Analisis Pengukuran Berulang Kadar Asam Urat: 3 Kelompok x 4 Waktu
Lusiana
2026-10-03
Nama: Lusiana
NIM: 2611018017
Mata Kuliah: Biostatistika Intermediete
Program Studi: Magister Kesehatan Masyarakat
Universitas Mulawarman
Metode
Data mencakup tiga kelompok dan empat pengukuran kadar asam urat per peserta. Analisis meliputi repeated-measures ANOVA satu arah pada kelompok DASH+AF, mixed ANOVA kelompok x waktu, koreksi Greenhouse-Geisser, pendekatan multivariat, post hoc, kontras interaksi, alternatif nonparametrik, dan linear mixed model (LMM). Baseline, minggu ke-4, minggu ke-8 dan minggu ke-12 berjarak sama. Sumber: data_asam_urat_wide.xlsx, sheet Data Asam Urat. Satuan pengukuran dan status asal data tidak dicantumkan dalam workbook.
analisis_asam_urat <- function() {
paket <- c("dplyr", "tidyr", "ggplot2", "afex", "emmeans", "rstatix",
"car", "effectsize", "lme4", "lmerTest", "pbkrtest",
"performance", "ggpubr", "WRS2", "htmltools", "base64enc")
belum <- paket[!vapply(paket, requireNamespace, logical(1), quietly = TRUE)]
if (length(belum)) install.packages(belum, repos = "https://cloud.r-project.org")
gagal <- paket[!vapply(paket, requireNamespace, logical(1), quietly = TRUE)]
if (length(gagal)) stop("Paket belum tersedia: ", paste(gagal, collapse = ", "))
suppressPackageStartupMessages({
library(dplyr); library(tidyr); library(ggplot2)
})
opsi_lama <- options(contrasts = c("contr.sum", "contr.poly"), width = 120)
on.exit(options(opsi_lama), add = TRUE)
afex::afex_options(emmeans_model = "multivariate")
theme_set(theme_bw(base_size = 12))
set.seed(2026)
folder <- file.path(getwd(), "Hasil_Analisis_Asam_Urat")
dir.create(folder, recursive = TRUE, showWarnings = FALSE)
hasil <- list(); isi <- list(); grafik <- list()
teks <- function(judul, x) {
cat("\n", judul, "\n", sep = "")
cat(x, "\n")
isi[[length(isi) + 1L]] <<- htmltools::tagList(
htmltools::tags$h2(judul), htmltools::tags$p(x))
}
tampil <- function(nama, x) {
hasil[[nama]] <<- x
cat("\n", gsub("_", " ", nama), "\n", sep = "")
print(x)
cetak <- capture.output(print(x))
writeLines(cetak, file.path(folder, paste0(nama, ".txt")))
if (is.data.frame(x)) write.csv(x, file.path(folder, paste0(nama, ".csv")), row.names = FALSE)
isi[[length(isi) + 1L]] <<- htmltools::tagList(
htmltools::tags$h2(gsub("_", " ", nama)),
htmltools::tags$pre(paste(cetak, collapse = "\n")))
invisible(x)
}
gambar <- function(nama, p, lebar = 11, tinggi = 6) {
print(p)
berkas <- file.path(folder, paste0(nama, ".png"))
ggsave(berkas, plot = p, width = lebar, height = tinggi, dpi = 300, bg = "white")
grafik[[nama]] <<- p
isi[[length(isi) + 1L]] <<- htmltools::tagList(
htmltools::tags$h2(gsub("_", " ", nama)),
htmltools::tags$img(src = base64enc::dataURI(file = berkas, mime = "image/png"),
style = "max-width:100%;height:auto;"))
invisible(p)
}
opsional <- function(nama, ekspresi) {
tryCatch(tampil(nama, ekspresi), error = function(e) {
teks(nama, paste("Analisis tidak dapat diestimasi:", conditionMessage(e)))
invisible(NULL)
})
}
fp <- function(p) ifelse(is.na(p), "NA", ifelse(p < .001, "< 0,001",
paste0("= ", formatC(p, digits = 3, format = "f", decimal.mark = ","))))
csv_asli <- 'id,kelompok,usia,jk,AU_M0,AU_M4,AU_M8,AU_M12
P001,Kontrol,42,L,6.7,7.2,6.4,6.5
P002,Kontrol,43,L,7.0,6.3,7.0,6.8
P003,Kontrol,62,P,5.1,5.2,4.7,6.3
P004,Kontrol,40,P,6.6,6.4,6.0,7.0
P005,Kontrol,53,P,6.1,5.8,6.6,6.2
P006,Kontrol,42,L,7.1,6.6,6.9,6.5
P007,Kontrol,60,P,7.7,7.4,8.5,7.0
P008,Kontrol,63,L,6.5,6.7,6.2,5.9
P009,Kontrol,53,P,6.1,5.9,5.5,6.4
P010,Kontrol,62,P,7.3,7.4,7.2,6.2
P011,Kontrol,57,P,5.5,5.2,5.4,5.5
P012,Kontrol,52,P,6.9,6.9,5.5,6.9
P013,Kontrol,63,L,8.9,9.1,9.2,8.6
P014,Kontrol,50,P,4.6,4.5,3.6,5.8
