lsttheory_es: Estimate latent state-trait models for experience sampling...
In amayer2010/lsttheory: Latent State-Trait Theory

lsttheory_es

R Documentation

Estimate latent state-trait models for experience sampling data

Description

Estimate latent state-trait models for experience sampling data

Usage

lsttheory_es(
  model,
  ntimepoints,
  nperiods = 1,
  data,
  addsyntax = "",
  la_t_equiv = NULL,
  la_o_equiv = NULL,
  la_s_equiv = NULL,
  vzeta_eqiv = NULL,
  veps_equiv = NULL,
  vtheta_equiv = NULL,
  nu_equiv = NULL,
  alpha_equiv = NULL,
  mtheta_equiv = NULL,
  gamma_t_equiv = NULL,
  manifest_thetacovariates = NULL,
  ...
)

Arguments

`model`	integer or character. Can be 1-9 or one of the following: c("MSSTinvar", "STARinvar", "PTARinvar", "ITARinvar", "PITARinvar", "STAR2", "PTAR2", "ITAR2", "PITAR2"). Each model implies invariance assumptions, but details can be changed with the options below. Model 1 ("STinvar" or "MSSTinvar") is a multistate-singletrait model (no autoregression). Models 1-5 assume state (residual)- and trait-equivalence, i.e., all factor loadings are fixed to 1 and all intercepts are fixed to 0. Model 2 ("STARinvar") is a singletrait model with autoregression. Model 3 ("PTARinvar") is a model with period-specific (usually day-specific) traits. Model 4 ("ITARinvar") is a model with indicator-specific traits. Model 5 ("PITARinvar") is a model with period- and indicator-specific traits. Models 6 - 9 ("STAR2", "PTAR2", "ITAR2", "PITAR2") are the same as models 2-5, but with state (residual)-congenericity and measurement invariance within periods. In experience sampling data, periods usually correspond to days.
`ntimepoints`	integer. The total number of measurement occasions on which data was collected.
`nperiods`	integer. The number of periods (e.g. days or weeks) on which data was collected.
`data`	a data.frame. This data frame contains the observed variables, sorted by time t and then by indicator i, i.e., Y11, Y21, Y31, ... Y12, Y22, Y32 ... Y15, Y25, Y35 ... etc.
`addsyntax`	character string. Will be added to generated lavaan syntax.
`la_t_equiv`	Character. Invariance option for factor loadings of the latent trait. Can be one of `c("one", "period.invar", "free")`.
`la_o_equiv`	Character. Invariance option for factor loadings of the occasion factor (OCC). Can be one of `c("one", "invar", "period.invar", "free")`. In models without autoregression this corresponds to the factor loadings of the state variables (for models with a single trait or period-specific traits) or state residual variables (for models with indicator-specific traits).
`la_s_equiv`	Character. Invariance option for autoregression between occasion factors. Can be one of `c("zero", "invar", "interval.invar", "free")`.
`vzeta_eqiv`	Character. Invariance option for variances of the state residual (zeta) variables. Can be one of `c("invar", "period.invar", "free")`.
`veps_equiv`	Character. Invariance option for variances of the residual (epsilon) variables. Can be one of `c("invar", "time.invar", "indicator.invar", "period.invar", "free")`.
`vtheta_equiv`	Character. Invariance option for variances of the latent trait. Can be one of `c("invar","indicator.invar", "free")`.
`nu_equiv`	Character. Invariance option for intercepts of the indicators. Can be one of `c("zero","period.invar", "free")`.
`alpha_equiv`	Character. Invariance option for intercepts of the latent states. Only relevant for models with a single trait or period-specific traits. Can be one of `c("zero","period.invar", "free")`.
`mtheta_equiv`	Character. Invariance option for means of the latent traits. Character. Can be one of `c("invar","indicator.invar", "free")`.
`gamma_t_equiv`	Character. Invariance option for regression coefficient from covariates to the latent trait variables.
`manifest_thetacovariates`	Vector or single character. Name or the variable (or variables) in the dataset which are covariates that further explain the trait variables in the model.
`...`	Further arguments passed to lower-level functions