sim_trial: Simulate a single cross-sectional trial

In taylordunn/gasr: Simulate and Analyze Goal Attainment Scaling Data

Simulate a single cross-sectional trial

Description

Simulate a single cross-sectional trial

Usage

sim_trial(
  n_subjects = 40,
  sigma_u = 0.5,
  group_allocation = list(control = 0.5, treatment = 0.5),
  random_allocation = TRUE,
  n_goals = NULL,
  n_goals_range = c(3, 6),
  n_goals_prob = NULL,
  sigma_e = 0.5,
  delta = 0.3,
  mean_control_response = -0.3,
  weight_type = "unweighted",
  n_levels = 5,
  score_dist = "norm",
  centre = 0
)

Arguments

n_subjects	Number of subjects.
sigma_u	Standard deviation of the subject-level random effect.
group_allocation	A named list of groups and their probability of assignment. By default, will give equal probability to "control" and "treatment" groups. If NULL, will returns subjects without groups.
random_allocation	Logical. If TRUE (default), randomly assigns to groups based on probabilities in group_allocation. If FALSE, assigns with equal probability. Note that this option will return an error if the subjects cannot be neatly divided. For example, n_subjects = 21 with two groups of probabilities 0.5 and 0.5, will return an error if random_allocation = FALSE because 21 cannot be equally divided in two.
n_goals	The exact number of goals for each subject.
n_goals_range	A vector defining the minimum and maximum number of goals per subjects. Randomly samples the range of goals with uniform probability.
n_goals_prob	A named list with element n_goals defining the possible numbers of goals, and element prob defining the probabilities of getting each number. For example: n_goals_prob = list(n_goals = 1:3, probs = c(0.3, 0.3, 0.4).
sigma_e	Standard deviation of the goal-level random effect, i.e. the random error of each latent goal score.
delta	The size of the treatment effect.
mean_control_response	The mean response in the control group.
weight_type	The type of weights to apply. See 'Details'.
n_levels	The number of levels to use. Defaults to the traditional 5 levels.
score_dist	The distribution by which scores should be approximated. The default "norm" will return thresholds which approximate a standard normal distribution. Choosing "unif" will return thresholds which approximate a uniform distribution.
centre	The centre of the normal distribution from which the thresholds are taken.

Value

A data frame with a row per trial (numbered 1 to n_sim) with list column data containing the simulated data.

taylordunn/gasr documentation built on April 5, 2022, 1:37 a.m.