designICT: 'designICT' class generator

In ICTatRTI/PersonAlyticsPower: Power Analysis and Simulation for PersonAlytics

designICT class generator

Description

The designICT class is a parent class to polyICT and other future planned classes such as expICT for exponential growth models and logisticICT for logistic ICT growth models. The designICT class contains methods (functions common to all child classes) and fields (variables common to all child classes) and is not to be used directly.

Methods

print

A method for printing a the primary inputs for an ICT design.

designCheck

This method simulates one dataset using the current ICT design and plots average trajectories and histograms of the outcome by phase and by group. This allows the user to visually check whether the inputs they provided match their intendend design. The default is to do this with large samples (N=5,000 per group) to show the asypmtotic behavior. For comparison, smaller sample sizes can also be used to see the possible deviations from the population. If smaller sample sizes are used, this should be repeated several times to prevent over-interpretation of one sample. See the examples in polyICT.

file Character. A file name for saving the resulting data. The default is NULL, in which case the data are not saved.

ylim Numeric vector of length 2. Limits for the y-axis. This is useful when variance is high and/or effect sizes are small, making it difficult to visualize average change over time. The default is NULL, in which case default limits are used.

fitMod Logical. Should the model implied by the design be fit to the simulated data? The default is FALSE.

seed Numeric. A random seed for enabling replication. The default is 123.

npg Numeric (integer). The number per group. The default is n=5,000 per group. If a small value is used, the user should repeat the check with several different seeds. See the examples in polyICT.

return Character. What should be returned. The default is the designCheck plot via return='plot'. Other options are 'model' which returns model fit and descriptive statistics; 'stats' which returns descriptive statistics only; 'data' which returns the simulated data; 'datastat' which returns the data and the descriptive statistics; and 'all' which returns everything in a list.

justData Logical. Should only the data be generated (and not a model or plot). This speeds up other calls to designCheck, e.g., by setupDist.

Methods

Public methods

• designICT$print()

• designICT$designCheck()

• designICT$clone()

---

Method print()

Usage

designICT$print(...)

---

Method designCheck()

Usage

designICT$designCheck(
  file = NULL,
  ylim = NULL,
  fitMod = FALSE,
  seed = 123,
  npg = 2000,
  return = "plot",
  justData = FALSE,
  type = "histogram",
  title = "Design Check"
)

---

Method clone()

The objects of this class are cloneable with this method.

Usage

designICT$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Author(s)

Stephen Tueller [e-mail]

ICTatRTI/PersonAlyticsPower documentation built on Dec. 13, 2024, 11:08 p.m.