P015,Kontrol,57,P,6.7,7.5,6.4,6.4
P016,Kontrol,39,P,6.6,7.1,6.8,6.7
P017,Kontrol,54,P,7.1,7.3,7.1,6.4
P018,Kontrol,38,P,6.7,6.3,7.0,7.2
P019,Kontrol,42,P,7.1,6.9,7.0,7.2
P020,Kontrol,41,L,8.7,8.7,8.9,9.2
P021,Kontrol,57,L,5.0,5.0,5.4,5.6
P022,Kontrol,62,P,8.5,8.1,8.3,9.4
P023,Kontrol,45,P,7.0,6.8,6.4,7.1
P024,Kontrol,51,L,6.4,6.2,6.6,6.2
P025,Kontrol,65,P,7.4,7.4,7.1,7.0
P026,Kontrol,56,P,7.3,7.9,7.4,6.8
P027,Kontrol,54,P,6.6,6.6,6.6,6.6
P028,Kontrol,42,P,5.5,5.8,5.3,5.1
P029,Kontrol,41,L,8.7,9.0,8.5,8.3
P030,Kontrol,50,P,6.0,6.2,5.7,5.6
P031,DASH,36,P,6.5,6.8,5.7,5.6
P032,DASH,61,L,7.4,7.7,7.1,6.4
P033,DASH,64,L,5.4,5.7,4.6,4.8
P034,DASH,36,L,5.9,5.1,5.4,5.3
P035,DASH,35,L,6.6,6.4,6.2,5.4
P036,DASH,60,L,6.8,5.9,6.3,6.1
P037,DASH,54,L,7.6,6.9,6.4,7.0
P038,DASH,44,P,7.8,7.5,7.4,7.3
P039,DASH,47,P,6.1,6.1,6.4,5.3
P040,DASH,48,L,6.7,6.1,6.1,5.6
P041,DASH,39,L,7.0,6.9,6.7,6.2
P042,DASH,39,P,5.3,5.2,4.6,4.6
P043,DASH,54,P,6.9,6.3,6.4,5.8
P044,DASH,39,P,6.4,5.9,6.3,5.3
P045,DASH,45,L,9.2,8.7,8.7,8.9
P046,DASH,57,L,8.8,8.9,8.1,7.5
P047,DASH,63,L,7.4,6.5,6.2,6.6
P048,DASH,53,P,5.5,5.6,4.9,4.7
P049,DASH,39,L,8.8,8.8,8.2,8.0
P050,DASH,45,P,5.9,5.9,5.7,4.8
P051,DASH,62,P,7.3,6.3,7.4,6.6
P052,DASH,45,P,7.1,6.9,6.7,6.6
P053,DASH,56,L,7.1,6.6,7.4,7.0
P054,DASH,59,L,7.5,7.3,6.2,6.4
P055,DASH,39,L,8.1,8.0,7.6,7.2
P056,DASH,63,L,6.2,6.3,5.9,6.1
P057,DASH,48,L,7.4,6.5,6.1,6.3
P058,DASH,62,L,6.5,6.3,6.3,5.1
P059,DASH,40,P,7.9,7.1,7.2,7.6
P060,DASH,59,L,5.9,6.2,5.4,5.2
P061,DASH+AF,60,P,8.2,8.8,7.2,7.5
P062,DASH+AF,44,L,8.1,7.7,7.7,8.2
P063,DASH+AF,48,P,4.4,4.0,3.2,3.6
P064,DASH+AF,39,L,7.7,7.7,6.9,6.6
P065,DASH+AF,50,P,7.9,7.7,7.7,7.0
P066,DASH+AF,59,P,5.3,4.7,4.4,4.3
P067,DASH+AF,56,L,7.6,7.2,5.8,5.7
P068,DASH+AF,57,L,8.6,7.9,8.7,7.6
P069,DASH+AF,45,L,7.7,7.4,7.2,6.6
P070,DASH+AF,42,P,5.4,5.4,4.2,3.8
P071,DASH+AF,47,L,6.5,5.9,6.1,5.8
P072,DASH+AF,53,L,6.0,5.6,5.4,5.6
P073,DASH+AF,44,P,6.7,6.8,6.1,5.8
P074,DASH+AF,47,P,5.9,5.3,4.8,4.4
P075,DASH+AF,46,P,7.4,7.2,7.4,6.6
P076,DASH+AF,36,P,6.2,6.0,5.8,5.6
P077,DASH+AF,39,L,6.9,6.7,5.9,6.0
P078,DASH+AF,63,P,5.9,5.5,5.1,5.0
P079,DASH+AF,59,L,7.8,7.2,7.0,6.1
P080,DASH+AF,43,P,7.1,7.7,6.0,5.9
P081,DASH+AF,57,L,9.1,8.8,8.6,8.1
P082,DASH+AF,44,P,6.1,5.4,5.3,4.6
P083,DASH+AF,41,P,6.7,6.1,5.2,5.9
P084,DASH+AF,41,L,6.8,5.8,6.4,5.3
P085,DASH+AF,64,L,6.9,5.9,5.6,6.3
P086,DASH+AF,44,P,6.2,5.9,5.8,4.9
P087,DASH+AF,47,L,6.5,6.3,6.0,6.6
P088,DASH+AF,64,P,6.5,5.6,5.4,4.7
P089,DASH+AF,46,P,6.7,6.5,6.0,6.0
P090,DASH+AF,48,L,6.1,5.3,5.6,5.0'
dat_wide <- read.csv(text = csv_asli, stringsAsFactors = FALSE)
kolom <- c("id", "kelompok", "usia", "jk", "AU_M0", "AU_M4", "AU_M8", "AU_M12")
stopifnot(all(kolom %in% names(dat_wide)), !anyDuplicated(dat_wide$id))
minggu <- c(0, 4, 8, 12)
kolom_au <- c("AU_M0", "AU_M4", "AU_M8", "AU_M12")
kelompok <- c("Kontrol", "DASH", "DASH+AF")
stopifnot(setequal(unique(dat_wide$kelompok), kelompok),
all(vapply(dat_wide[kolom_au], is.numeric, logical(1))),
all(is.finite(as.matrix(dat_wide[kolom_au]))))
dat_wide <- dat_wide |> mutate(id = factor(id), kelompok = factor(kelompok, levels = c("Kontrol", "DASH", "DASH+AF")))
dat_long <- dat_wide |>
pivot_longer(all_of(kolom_au), names_to = "waktu", values_to = "asam_urat") |>
mutate(waktu = factor(waktu, levels = kolom_au, labels = c("M0", "M4", "M8", "M12")),
minggu = minggu[as.integer(waktu)]) |> arrange(id, waktu)
teks("Identitas", paste("Nama: Lusiana | NIM: 2611018017 |",
"Mata Kuliah: Biostatistika Intermediete | Magister Kesehatan Masyarakat Universitas Mulawarman"))
teks("Desain", paste(nrow(dat_wide), "peserta; tiga kelompok; empat pengukuran berulang.",
"Tanpa nilai hilang. Satuan kadar asam urat tidak dicantumkan dalam workbook."))
