alternative HTML version deleted]]



More information about the R-help mailing list