LPA Classification-Mplus Atomation

Tutorial:https://github.com/immerse-ucsb/lca_enum Package:https://cran.r-project.org/web/packages/MplusAutomation/index.html ————————————————————————

#DATA ##Load packages

##Variable LPA RESS = ress_rum, ress_eng, ress_reap, ress_sup, ress_rel, ress_dis average frequency of use of each strategy (average of 4 items per strategy)

##Prepare Data

# Load your dataset
getwd()

[1] "C:/Users/Lenovo/Documents/data"

#Descriptive Analysis

Descriptive Statistics for Continuous Variables
Variable	N	Mean	SD	Min	Max	Median
ress_rum	244	4.34	1.51	1	7.00	4.50
ress_eng	244	2.73	1.18	1	6.25	2.50
ress_reap	244	3.75	1.33	1	7.00	3.75
ress_sup	244	4.12	1.64	1	7.00	4.00
ress_rel	244	3.36	1.54	1	7.00	3.25
ress_dis	244	4.18	1.63	1	7.00	4.25

descriptives_table <- purrr::map_dfr(
  c("ress_rum", "ress_eng", "ress_reap", "ress_sup", "ress_rel", "ress_dis"),
  ~ df_ress %>%
    summarise(
      variable = .x,
      n    = sum(!is.na(.data[[.x]])),
      mean = mean(.data[[.x]], na.rm = TRUE),
      sd   = sd(.data[[.x]], na.rm = TRUE),
      min  = min(.data[[.x]], na.rm = TRUE),
      max  = max(.data[[.x]], na.rm = TRUE),
      med  = median(.data[[.x]], na.rm = TRUE)
    )
) %>%
  dplyr::mutate(
    dplyr::across(
      c(mean, sd, min, max, med),
      ~ round(.x, 2)
    )
  ) %>%
  gt::gt(rowname_col = "variable") %>%
  gt::tab_stubhead(label = md("*Variable*")) %>%
  gt::tab_header(
    md("Descriptive Statistics for Continuous Variables")
  ) %>%
  gt::cols_label(
    n    = md("*N*"),
    mean = md("*Mean*"),
    sd   = md("*SD*"),
    min  = md("*Min*"),
    max  = md("*Max*"),
    med  = md("*Median*")
  )

# Display table
descriptives_table


# Save table as image
gtsave(descriptives_table, here::here("figures", "descriptives_table.png"))

#NUMERATION

IMPORTANT: Before moving forward, make sure to open each output document to ensure models were estimated normally. Review: files> data>enum los output (.out) revisar que no salga error (warning esta bien), que el n muestral es correcto y que diga “THE MODEL ESTIMATION TERMINATED NORMALLY”

Model Fit Summary Table1
Classes	Par	LL	BIC	aBIC	CAIC	AWE	BLRT	VLMR	BF	cmPk
1-Class	12	−2,630.01	5,325.99	5,287.95	5,337.99	5,427.96	–	–	0.00	<.001
2-Class	19	−2,545.99	5,196.43	5,136.20	5,215.43	5,357.88	<.001	<.001	0.00	<.001
3-Class	26	−2,515.47	5,173.86	5,091.44	5,199.86	5,394.78	<.001	0.14	0.00	<.001
4-Class	33	−2,485.63	5,152.66	5,048.05	5,185.66	5,433.07	<.001	0.24	0.14	0.12
5-Class	40	−2,464.42	5,148.73	5,021.93	5,188.73	5,488.61	<.001	0.10	>100	0.88
6-Class	47	−2,451.95	5,162.26	5,013.28	5,209.26	5,561.63	0.02	0.65	–	0.00
1 Note. Par = Parameters; LL = model log likelihood; BIC = Bayesian information criterion; aBIC = sample size adjusted BIC; CAIC = consistent Akaike information criterion; AWE = approximate weight of evidence criterion; BLRT = bootstrapped likelihood ratio test p-value; VLMR = Vuong-Lo-Mendell-Rubin adjusted likelihood ratio test p-value; cmPk = approximate correct model probability.

##Information Criteria Plot

#LPA

p3 <- plot_lpa(output_ress$c3_ress.out)
p3
ggsave(filename = here::here("figures", "plot_3ps.png"),plot = p3, dpi = 300,
  height = 6, width = 9, units = "in")


p4 <- plot_lpa(output_ress$c4_ress.out)
p4

ggsave(filename = here::here("figures", "plot_4ps.png"),plot = p4, dpi = 300,
  height = 6, width = 9, units = "in")

p5 <- plot_lpa(output_ress$c5_ress.out)
p5
ggsave(filename = here::here("figures", "plot_5ps.png"),plot = p5, dpi = 300,
  height = 6, width = 9, units = "in")

p6 <- plot_lpa(output_ress$c6_ress.out)
p6
ggsave(filename = here::here("figures", "plot_6ps.png"),plot = p6, dpi = 300,
  height = 6, width = 9, units = "in")