tampil("01_Jumlah_Peserta", dat_wide |> count(kelompok, name = "n"))
tampil("01a_Usia", dat_wide |> group_by(kelompok) |> summarise(n = n(), mean = mean(usia), sd = sd(usia), min = min(usia), max = max(usia), .groups = "drop"))
tampil("01b_Jenis_Kelamin", dat_wide |> count(kelompok, jk))
teks("Sumber data", "Analisis dihitung dari sheet Data Asam Urat. Sheet Ringkasan hanya berisi statistik agregat. Satuan pengukuran dan status data asli/simulasi tidak dicantumkan; interpretasi menggunakan skala numerik dalam workbook.")
tampil("02_Data_Wide", head(dat_wide))
tampil("03_Data_Long", head(dat_long, 12))
desk <- dat_long |> group_by(kelompok, waktu, minggu) |>
summarise(n = n(), mean = mean(asam_urat), sd = sd(asam_urat), median = median(asam_urat),
min = min(asam_urat), max = max(asam_urat), se = sd / sqrt(n),
lower = mean - qt(.975, n - 1) * se,
upper = mean + qt(.975, n - 1) * se, .groups = "drop")
tampil("04_Deskriptif", desk)
tampil("05_Kovarians", cov(dat_wide[kolom_au]))
tampil("06_Korelasi", cor(dat_wide[kolom_au]))
pasangan <- combn(kolom_au, 2)
tampil("07_Varians_Selisih", data.frame(
pasangan = apply(pasangan, 2, paste, collapse = " - "),
varians = apply(pasangan, 2, function(p) var(dat_wide[[p[1]]] - dat_wide[[p[2]]]))))
gambar("G01_Profil_Rerata_CI95", ggplot(desk, aes(minggu, mean, color = kelompok, group = kelompok)) +
geom_line(linewidth = 1) + geom_point(size = 3) +
geom_errorbar(aes(ymin = lower, ymax = upper), width = .5) +
scale_x_continuous(breaks = minggu) +
labs(title = "Profil rerata kadar asam urat dan interval kepercayaan 95%", x = "Minggu", y = "Kadar asam urat", color = "Kelompok") +
theme(legend.position = "bottom"))
gambar("G02_Lintasan_Individu", ggplot(dat_long, aes(minggu, asam_urat, group = id)) +
geom_line(alpha = .25) + stat_summary(aes(group = kelompok), fun = mean,
geom = "line", color = "firebrick", linewidth = 1.2) +
facet_wrap(~ kelompok) + scale_x_continuous(breaks = minggu) +
labs(title = "Lintasan individu dan rerata kelompok", x = "Minggu", y = "Kadar asam urat"))
gambar("G03_Boxplot", ggplot(dat_long, aes(waktu, asam_urat, fill = kelompok)) +
geom_boxplot() + labs(title = "Distribusi kadar asam urat berdasarkan kelompok dan waktu", x = "Waktu", y = "Kadar asam urat", fill = "Kelompok") +
theme(legend.position = "bottom"))
d1 <- droplevels(filter(dat_long, kelompok == "DASH+AF"))
tampil("08_Outlier_Satu_Kelompok", d1 |> group_by(waktu) |> rstatix::identify_outliers(asam_urat))
tampil("09_Shapiro_Satu_Kelompok", d1 |> group_by(waktu) |> rstatix::shapiro_test(asam_urat))
gambar("G04_QQ_Satu_Kelompok", ggpubr::ggqqplot(d1, "asam_urat", facet.by = "waktu"))
aov1_rs <- rstatix::anova_test(data = d1, dv = asam_urat, wid = id, within = waktu, effect.size = "pes")
tampil("10_ANOVA_Rstatix_Satu_Kelompok", aov1_rs)
aov1 <- afex::aov_ez(id = "id", dv = "asam_urat", data = d1, within = "waktu",
anova_table = list(es = c("ges", "pes"), correction = "GG"))
tampil("11_RM_ANOVA", aov1)
tampil("12_Sferisitas_dan_Koreksi_RM", summary(aov1))
tampil("13_Multivariat_RM", aov1$Anova)
tampil("14_Eta_Kuadrat_RM", effectsize::eta_squared(aov1, partial = TRUE))
em1 <- emmeans::emmeans(aov1, ~ waktu)
tampil("15_EMM_RM", as.data.frame(em1))
tampil("16_Posthoc_RM_Bonferroni", as.data.frame(pairs(em1, adjust = "bonferroni")))
tampil("17_RM_vs_Baseline_Holm", as.data.frame(emmeans::contrast(em1, "trt.vs.ctrl", ref = 1, adjust = "holm")))
tampil("18_Tren_Polinomial_RM", as.data.frame(emmeans::contrast(em1, "poly")))
tampil("19_Friedman", rstatix::friedman_test(d1, asam_urat ~ waktu | id))
tampil("20_Kendall_W", rstatix::friedman_effsize(d1, asam_urat ~ waktu | id))
tampil("21_Wilcoxon_Berpasangan", d1 |> arrange(waktu, id) |>
rstatix::wilcox_test(asam_urat ~ waktu, paired = TRUE, p.adjust.method = "bonferroni"))
opsional("22_Robust_RM_Trim20", WRS2::rmanova(d1$asam_urat, d1$waktu, d1$id, tr = .2))
tampil("23_Outlier_Mixed", dat_long |> group_by(kelompok, waktu) |> rstatix::identify_outliers(asam_urat))
tampil("24_Shapiro_Mixed", dat_long |> group_by(kelompok, waktu) |> rstatix::shapiro_test(asam_urat))
gambar("G05_QQ_Mixed", ggpubr::ggqqplot(dat_long, "asam_urat", ggtheme = theme_bw()) +
facet_grid(waktu ~ kelompok), tinggi = 9)
tampil("25_Levene", dat_long |> group_by(waktu) |> rstatix::levene_test(asam_urat ~ kelompok))
tampil("26_Box_M", rstatix::box_m(dat_wide[kolom_au], dat_wide$kelompok))
aov2 <- afex::aov_ez(id = "id", dv = "asam_urat", data = dat_long, between = "kelompok", within = "waktu",
anova_table = list(es = c("ges", "pes"), correction = "GG"))
tampil("27_Mixed_ANOVA", aov2)
tampil("28_Sferisitas_dan_Koreksi_Mixed", summary(aov2))
tampil("29_Multivariat_Mixed", aov2$Anova)
tampil("30_Eta_Kuadrat_Mixed", effectsize::eta_squared(aov2, partial = TRUE))
gambar("G06_Profil_Mixed_ANOVA", afex::afex_plot(aov2, x = "waktu", trace = "kelompok",
error = "within", mapping = c("colour", "shape", "linetype")) +
labs(title = "Profil mixed ANOVA", x = "Waktu", y = "Kadar asam urat"))
teks("Interval grafik mixed ANOVA", "Interval within-subject pada grafik afex berbeda dari interval kepercayaan rerata biasa pada grafik profil deskriptif.")
