ESEM {latentFactoR}

Estimates Exploratory Structural Equation Model

Description

A general function to estimate an Exploratory Structural Equation Model (ESEM) using the lavaan package. With latentFactoR objects, the function requires fewer inputs

Usage

ESEM(
  data,
  factors,
  variables,
  estimator = c("MLR", "WLSMV"),
  fit_measures = NULL,
  variable_polarity = NULL,
  wording_factor = c("none", "CTCM1", "CTCM1_each", "RI", "RI_each"),
  CTCM1_polarity = c("negative", "positive"),
  ...
)

Arguments

data

Numeric matrix, data frame, or latentFactoR object

factors

Numeric (length = 1). Number of ESEM factors to estimate

variables

Numeric (length = 1 or factors). Number of variables per factor. A vector the length of the number of factors can be specified to allow varying number of variables on each factor (necessary for some wording_factor arguments)

estimator

Character. Estimator to be used in cfa. Default options are "MLR" for continuous data and "WLSMV" for categorical data

fit_measures

Character. Fit measures to be computed using fitMeasures. Defaults to: "chisq", "df", "pvalue", "cfi", "tli", "rmsea", "rmsea.ci.lower", "rmsea.ci.upper", "rmsea.pvalue", and "srmr". Other measures can be added but these measures will always be produced.

If scaled values are available (not NA), then scaled fit measures will be used.

variable_polarity

Numeric/character (length = 1 or total variables). Whether all (length = 1) or each variable (length = total variables) are positive (1, "p", "pos", "positive") or negative (-1, "n", "neg", "negative") polarity on the factor

wording_factor

Character (length = 1). Whether wording factor(s) should be estimated. Defaults to "none". Options include:

• "CTCM1" — Description coming soon...

• "CTCM1_each" — Description coming soon...

• "RI" — Description coming soon...

• "RI_each" — Description coming soon...

CTCM1_polarity

Character. Polarity of the CTCM1 wording factor(s). Defaults to "negative" for negative polarity variables

...

Additional arguments to be passed on to cfa

Value

Returns a list containing:

model

Estimated ESEM model

fit

Fit measures of estimated ESEM model

Author(s)

Alexander P. Christensen <[e-mail]>, Luis Eduardo Garrido <[e-mail]>

Examples

# Generate factor data
two_factor <- simulate_factors(
  factors = 2, # factors = 2
  variables = 6, # variables per factor = 6
  loadings = 0.55, # loadings between = 0.45 to 0.65
  cross_loadings = 0.05, # cross-loadings N(0, 0.05)
  correlations = 0.30, # correlation between factors = 0.30
  sample_size = 1000, # number of cases = 1000
  variable_categories = 5 # 5-point Likert scale
)

## Not run: 
# Estimate ESEM model with no wording effects
esem_no_wording_effects <- ESEM(
  data = two_factor,
  estimator = "WLSMV"
)

# Add wording effects using acquiescence method
two_factor_acquiescence <- add_wording_effects(
  lf_object = two_factor,
  proportion_negative = 0.50,
  proportion_biased_cases = 0.10,
  method = "acquiescence"
)

# Estimate ESEM model with wording effects
esem_wording_effects <- ESEM(
  data = two_factor_acquiescence,
  estimator = "WLSMV",
  wording_factor = "RI"
)
## End(Not run)