bridgr: Implement bridging correction

In sidakyntiso/bridgr: Bridging the Grading Gap

Implement bridging correction

Description

Implements a Bayesian version of the Aldrich Mckelvey model to grade data.

Usage

bridgr(
  bridgr.dat,
  min_grade = NA,
  max_grade = NA,
  stan_model = NA,
  ITER = NA,
  WARMUP = NA,
  THIN = NA,
  CHAINS = NA,
  CORES = NA
)

Arguments

bridgr.dat	A bridgr.data object.
min_grade	The minimum possible grade (by default is zero.)
max_grade	The maximum possible grade (by default is the maximum observed grade.)
stan_model	Rstan model (see manuscript for default implementation.)
ITER	RSTAN parameter. A positive integer specifying the number of iterations for each chain (including warmup). The default is 20000.
WARMUP	= Rstan parameter. A positive integer specifying the number of warmup (aka burnin) iterations per chain. The default is 1000.
THIN	= A positive integer specifying the period for saving samples. The default is 1,
CHAINS	Rstan parameter. A positive integer specifying the number of Markov chains. The default is 4.
CORES	Rstan parameter. The number of cores to use when executing the Markov chains in parallel. The default is to use the value of the "mc.cores"

Value

bridgr.dat Processed version of the input dataset.

coregradmeds The post-processed (bridged) student grades on the scale of the input dataset.

zmedsrank The post-processed (bridged) student ranks.

student_id Student identifier that can be linked with the input dataset.

Examples

data("bridgr.sim.data")
# Re-structure the input grading dataset.
bridgr.dat <- bridgr.data(df=bridgr.sim.data,student="student",
grader.assigned = "grader.assigned", grader="grader",grade="grade")

# Correct grading bias using bridging observations. Set cores = NA to utilize more CPU cores.
bridgr.sim.results = bridgr(bridgr.dat=bridgr.dat,min_grade=NA,max_grade=NA, CORES = 2)

sidakyntiso/bridgr documentation built on Feb. 11, 2023, 4:18 p.m.