Multigroup Analysis for Piecewise SEM
multigroup(
modelList,
group,
standardize = "scale",
standardize.type = "latent.linear",
test.type = "III"
)a list of structural equations
the name of the grouping variable in quotes
The type of standardization: none, scale, range.
Default is scale.
The type of standardized for non-Gaussian responses:
latent.linear, Menard.OE. Default is latent.linear.
what kind of ANOVA should be reported. Default is type III
data(meadows)
jutila <- psem(
lm(rich ~ elev + mass, data = meadows),
lm(mass ~ elev, data = meadows)
)
jutila.multigroup <- multigroup(jutila, group = "grazed")
jutila.multigroup
#>
#> Structural Equation Model of jutila
#>
#> Groups = grazed [ 1, 0 ]
#>
#> ---
#>
#> Global goodness-of-fit:
#>
#> Fisher's C = NA with P-value = NA and on 0 degrees of freedom
#>
#> ---
#>
#> Model-wide Interactions:
#>
#> Response Predictor Test.Stat DF P.Value
#> rich elev:grazed 1556.7 1 0.3358
#> rich grazed:mass 1556.7 1 0.0026 **
#> mass elev:grazed 1416938.0 1 0.0055 **
#>
#> elev -> rich constrained to the global model
#>
#> ---
#>
#> Group [1] coefficients:
#>
#> Response Predictor Estimate Std.Error DF Crit.Value P.Value Std.Estimate
#> rich elev 0.0731 0.0081 351 8.9882 0 0.4967 ***
#> rich mass -0.0007 0.0017 162 -0.4198 0.6752 -0.0291
#> mass elev -1.2028 0.4728 163 -2.5438 0.0119 -0.1954 *
#>
#> c
#>
#>
#>
#> Group [0] coefficients:
#>
#> Response Predictor Estimate Std.Error DF Crit.Value P.Value Std.Estimate
#> rich elev 0.0731 0.0081 351 8.9882 0 0.3933 ***
#> rich mass -0.0072 0.0013 186 -5.4216 0 -0.3222 ***
#> mass elev -3.2735 0.5571 187 -5.8764 0 -0.3948 ***
#>
#> c
#>
#>
#>
#> Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 c = constrained