tampil("31_Efek_Waktu_per_Kelompok", as.data.frame(emmeans::joint_tests(aov2, by = "kelompok")))
tampil("32_Efek_Kelompok_per_Waktu", as.data.frame(emmeans::joint_tests(aov2, by = "waktu")))
em2 <- emmeans::emmeans(aov2, ~ waktu | kelompok)
em2b <- emmeans::emmeans(aov2, ~ kelompok | waktu)
tampil("33_EMM_Waktu_per_Kelompok", as.data.frame(em2))
tampil("34_Waktu_vs_Baseline_Holm", as.data.frame(emmeans::contrast(em2, "trt.vs.ctrl", ref = 1, adjust = "holm")))
tampil("35_Antarkelompok_Tukey", as.data.frame(pairs(em2b, adjust = "tukey")))
em_full <- emmeans::emmeans(aov2, ~ waktu * kelompok)
ki <- emmeans::contrast(em_full, interaction = list(
waktu = list("M12-M0" = c(-1, 0, 0, 1)), kelompok = "pairwise"), adjust = "holm")
tampil("36_Kontras_Perbedaan_Perubahan", as.data.frame(ki))
tren <- as.data.frame(emmeans::contrast(em_full,
interaction = c(waktu = "poly", kelompok = "pairwise"), adjust = "none"))
tren_lin <- subset(tren, waktu_poly == "linear")
tren_lin$p.holm <- p.adjust(tren_lin$p.value, "holm")
tampil("37_Kontras_Tren_Linear", tren_lin)
teks("Koreksi perbandingan", "Holm untuk waktu versus baseline berlaku dalam tiap kelompok; Tukey berlaku dalam tiap waktu. Kontras perbedaan perubahan dan tren linear memakai Holm pada masing-masing keluarga kontras.")
tabel_gg <- as.data.frame(afex::nice(aov2, correction = "GG", es = "pes"))
tampil("38_Tabel_ANOVA_GG", tabel_gg)
tab_num <- as.data.frame(aov2$anova_table)
if ("kelompok:waktu" %in% rownames(tab_num)) {
r <- tab_num["kelompok:waktu", , drop = FALSE]
teks("Interpretasi interaksi", paste0("Mixed ANOVA dengan koreksi Greenhouse-Geisser menghasilkan interaksi kelompok x waktu: F(",
round(r[["num Df"]], 2), ", ", round(r[["den Df"]], 2), ") = ", round(r[["F"]], 3),
"; p ", fp(r[["Pr(>F)"]]), ". ",
if (r[["Pr(>F)"]] < .05) "Pola perubahan kadar asam urat berbeda secara statistik antar kelompok."
else "Belum ditemukan bukti statistik perbedaan pola perubahan kadar asam urat antar kelompok.",
" Arah dan besarnya perbedaan dinilai dari rerata serta kontras perubahan M12-M0."))
}
kontrol <- lme4::lmerControl(optimizer = "bobyqa", optCtrl = list(maxfun = 200000))
lmm1 <- lmerTest::lmer(asam_urat ~ kelompok * waktu + (1 | id), data = dat_long, REML = TRUE, control = kontrol)
lmm2 <- lmerTest::lmer(asam_urat ~ kelompok * waktu + (1 + minggu | id), data = dat_long, REML = TRUE, control = kontrol)
tampil("39_LMM_Intersep", summary(lmm1))
tampil("40_LMM_Intersep_Slope", summary(lmm2))
tampil("41_Perbandingan_LMM_REML", anova(lmm1, lmm2, refit = FALSE))
teks("Perbandingan model", "Kedua model mempunyai fixed effects yang sama. Perbandingan REML mengikuti materi; nilai p likelihood-ratio bersifat pendekatan karena varians acak diuji pada batas ruang parameter.")
