A Function That Writes, Saves, and Exports Syntax for Fitting Latent Dyadic Confirmatory Factor Analysis (CFA) Models with Multiple Factors

This function takes the outputted object from scrapeVarCross() when the var_list argument has been used, and automatically writes, returns, and exports (.txt) lavaan() syntax for specifying dyadic configural, loading, and intercept invariant measurement models for either a group of latent variables (e.g., different sub-scales from a self-report measures).

Usage

scriptCFA(
  dvn,
  scaleset = "FF",
  constr_dy_meas = c("loadings", "intercepts", "residuals"),
  constr_dy_struct = c("variances", "means"),
  writeTo = NULL,
  fileName = NULL
)

Arguments

dvn: input dvn list from scrapeVarCross(); scrapeVarCross must have been run with the var_list argument
scaleset: Input character to specify how to set the scale of the latent variable(s). Default is "FF" (fixed-factor; see Details for rationale), but user can specify "MV" (Marker Variable).
constr_dy_meas: input character vector detailing which measurement model parameters to constrain across dyad members. Default is c("loadings", "intercepts", "residuals")(in combination with defaults for constr_dy_struct, an indistinguishable dyadic CFA), but user can specify any combination of "loadings", "intercepts", and "residuals", or "none" to specify an otherwise unconstrained dyadic configural invariance model
constr_dy_struct: input character vector detailing which structural model parameters to constrain across dyad members. Default is c("variances", "means")(in combination with defaults for constr_dy_meas, an indistinguishable dyadic CFA), but user can specify any combination of "variances" and "means", or "none".
writeTo: A character string specifying a directory path to where the output file(s) should be saved. If set to ".", the file(s) will be written to the current working directory. The default is NULL, and examples use a temporary directory created by tempdir(). When dealing with tabular output, writeTo is only relevant if gtTab = TRUE.
fileName: A character string specifying a desired base name for the output file. The default is NULL. The specified name will be automatically appended with the appropriate file extension (e.g., .txt for lavaan scripts, .rtf for tabular output when gtTab = TRUE, or other extensions as appropriate for the output type). If a file with the same name already exists in the user's chosen directory, it will be overwritten.

Value

character object of lavaan script that can be passed immediately to lavaan functions

Details

By default, many dySEM:: functions (including scriptCFA() default to a fixed-factor method of scale-setting, whereby the latent variance of a given factor is constrained to 1 for both partners in the configurally invariant #model, and then one of these variances is freely estimated in subsequent #models of the invariance testing sequence. We have selected this default for two reasons: (1) the selection of a marker-variable is usually arbitrary, yet can have a large influence on the estimation and testing of of structural parameters (see https://stats.stackexchange.com/questions/402133/in-cfa-does-it-matter-which-factor-loading-is-set-to-1/402732#402732); and (2) the selection of a non-invariant marker-variable can have disastrous down-stream consequences for the identification of non-invariant measurement parameters, following a the rejection of an omnibus #invariance constraint set (see Lee, Preacher, & Little, 2011).

Examples


# When different factor use distinct stems:
imsList <- list(
  lvnames = c("Sat", "Q_Alt", "Invest", "Comm"),
  stem = c("sat.g", "qalt.g", "invest.g", "com"),
  delim1 = c("", "", "", ""),
  delim2 = c("_", "_", "_", "_")
)

dvnIMS <- scrapeVarCross(imsM,
  var_list = imsList,
  var_list_order = "sip",
  distinguish_1 = "f",
  distinguish_2 = "m"
)
#> 
#> ── Variable Scraping Summary ──
#> 
#> ✔ Successfully scraped 4 latent variables
#> ℹ Sat: 5 indicators for P1 (f), 5 indicators for P2 (m)
#> ℹ Q_Alt: 5 indicators for P1 (f), 5 indicators for P2 (m)
#> ℹ Invest: 5 indicators for P1 (f), 5 indicators for P2 (m)
#> ℹ Comm: 5 indicators for P1 (f), 5 indicators for P2 (m)
#> ℹ Total indicators: 40

