sim_control: Set control parameters for simulated data

In strategicdataproject/OpenSDPsynthR: Package to generate synthetic data for the OpenSDP Project

Description

Set control parameters for simulated data

Usage

sim_control(nschls = 2L, best_school = NULL, race_groups = NULL,
  race_prob = NULL, ses_list = NULL, minyear = 2002,
  maxyear = 2017, n_cohorts = 8L, gifted_list = NULL,
  iep_list = NULL, ell_list = NULL, ps_transfer_list = NULL,
  grade_levels = NULL, school_means = NULL, school_cov_mat = NULL,
  school_names = NULL, postsec_names = NULL,
  gpa_sim_parameters = NULL, grad_sim_parameters = NULL,
  ps_sim_parameters = NULL, assess_sim_par = NULL,
  assessment_adjustment = NULL, grad_adjustment = NULL,
  ps_adjustment = NULL, gpa_adjustment = NULL, assess_grades = NULL,
  n_postsec = 35L, postsec_method = "scorecard")

Arguments

nschls	integer- number of schools to create, default is 2
best_school	character, format is a number, padded with a leading 0, indicates the school that will be the highest performing
race_groups	character - vector of labels for race groups
race_prob	numerics - same length as race_groups
ses_list	a probability list defining probabilities of being low socioeconomic status, see Details
minyear	integer - the first year student records are observed
maxyear	integer - the last year student records are observed
n_cohorts	integer - the number of birth-year cohorts to produce
gifted_list	a probability list defining the probability of being in a gifted and talented program
iep_list	a probability list defining probabilities of being on an individualized education plan
ell_list	a probability list defining probabilities for being an English language learner
ps_transfer_list	a probability list for transferring postsecondary institutions after enrolling
grade_levels	a probability list for grade promotion and retention
school_means	numeric - a named vector of means for school level attributes
school_cov_mat	matrix - a covariance matrix for the school level attributes
school_names	character - a vector to draw school names from
postsec_names	character - a vector to draw postsecondary institution names from
gpa_sim_parameters	list - parameters to pass to gen_outcome_model
grad_sim_parameters	list - parameters to pass to gen_outcome_model
ps_sim_parameters	list - parameters to pass to gen_outcome_model
assess_sim_par	list - parameters to pass to gen_outcome_model
assessment_adjustment	list - parameters to adjust assessment scores by for bias
grad_adjustment	list - parameters to adjust graduation probabilities by for bias
ps_adjustment	list - parameters to adjust postsecondary enrollment probabilities by for bias
gpa_adjustment	list - parameters to adjust gpa for bias
assess_grades	character - grade levels to generate assessment scores for
n_postsec	numeric - number of postsecondary schools to assign to
postsec_method	character - options "scorecard" or NULL

Details

This function has a full set of default values that are designed to produce realistic data. These defaults can be overridden by specifying any of the arguments to be overridden as an option to the function call.

There are two unique data structures that are used to set options for simulations. The first is a probability_list, a list which defines a grouping factor, and for each level of the grouping factor a function and parameters to generate probability distribution from.

The sim_parameters data structure defines the parameters for the outcome simulation. Outcomes are simulated using a simulated multilevel model structure and this data structure contains the parameters that describe the model and the error structure of data generated from that model.

To modify either of these elements, use the validate_probability_list or validate_sim_parameter helper functions to ensure that all of the parameters are defined with valid values.

Value

a named list

strategicdataproject/OpenSDPsynthR documentation built on June 20, 2020, 6:17 a.m.