tampil("42_Singularitas_LMM", data.frame(model = c("Intersep", "Intersep dan slope"),
singular = c(lme4::isSingular(lmm1), lme4::isSingular(lmm2))))
opsional("43_LMM_Kenward_Roger", anova(lmm2, ddf = "Kenward-Roger"))
opsional("44_ICC", performance::icc(lmm1))
rd <- data.frame(prediksi = fitted(lmm2), residual = resid(lmm2))
gambar("G07_QQ_Residual_LMM", ggplot(rd, aes(sample = residual)) + stat_qq() + stat_qq_line() +
labs(title = "Q-Q residual LMM", x = "Kuantil teoretis", y = "Kuantil residual"))
ra <- data.frame(intersep = lme4::ranef(lmm2)$id[, 1])
gambar("G08_QQ_Intersep_Acak", ggplot(ra, aes(sample = intersep)) + stat_qq() + stat_qq_line() +
labs(title = "Q-Q intersep acak LMM", x = "Kuantil teoretis", y = "Kuantil intersep acak"))
gambar("G09_Residual_vs_Prediksi", ggplot(rd, aes(prediksi, residual)) + geom_point(alpha = .5) +
geom_hline(yintercept = 0, linetype = 2) + labs(title = "Residual versus nilai prediksi LMM", x = "Prediksi kadar asam urat", y = "Residual"))
em_lmm <- emmeans::emmeans(lmm2, ~ waktu | kelompok, lmer.df = "kenward-roger")
tampil("45_EMM_LMM", as.data.frame(em_lmm))
set.seed(1)
dat_miss <- dat_long
indeks_hilang <- sample(which(dat_miss$waktu != "M0"), 30)
dat_miss$asam_urat[indeks_hilang] <- NA_real_
teks("Demonstrasi data hilang", "Sebanyak 30 pengukuran pascabaseline dihapus secara acak hanya pada salinan demonstrasi. Analisis utama tetap menggunakan seluruh data yang dilampirkan; demonstrasi ini mengikuti bagian LMM data hilang dalam materi.")
lmm_miss <- lmerTest::lmer(asam_urat ~ kelompok * waktu + (1 + minggu | id), data = dat_miss,
REML = TRUE, control = kontrol, na.action = na.omit)
opsional("46_LMM_Data_Hilang_KR", anova(lmm_miss, ddf = "Kenward-Roger"))
tampil("47_Ringkasan_Data_Hilang", data.frame(pengukuran_hilang = sum(is.na(dat_miss$asam_urat)),
peserta_terdampak = n_distinct(dat_miss$id[is.na(dat_miss$asam_urat)]),
peserta_lengkap = n_distinct(dat_miss$id) - n_distinct(dat_miss$id[is.na(dat_miss$asam_urat)]),
pengukuran_dianalisis_LMM = nobs(lmm_miss)))
tampil("48_Informasi_Sesi", sessionInfo())
write.csv(dat_wide, file.path(folder, "Data_Asam_Urat_Wide.csv"), row.names = FALSE)
write.csv(dat_long, file.path(folder, "Data_Asam_Urat_Long.csv"), row.names = FALSE)
write.csv(dat_miss, file.path(folder, "Data_Demonstrasi_Missing.csv"), row.names = FALSE)
saveRDS(list(hasil = hasil, data_wide = dat_wide, data_long = dat_long,
aov1 = aov1, aov2 = aov2, lmm1 = lmm1, lmm2 = lmm2, lmm_miss = lmm_miss),
file.path(folder, "Objek_Analisis_Asam_Urat.rds"))
laporan <- htmltools::tags$html(lang = "id",
htmltools::tags$head(htmltools::tags$meta(charset = "UTF-8"),
htmltools::tags$title("Analisis Kadar Asam Urat: 3 x 4"),
htmltools::tags$style("body{font-family:Arial,sans-serif;max-width:1100px;margin:40px auto;padding:20px;line-height:1.6;color:#17342e}h1,h2{color:#0b6b4f}h2{border-bottom:1px solid #d9e4e0;padding-top:24px}pre{background:#f3f7f5;padding:16px;overflow:auto;font-size:13px}img{display:block;margin:15px auto}")),
htmltools::tags$body(htmltools::tags$h1("Analisis Pengukuran Berulang Kadar Asam Urat: 3 x 4"), isi))
htmltools::save_html(laporan, file.path(folder, "Laporan_Analisis_Asam_Urat.html"))
cat("\nHasil analisis: ", normalizePath(folder), "\n", sep = "")
invisible(list(hasil = hasil, grafik = grafik, data_wide = dat_wide,
data_long = dat_long, aov1 = aov1, aov2 = aov2,
lmm1 = lmm1, lmm2 = lmm2, lmm_miss = lmm_miss))
}
hasil_asam_urat <- analisis_asam_urat()## Registered S3 method overwritten by 'lme4':
## method from
## na.action.merMod car##
## Identitas
## Nama: Lusiana | NIM: 2611018017 | Mata Kuliah: Biostatistika Intermediete | Magister Kesehatan Masyarakat Universitas Mulawarman
##
## Desain
## 90 peserta; tiga kelompok; empat pengukuran berulang. Tanpa nilai hilang. Satuan kadar asam urat tidak dicantumkan dalam workbook.
##
## 01 Jumlah Peserta
## kelompok n
## 1 Kontrol 30
## 2 DASH 30
## 3 DASH+AF 30
##
## 01a Usia
## # A tibble: 3 × 6
## kelompok n mean sd min max
## <fct> <int> <dbl> <dbl> <int> <int>
## 1 Kontrol 30 51.2 8.60 38 65
## 2 DASH 30 49.7 9.79 35 64
## 3 DASH+AF 30 49.1 8.07 36 64
##
## 01b Jenis Kelamin
## kelompok jk n
## 1 Kontrol L 9
## 2 Kontrol P 21
## 3 DASH L 19
## 4 DASH P 11
## 5 DASH+AF L 14
## 6 DASH+AF P 16
##
## Sumber data
## Analisis dihitung dari sheet Data Asam Urat. Sheet Ringkasan hanya berisi statistik agregat. Satuan pengukuran dan status data asli/simulasi tidak dicantumkan; interpretasi menggunakan skala numerik dalam workbook.