script.ims.config <- scriptCFA(dvnIMS,
  scaleset = "FF",
  constr_dy_meas = "none",
  constr_dy_struct = "none", writeTo = tempdir(), fileName = "ims_config"
)

script.ims.load <- scriptCFA(dvnIMS,
  scaleset = "FF",
  constr_dy_meas = c("loadings"),
  constr_dy_struct = "none", writeTo = tempdir(), fileName = "ims_load"
)

script.ims.int <- scriptCFA(dvnIMS,
  scaleset = "FF",
  constr_dy_meas = c("loadings", "intercepts"),
  constr_dy_struct = "none", writeTo = tempdir(), fileName = "ims_int"
)

script.ims.res <- scriptCFA(dvnIMS,
  scaleset = "FF",
  constr_dy_meas = c("loadings", "intercepts", "residuals"),
  constr_dy_struct = "none", writeTo = tempdir(), fileName = "ims_res"
)

script.ims.indist <- scriptCFA(dvnIMS,
  scaleset = "FF",
  constr_dy_meas = c("loadings", "intercepts", "residuals"),
  constr_dy_struct = c("variances", "means"), writeTo = tempdir(), fileName = "ims_indist"
)

# When different factor use the same stem and muse be detected through numeric position

prqcList <- list(
  lvnames = c("Sat", "Comm", "Intim", "Trust", "Pass", "Love"),
  stem = c("prqc", "prqc", "prqc", "prqc", "prqc", "prqc"),
  delim1 = c(".", ".", ".", ".", ".", "."),
  delim2 = c("_", "_", "_", "_", "_", "_"),
  min_num = c(1, 4, 7, 10, 13, 16),
  max_num = c(3, 6, 9, 12, 15, 18)
)

dvnPRQC <- scrapeVarCross(prqcQ,
  var_list = prqcList,
  var_list_order = "spi",
  distinguish_1 = "1",
  distinguish_2 = "2"
)
#> 
#> ── Variable Scraping Summary ──
#> 
#> ✔ Successfully scraped 6 latent variables
#> ℹ Sat: 3 indicators for P1 (1), 3 indicators for P2 (2)
#> ℹ Comm: 3 indicators for P1 (1), 3 indicators for P2 (2)
#> ℹ Intim: 3 indicators for P1 (1), 3 indicators for P2 (2)
#> ℹ Trust: 3 indicators for P1 (1), 3 indicators for P2 (2)
#> ℹ Pass: 3 indicators for P1 (1), 3 indicators for P2 (2)
#> ℹ Love: 3 indicators for P1 (1), 3 indicators for P2 (2)
#> ℹ Total indicators: 36

script.prqc.config <- scriptCFA(dvnPRQC,
  scaleset = "FF",
  constr_dy_meas = "none",
  constr_dy_struct = "none", writeTo = tempdir(), fileName = "prqc_config"
)

script.prqc.load <- scriptCFA(dvnPRQC,
  scaleset = "FF",
  constr_dy_meas = c("loadings"),
  constr_dy_struct = "none", writeTo = tempdir(), fileName = "prqc_load"
)

script.prqc.int <- scriptCFA(dvnPRQC,
  scaleset = "FF",
  constr_dy_meas = c("loadings", "intercepts"),
  constr_dy_struct = "none", writeTo = tempdir(), fileName = "prqc_int"
)

script.prqc.res <- scriptCFA(dvnPRQC,
  scaleset = "FF",
  constr_dy_meas = c("loadings", "intercepts", "residuals"),
  constr_dy_struct = "none", writeTo = tempdir(), fileName = "prqc_res"
)

script.prqc.indist <- scriptCFA(dvnPRQC,
  scaleset = "FF",
  constr_dy_meas = c("loadings", "intercepts", "residuals"),
  constr_dy_struct = c("variances", "means"), writeTo = tempdir(), fileName = "prqc_indist"
)