id_plot_cov: Marginal Effects Plot for Hierarchical Covariates

In idealstan: Robust Measurement with 'Stan'

Marginal Effects Plot for Hierarchical Covariates

Description

This function will calculate and plot the ideal point marginal effects, or the first derivative of the IRT/ideal point model with respect to the hierarchical covariate, for each item in the model. The function id_me() is used to first calculate the ideal point marginal effects.

Usage

id_plot_cov(
  object,
  calc_param = NULL,
  label_high = "High",
  label_low = "Low",
  group_effects = NULL,
  plot_model_id = NULL,
  pred_outcome = NULL,
  lb = 0.05,
  upb = 0.95,
  facet_ncol = 2,
  cov_type = "person_cov",
  ...
)

Arguments

object	A fitted idealstan object
calc_param	Whether to calculate ideal point marginal effects for a given covariate. If NULL, the default, the function will instead produce a plot of the raw coefficients from the ideal point model. If passing the name of a covariate, should be a character value of a column in the data passed to the id_make function.
label_high	What label to use on the plot for the high end of the latent scale
label_low	What label to use on the plot for the low end of the latent scale
group_effects	Character value for name of column in data by which to subset the data. Must be a column passed to the id_make function
plot_model_id	The integer of the model ID to plot. If NULL and there are multiple model types, facet_wrap will be used to produce multiple plots with one for each model type.
pred_outcome	For discrete models with more than 2 categories, or binary models with missing data, which outcome to predict. This should be the value that matches what the outcome was coded as in the data passed to id_make().
lb	The lower limit of the posterior density to use for calculating credible intervals
upb	The upper limit of the posterior density to use for calculating credible intervals
facet_ncol	If facetting by multiple models or grouped factors, sets the number of columns in the multiple plots
cov_type	Either 'person_cov' for person or group-level hierarchical parameters, 'discrim_reg_cov' for bill/item discrimination parameters from regular (non-inflated) model, and 'discrim_infl_cov' for bill/item discrimination parameters from inflated model.
...	Additional argument passed on to id_me

Details

The ends of the latent variable can be specified via the label_low and label_high options, which will use those labels for item discrimination.

Note that the function produces a ggplot2 object, which can be further modified with ggplot2 functions.

Value

A ggplot2 plot that can be further customized with ggplot2 functions if need be.

Examples

# Fit a model with a person-level covariate
data('senate114')
senate114$cast_code <- ifelse(senate114$cast_code=="Absent", NA,
                              as.integer(senate114$cast_code) - 1L)
senate114$age <- (2018 - senate114$born - mean(2018 - senate114$born)) / 10
sen_cov <- id_make(senate114, outcome_disc='cast_code',
                   person_id='bioname', item_id='rollnumber',
                   group_id='party_code', person_cov=~party_code+age)
sen_cov_est <- id_estimate(sen_cov, model_type=1, fixtype='vb_full', 
                use_method="pathfinder", nchains=2, ncores=4)

# Plot marginal effects of age on ideal points
id_plot_cov(sen_cov_est, calc_param='age')

# Plot the raw person-level covariate coefficients
id_plot_cov(sen_cov_est)

idealstan documentation built on May 13, 2026, 1:08 a.m.