##
## 02 Data Wide
## id kelompok usia jk AU_M0 AU_M4 AU_M8 AU_M12
## 1 P001 Kontrol 42 L 6.7 7.2 6.4 6.5
## 2 P002 Kontrol 43 L 7.0 6.3 7.0 6.8
## 3 P003 Kontrol 62 P 5.1 5.2 4.7 6.3
## 4 P004 Kontrol 40 P 6.6 6.4 6.0 7.0
## 5 P005 Kontrol 53 P 6.1 5.8 6.6 6.2
## 6 P006 Kontrol 42 L 7.1 6.6 6.9 6.5
##
## 03 Data Long
## # A tibble: 12 × 7
## id kelompok usia jk waktu asam_urat minggu
## <fct> <fct> <int> <chr> <fct> <dbl> <dbl>
## 1 P001 Kontrol 42 L M0 6.7 0
## 2 P001 Kontrol 42 L M4 7.2 4
## 3 P001 Kontrol 42 L M8 6.4 8
## 4 P001 Kontrol 42 L M12 6.5 12
## 5 P002 Kontrol 43 L M0 7 0
## 6 P002 Kontrol 43 L M4 6.3 4
## 7 P002 Kontrol 43 L M8 7 8
## 8 P002 Kontrol 43 L M12 6.8 12
## 9 P003 Kontrol 62 P M0 5.1 0
## 10 P003 Kontrol 62 P M4 5.2 4
## 11 P003 Kontrol 62 P M8 4.7 8
## 12 P003 Kontrol 62 P M12 6.3 12
##
## 04 Deskriptif
## # A tibble: 12 × 12
## kelompok waktu minggu n mean sd median min max se lower upper
## <fct> <fct> <dbl> <int> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 Kontrol M0 0 30 6.78 1.07 6.7 4.6 8.9 0.194 6.38 7.18
