library(testthat) library(lavaan) mod <- ' m1 ~ iv1 + a2 * iv2 dv ~ b * m1 a1b := a2 * b ' dat <- pa_dat dat0 <- dat[1:50, ] fit <- lavaan::sem(mod, dat0) rerun_out <- lavaan_rerun(fit, parallel = FALSE, to_rerun = 1:10) fit_est_change <- est_change(rerun_out) fit_est_change_raw <- est_change_raw(rerun_out) inf_out <- influence_stat(rerun_out) p <- index_plot(fit_est_change, "gcd") test_that("index_plot", { expect_equal(p$data$x, fit_est_change[, "gcd"], ignore_attr = TRUE) expect_equal(p$layers[[4]]$data[1, "row_id"], 9) }) p <- index_plot(fit_est_change_raw, "m1~iv2") test_that("index_plot", { expect_equal(p$data$x, fit_est_change_raw[, "m1~iv2"], ignore_attr = TRUE) expect_equal(p$layers[[4]]$data[1, "row_id"], 8) }) p <- index_plot(fit_est_change_raw, "m1~iv2", largest_x = 5) test_that("index_plot", { expect_equal(p$data$x, fit_est_change_raw[, "m1~iv2"], ignore_attr = TRUE) expect_equal(nrow(p$layers[[4]]$data), 5) expect_equal(p$labels$y, "Statistic") }) p <- index_plot(inf_out, "chisq", x_label = "Chi-Square Influence", largest_x = 3, cutoff_x_high = .08, cutoff_x_low = -.25) test_that("index_plot", { expect_equal(p$data$x, inf_out[, "chisq"], ignore_attr = TRUE) expect_equal(p$layers[[4]]$data[1, 1], -.25) expect_equal(p$layers[[5]]$data[1, 1], .08) expect_equal(nrow(p$layers[[6]]$data), 6) expect_equal(p$labels$y, "Chi-Square Influence") }) p <- index_plot(inf_out, "chisq", absolute = TRUE, largest_x = 3, cutoff_x_high = .08) test_that("index_plot", { expect_equal(p$data$x, abs(inf_out[, "chisq"]), ignore_attr = TRUE) expect_equal(p$layers[[4]]$data[1, 1], .08) expect_equal(nrow(p$layers[[5]]$data), 6) expect_equal(p$labels$y, "Absolute(Statistic)") })