# effect size for binomial mixed model's fixed-effects

I have a binomial mixed-effects model that I fit using glmer (package lme4) in R. I want some sort of descriptive statistic for the effect size of each of the two fixed effects (something like Cohen's d for each explanatory variable--var1 and var2 in the code below). Is there a simple way to do this using the model output from glmer? My model and its output below. Thanks for your thoughts!

mod1<-glmer(formula = "response ~ var1 + var2 + (1|population)", data =
data.anon, family = binomial)
summary(mod1)

Generalized linear mixed model fit by maximum likelihood (Laplace
Approximation) [glmerMod]
Family: binomial  ( logit )
Formula: response ~ var1 + var2 + (1 | population)
Data: data.anon

AIC      BIC   logLik deviance df.resid
139.4    149.9    -65.7    131.4       99

Scaled residuals:
Min      1Q  Median      3Q     Max
-1.5010 -0.8794 -0.5695  1.0912  1.8554

Random effects:
Groups     Name        Variance Std.Dev.
population (Intercept) 0        0
Number of obs: 103, groups:  population, 14

Fixed effects:
Estimate Std. Error z value Pr(>|z|)
(Intercept)  -0.2437     0.2374  -1.026   0.3046
var1         -0.7521     0.3317  -2.267   0.0234 *
var2          0.4655     0.2084   2.233   0.0255 *
---
Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

Correlation of Fixed Effects:
(Intr) var1
var1 -0.360
var2 -0.182 -0.322

• what are your independent variables and response variable ? – Subhash C. Davar Dec 29 '17 at 6:38
• My response variable is binomial. My independent variables are var1 and var2. Var1 is an area measurement (originally in meters-squared) and var2 is a metric about the average genetic makeup of a population (originally continuous ranging 1-5). Both var1 and var2 are standardized (centered to their respective means and re-scaled to 1 SD). – G. Spea Dec 29 '17 at 18:59
• construction of var2 is not clear to me ? why do you take a recourse to the standardized variables ? And rescaling. Your respinse variable is binomial? – Subhash C. Davar Dec 30 '17 at 7:43