## 2 Kontrol M4 4 30 6.78 1.12 6.75 4.5 9.1 0.205 6.36 7.20
## 3 Kontrol M8 8 30 6.64 1.25 6.6 3.6 9.2 0.228 6.17 7.11
## 4 Kontrol M12 12 30 6.75 1.01 6.55 5.1 9.4 0.185 6.37 7.13
## 5 DASH M0 0 30 6.97 1.00 6.95 5.3 9.2 0.183 6.59 7.34
## 6 DASH M4 4 30 6.68 0.982 6.45 5.1 8.9 0.179 6.31 7.05
## 7 DASH M8 8 30 6.45 1.01 6.3 4.6 8.7 0.184 6.08 6.83
## 8 DASH M12 12 30 6.18 1.07 6.15 4.6 8.9 0.195 5.78 6.58
## 9 DASH+AF M0 0 30 6.83 1.04 6.7 4.4 9.1 0.190 6.44 7.22
## 10 DASH+AF M4 4 30 6.47 1.18 6.2 4 8.8 0.215 6.03 6.91
## 11 DASH+AF M8 8 30 6.08 1.24 5.95 3.2 8.7 0.227 5.62 6.55
## 12 DASH+AF M12 12 30 5.84 1.17 5.85 3.6 8.2 0.214 5.40 6.27
##
## 05 Kovarians
## AU_M0 AU_M4 AU_M8 AU_M12
## AU_M0 1.0548077 1.041194 1.104733 0.9831161
## AU_M4 1.0411935 1.193703 1.131793 1.0251386
## AU_M8 1.1047328 1.131793 1.394433 1.1479476
## AU_M12 0.9831161 1.025139 1.147948 1.2955506
##
## 06 Korelasi
## AU_M0 AU_M4 AU_M8 AU_M12
## AU_M0 1.0000000 0.9278904 0.9109022 0.8409903
## AU_M4 0.9278904 1.0000000 0.8772430 0.8243417
## AU_M8 0.9109022 0.8772430 1.0000000 0.8540750
## AU_M12 0.8409903 0.8243417 0.8540750 1.0000000
##
## 07 Varians Selisih
## pasangan varians
## 1 AU_M0 - AU_M4 0.1661236
## 2 AU_M0 - AU_M8 0.2397753
## 3 AU_M0 - AU_M12 0.3841261
## 4 AU_M4 - AU_M8 0.3245506
## 5 AU_M4 - AU_M12 0.4389763
## 6 AU_M8 - AU_M12 0.3940886
##
## 08 Outlier Satu Kelompok
## [1] waktu id kelompok usia jk asam_urat minggu is.outlier is.extreme
## <0 rows> (or 0-length row.names)
##
## 09 Shapiro Satu Kelompok
## # A tibble: 4 × 4
## waktu variable statistic p
## <fct> <chr> <dbl> <dbl>
## 1 M0 asam_urat 0.986 0.956
## 2 M4 asam_urat 0.965 0.424
## 3 M8 asam_urat 0.973 0.612
## 4 M12 asam_urat 0.979 0.808
##
## 10 ANOVA Rstatix Satu Kelompok
## ANOVA Table (type III tests)
##
## $ANOVA
## Effect DFn DFd F p p<.05 pes
## 1 waktu 3 87 45.869 7.15e-18 * 0.613
##
## $`Mauchly's Test for Sphericity`
## Effect W p p<.05
## 1 waktu 0.694 0.072
##
## $`Sphericity Corrections`
## Effect GGe DF[GG] p[GG] p[GG]<.05 HFe DF[HF] p[HF] p[HF]<.05
## 1 waktu 0.845 2.54, 73.54 1.84e-15 * 0.933 2.8, 81.18 7.88e-17 *
##
##
## 11 RM ANOVA
## Anova Table (Type 3 tests)
##
## Response: asam_urat
## Effect df MSE F ges pes p.value
## 1 waktu 2.54, 73.54 0.15 45.87 *** .099 .613 <.001
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
##
## Sphericity correction method: GG
##
## 12 Sferisitas dan Koreksi RM
##
## Univariate Type III Repeated-Measures ANOVA Assuming Sphericity
##
## Sum Sq num Df Error SS den Df F value Pr(>F)
## (Intercept) 4769.1 1 144.965 29 954.048 < 2.2e-16 ***
## waktu 17.1 3 10.816 87 45.869 < 2.2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
##
## Mauchly Tests for Sphericity
##
## Test statistic p-value
## waktu 0.69402 0.071917
##
##
## Greenhouse-Geisser and Huynh-Feldt Corrections
## for Departure from Sphericity
##
## GG eps Pr(>F[GG])
## waktu 0.84532 1.839e-15 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## HF eps Pr(>F[HF])
## waktu 0.9330749 7.880193e-17
##
## 13 Multivariat RM
##
## Type III Repeated Measures MANOVA Tests: Pillai test statistic
## Df test stat approx F num Df den Df Pr(>F)
## (Intercept) 1 0.97050 954.05 1 29 < 2.2e-16 ***
## waktu 1 0.85808 54.42 3 27 1.416e-11 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## 14 Eta Kuadrat RM
## # Effect Size for ANOVA (Type III)
##
## Parameter | Eta2 (partial) | 95% CI
## -----------------------------------------
## waktu | 0.61 | [0.50, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].
## 15 EMM RM
## waktu emmean SE df lower.CL upper.CL
## M0 6.830000 0.1897457 29 6.441926 7.218074
## M4 6.466667 0.2147082 29 6.027539 6.905794
## M8 6.083333 0.2265221 29 5.620044 6.546623
## M12 5.836667 0.2136429 29 5.399718 6.273615
##
## Confidence level used: 0.95
##
## 16 Posthoc RM Bonferroni
## contrast estimate SE df t.ratio p.value
## M0 - M4 0.3633333 0.07038210 29 5.162 <0.0001
## M0 - M8 0.7466667 0.07874494 29 9.482 <0.0001
## M0 - M12 0.9933333 0.08800906 29 11.287 <0.0001
## M4 - M8 0.3833333 0.10732734 29 3.572 0.0076
## M4 - M12 0.6300000 0.10138693 29 6.214 <0.0001
## M8 - M12 0.2466667 0.09501764 29 2.596 0.0879
##
## P value adjustment: bonferroni method for 6 tests
##
## 17 RM vs Baseline Holm
## contrast estimate SE df t.ratio p.value
## M4 - M0 -0.3633333 0.07038210 29 -5.162 <0.0001
## M8 - M0 -0.7466667 0.07874494 29 -9.482 <0.0001
## M12 - M0 -0.9933333 0.08800906 29 -11.287 <0.0001
##
## P value adjustment: holm method for 3 tests
##
## 18 Tren Polinomial RM
## contrast estimate SE df t.ratio p.value
## linear -3.363333 0.2978653 29 -11.291 <0.0001
## quadratic 0.116667 0.1058174 29 1.103 0.2793
## cubic 0.156667 0.3223721 29 0.486 0.6306
##
##
## 19 Friedman
## # A tibble: 1 × 6
## .y. n statistic df p method
## * <chr> <int> <dbl> <dbl> <dbl> <chr>
## 1 asam_urat 30 55.6 3 5.20e-12 Friedman test
##
## 20 Kendall W
## # A tibble: 1 × 5
## .y. n effsize method magnitude
## * <chr> <int> <dbl> <chr> <ord>
## 1 asam_urat 30 0.617 Kendall W large
##
## 21 Wilcoxon Berpasangan
## # A tibble: 6 × 9
## .y. group1 group2 n1 n2 statistic p p.adj p.adj.signif
## * <chr> <chr> <chr> <int> <int> <dbl> <dbl> <dbl> <chr>
## 1 asam_urat M0 M4 30 30 413 0.0000558 0.000335 ***
## 2 asam_urat M0 M8 30 30 462 0.00000000745 0.0000000447 ****
## 3 asam_urat M0 M12 30 30 462 0.00000000931 0.0000000559 ****
## 4 asam_urat M4 M8 30 30 390. 0.000587 0.00352 **
## 5 asam_urat M4 M12 30 30 442. 