This vignette shows how to estimate interaction models, with both continuous and ordered (categorical) data.
fit_cont <- pls(
m,
data = modsem::oneInt,
bootstrap = TRUE,
boot.R = 50
)
summary(fit_cont)
#> plssem (0.1.2) ended normally after 3 iterations
#> Estimator PLSc
#> Link LINEAR
#>
#> Number of observations 2000
#> Number of iterations 3
#> Number of latent variables 3
#> Number of observed variables 9
#>
#> Fit Measures:
#> Chi-Square 56.757
#> Degrees of Freedom 21
#> SRMR 0.006
#> RMSEA 0.029
#>
#> R-squared (indicators):
#> x1 0.863
#> x2 0.819
#> x3 0.809
#> z1 0.830
#> z2 0.827
#> z3 0.843
#> y1 0.934
#> y2 0.919
#> y3 0.923
#>
#> R-squared (latents):
#> Y 0.604
#>
#> Latent Variables:
#> Estimate Std.Error z.value P(>|z|)
#> X =~
#> x1 0.929 0.012 75.834 0.000
#> x2 0.905 0.015 61.806 0.000
#> x3 0.899 0.013 70.530 0.000
#> Z =~
#> z1 0.911 0.011 79.973 0.000
#> z2 0.909 0.015 61.911 0.000
#> z3 0.918 0.012 77.681 0.000
#> Y =~
#> y1 0.966 0.006 174.278 0.000
#> y2 0.959 0.008 116.493 0.000
#> y3 0.961 0.006 147.962 0.000
#>
#> Regressions:
#> Estimate Std.Error z.value P(>|z|)
#> Y ~
#> X 0.423 0.020 21.233 0.000
#> Z 0.361 0.017 21.096 0.000
#> X:Z 0.452 0.017 27.255 0.000
#>
#> Covariances:
#> Estimate Std.Error z.value P(>|z|)
#> X ~~
#> Z 0.201 0.023 8.869 0.000
#> X:Z 0.018 0.040 0.455 0.649
#> Z ~~
#> X:Z 0.060 0.048 1.267 0.205
#>
#> Variances:
#> Estimate Std.Error z.value P(>|z|)
#> X 1.000 0.022 45.244 0.000
#> Z 1.000 0.033 30.396 0.000
#> .Y 0.396 0.017 23.234 0.000
#> X:Z 1.013 0.061 16.594 0.000
#> .x1 0.137 0.023 6.034 0.000
#> .x2 0.181 0.026 6.834 0.000
#> .x3 0.191 0.023 8.333 0.000
#> .z1 0.170 0.021 8.181 0.000
#> .z2 0.173 0.027 6.493 0.000
#> .z3 0.157 0.022 7.226 0.000
#> .y1 0.066 0.011 6.167 0.000
#> .y2 0.081 0.016 5.157 0.000
#> .y3 0.077 0.012 6.201 0.000
fit_ord <- pls(
m,
data = oneIntOrdered,
bootstrap = TRUE,
boot.R = 50,
ordered = colnames(oneIntOrdered) # explicitly specify variables as ordered
)
summary(fit_ord)
#> plssem (0.1.2) ended normally after 13 iterations
#> Estimator MCOrdPLSc
#> Link PROBIT
#>
#> Number of observations 2000
#> Number of iterations 13
#> Number of latent variables 3
#> Number of observed variables 9
#>
#> Fit Measures:
#> Chi-Square 20.221
#> Degrees of Freedom 21
#> SRMR 0.012
#> RMSEA 0.000
#>
#> R-squared (indicators):
#> x1 0.929
#> x2 0.900
#> x3 0.906
#> z1 0.936
#> z2 0.903
#> z3 0.912
#> y1 0.970
#> y2 0.953
#> y3 0.962
#>
#> R-squared (latents):
#> Y 0.554
#>
#> Latent Variables:
#> Estimate Std.Error z.value P(>|z|)
#> X =~
#> x1 0.929 0.007 136.264 0.000
#> x2 0.900 0.009 99.157 0.000
#> x3 0.906 0.008 119.169 0.000
#> Z =~
#> z1 0.936 0.009 106.614 0.000
#> z2 0.903 0.009 98.216 0.000
#> z3 0.912 0.008 120.827 0.000
#> Y =~
#> y1 0.970 0.004 225.334 0.000
#> y2 0.953 0.005 177.485 0.000
#> y3 0.962 0.004 217.936 0.000
#>
#> Regressions:
#> Estimate Std.Error z.value P(>|z|)
#> Y ~
#> X 0.419 0.020 21.466 0.000
#> Z 0.355 0.017 21.328 0.000
#> X:Z 0.445 0.023 19.293 0.000
#>
#> Covariances:
#> Estimate Std.Error z.value P(>|z|)
#> X ~~
#> Z 0.193 0.027 7.025 0.000
#> X:Z -0.006 0.015 -0.410 0.682
#> Z ~~
#> X:Z 0.006 0.018 0.321 0.748
#>
#> Variances:
#> Estimate Std.Error z.value P(>|z|)
#> X 1.000
#> Z 1.000
#> .Y 0.446 0.033 13.489 0.000
#> X:Z 1.000
#> .x1 0.071 0.007 10.463 0.000
#> .x2 0.100 0.009 10.957 0.000
#> .x3 0.094 0.008 12.417 0.000
#> .z1 0.064 0.009 7.300 0.000
#> .z2 0.097 0.009 10.601 0.000
#> .z3 0.088 0.008 11.683 0.000
#> .y1 0.030 0.004 6.855 0.000
#> .y2 0.047 0.005 8.838 0.000
#> .y3 0.038 0.004 8.602 0.000