Main effect in a bayesian model

Modeling
1

Hello,

I have the following bayesian model:

fit <- brm(response ~  0  + Intercept + language:contrast + (1 | subject), data = test, family = bernoulli("logit"), prior = prior, sample_prior = "yes")

Family: bernoulli 
  Links: mu = logit 
Formula: response ~ 0 + Intercept + language:contrast + (1 | subject) 
   Data: test (Number of observations: 2400) 
  Draws: 4 chains, each with iter = 2000; warmup = 1000; thin = 1;
         total post-warmup draws = 4000

Group-Level Effects: 
~subject (Number of levels: 50) 
              Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS Tail_ESS
sd(Intercept)     0.92      0.12     0.71     1.18 1.00     1133     1462

Population-Level Effects: 
                              Estimate Est.Error l-95% CI u-95% CI Rhat Bulk_ESS
Intercept                         0.56      0.62    -0.62     1.81 1.01      461
languageCG:contrasthad_hard      -0.86      0.65    -2.15     0.43 1.00      492
languageRP:contrasthad_hard       0.90      0.71    -0.47     2.25 1.00      595
languageSMG:contrasthad_hard     -1.88      0.67    -3.23    -0.57 1.01      472
languageCG:contrasthead_heed      4.12      0.95     2.42     6.14 1.00      791
languageRP:contrasthead_heed      3.99      1.21     1.85     6.58 1.00     1426
languageSMG:contrasthead_heed     3.52      0.81     1.96     5.12 1.00      686
languageCG:contrasthid_heed      -1.06      0.64    -2.32     0.22 1.00      466
languageRP:contrasthid_heed       2.09      0.77     0.57     3.56 1.00      667
languageSMG:contrasthid_heed     -2.32      0.68    -3.69    -1.06 1.01      480
languageCG:contrasthoard_hod      0.05      0.64    -1.24     1.30 1.00      484
languageRP:contrasthoard_hod      1.08      0.71    -0.32     2.46 1.00      581
languageSMG:contrasthoard_hod    -0.73      0.66    -2.06     0.51 1.01      473
languageCG:contrasthud_hard      -1.00      0.64    -2.28     0.25 1.01      469
languageRP:contrasthud_hard       0.51      0.70    -0.89     1.88 1.00      586
languageSMG:contrasthud_hard     -1.14      0.67    -2.48     0.15 1.00      489
languageCG:contrastwhod_hood     -0.98      0.65    -2.23     0.25 1.01      493
languageRP:contrastwhod_hood      0.74      0.71    -0.67     2.10 1.00      587
languageSMG:contrastwhod_hood    -1.84      0.67    -3.20    -0.59 1.00      484
                              Tail_ESS
Intercept                          888
languageCG:contrasthad_hard        952
languageRP:contrasthad_hard       1274
languageSMG:contrasthad_hard       947
languageCG:contrasthead_heed      1837
languageRP:contrasthead_heed      2151
languageSMG:contrasthead_heed     1691
languageCG:contrasthid_heed       1121
languageRP:contrasthid_heed       1571
languageSMG:contrasthid_heed      1112
languageCG:contrasthoard_hod      1191
languageRP:contrasthoard_hod      1077
languageSMG:contrasthoard_hod      951
languageCG:contrasthud_hard       1127
languageRP:contrasthud_hard       1241
languageSMG:contrasthud_hard      1032
languageCG:contrastwhod_hood      1189
languageRP:contrastwhod_hood      1277
languageSMG:contrastwhod_hood      929

I would like to find the main effect of contrast (all contrasts) on languageCG. I’m a little bit confused about the code, so any help would be appreciated.

Best,

George

2

You might find that checking out the emmeans package helps you out here.

You could try looking at something like:

emmeans(fit, specs = c("contrast")
emmeans(fit, specs = c("contrast", "language")