0.00000114 0.00000686 ****
## 6 asam_urat M8 M12 30 30 349 0.0147 0.0880 ns
##
## 22 Robust RM Trim20
## Call:
## WRS2::rmanova(y = d1$asam_urat, groups = d1$waktu, blocks = d1$id,
## tr = 0.2)
##
## Test statistic: F = 23.6353
## Degrees of freedom 1: 2.97
## Degrees of freedom 2: 50.55
## p-value: 0
##
##
## 23 Outlier Mixed
## # A tibble: 10 × 9
## kelompok waktu id usia jk asam_urat minggu is.outlier is.extreme
## <fct> <fct> <fct> <int> <chr> <dbl> <dbl> <lgl> <lgl>
## 1 Kontrol M0 P013 63 L 8.9 0 TRUE FALSE
## 2 Kontrol M8 P013 63 L 9.2 8 TRUE FALSE
## 3 Kontrol M8 P014 50 P 3.6 8 TRUE FALSE
## 4 Kontrol M12 P013 63 L 8.6 12 TRUE FALSE
## 5 Kontrol M12 P020 41 L 9.2 12 TRUE FALSE
## 6 Kontrol M12 P022 62 P 9.4 12 TRUE TRUE
## 7 Kontrol M12 P029 41 L 8.3 12 TRUE FALSE
## 8 DASH M4 P045 45 L 8.7 4 TRUE FALSE
## 9 DASH M4 P046 57 L 8.9 4 TRUE FALSE
## 10 DASH M4 P049 39 L 8.8 4 TRUE FALSE
##
## 24 Shapiro Mixed
## # A tibble: 12 × 5
## kelompok waktu variable statistic p
## <fct> <fct> <chr> <dbl> <dbl>
## 1 Kontrol M0 asam_urat 0.960 0.309
## 2 Kontrol M4 asam_urat 0.980 0.825
## 3 Kontrol M8 asam_urat 0.969 0.524
## 4 Kontrol M12 asam_urat 0.896 0.00680
## 5 DASH M0 asam_urat 0.973 0.624
## 6 DASH M4 asam_urat 0.929 0.0459
## 7 DASH M8 asam_urat 0.971 0.555
## 8 DASH M12 asam_urat 0.962 0.355
## 9 DASH+AF M0 asam_urat 0.986 0.956
## 10 DASH+AF M4 asam_urat 0.965 0.424
## 11 DASH+AF M8 asam_urat 0.973 0.612
## 12 DASH+AF M12 asam_urat 0.979 0.808
##
## 25 Levene
## # A tibble: 4 × 5
## waktu df1 df2 statistic p
## <fct> <int> <int> <dbl> <dbl>
## 1 M0 2 87 0.00953 0.991
## 2 M4 2 87 0.959 0.387
## 3 M8 2 87 0.505 0.605
## 4 M12 2 87 0.634 0.533
##
## 26 Box M
## # A tibble: 1 × 4
## statistic p.value parameter method
## <dbl> <dbl> <dbl> <chr>
## 1 27.2 0.129 20 Box's M-test for Homogeneity of Covariance Matrices
##
## 27 Mixed ANOVA
## Anova Table (Type 3 tests)
##
## Response: asam_urat
## Effect df MSE F ges pes p.value
## 1 kelompok 2, 87 4.42 1.29 .026 .029 .280
## 2 waktu 2.42, 210.37 0.17 48.14 *** .044 .356 <.001
## 3 kelompok:waktu 4.84, 210.37 0.17 10.13 *** .019 .189 <.001
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '+' 0.1 ' ' 1
##
## Sphericity correction method: GG
##
## 28 Sferisitas dan Koreksi Mixed
##
## Univariate Type III Repeated-Measures ANOVA Assuming Sphericity
##
## Sum Sq num Df Error SS den Df F value Pr(>F)
## (Intercept) 15382.1 1 384.78 87 3477.9614 < 2.2e-16 ***
## kelompok 11.4 2 384.78 87 1.2903 0.2804
## waktu 19.5 3 35.15 261 48.1420 < 2.2e-16 ***
## kelompok:waktu 8.2 6 35.15 261 10.1306 4.522e-10 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
##
## Mauchly Tests for Sphericity
##
## Test statistic p-value
## waktu 0.57481 4.5869e-09
## kelompok:waktu 0.57481 4.5869e-09
##
##
## Greenhouse-Geisser and Huynh-Feldt Corrections
## for Departure from Sphericity
##
## GG eps Pr(>F[GG])
## waktu 0.80601 < 2.2e-16 ***
## kelompok:waktu 0.80601 1.62e-08 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## HF eps Pr(>F[HF])
## waktu 0.8306986 5.013127e-21
## kelompok:waktu 0.8306986 1.026115e-08
##
## 29 Multivariat Mixed
##
## Type III Repeated Measures MANOVA Tests: Pillai test statistic
## Df test stat approx F num Df den Df Pr(>F)
## (Intercept) 1 0.97560 3478.0 1 87 < 2.2e-16 ***
## kelompok 2 0.02881 1.3 2 87 0.2804
## waktu 1 0.76354 91.5 3 85 < 2.2e-16 ***
## kelompok:waktu 2 0.57739 11.6 6 172 6.387e-11 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## 30 Eta Kuadrat Mixed
## # Effect Size for ANOVA (Type III)
##
## Parameter | Eta2 (partial) | 95% CI
## ----------------------------------------------
## kelompok | 0.03 | [0.00, 1.00]
## waktu | 0.36 | [0.28, 1.00]
## kelompok:waktu | 0.19 | [0.11, 1.00]
##
## - One-sided CIs: upper bound fixed at [1.00].## Warning: Panel(s) show a mixed within-between-design.
## Error bars do not allow comparisons across all means.
## Suppress error bars with: error = "none"
##
## Interval grafik mixed ANOVA
## Interval within-subject pada grafik afex berbeda dari interval kepercayaan rerata biasa pada grafik profil deskriptif.
##
## 31 Efek Waktu per Kelompok
## kelompok = Kontrol:
## model term df1 df2 F.ratio p.value
## waktu 3 87 1.110 0.3494
##
## kelompok = DASH:
## model term df1 df2 F.ratio p.value
## waktu 3 87 47.894 <0.0001
##
## kelompok = DASH+AF:
## model term df1 df2 F.ratio p.value
## waktu 3 87 82.977 <0.0001
##
##
## 32 Efek Kelompok per Waktu
## waktu = M0:
## model term df1 df2 F.ratio p.value
## kelompok 2 87 0.261 0.7708
##
## waktu = M4:
## model term df1 df2 F.ratio p.value
## kelompok 2 87 0.639 0.5305
##
## waktu = M8:
## model term df1 df2 F.ratio p.value
## kelompok 2 87 1.756 0.1787
##
## waktu = M12:
## model term df1 df2 F.ratio p.value
## kelompok 2 87 5.378 0.0063
##
##
## 33 EMM Waktu per Kelompok
## kelompok = Kontrol:
## waktu emmean SE df lower.CL upper.CL
## M0 6.780000 0.1890870 87 6.404169 7.155831
## M4 6.780000 0.2002897 87 6.381903 7.178097
## M8 6.640000 0.2137857 87 6.215078 7.064922
## M12 6.746667 0.1982858 87 6.352552 7.140781
##
## kelompok = DASH:
## waktu emmean SE df lower.CL upper.CL
## M0 6.966667 0.1890870 87 6.590836 7.342498
## M4 6.680000 0.2002897 87 6.281903 7.078097
## M8 6.453333 0.2137857 87 6.028411 6.878256
## M12 6.176667 0.1982858 87 5.782552 6.570781
##
## kelompok = DASH+AF:
## waktu emmean SE df lower.CL upper.CL
## M0 6.830000 0.1890870 87 6.454169 7.205831
## M4 6.466667 0.2002897 87 6.068569 6.864764
## M8 6.083333 0.2137857 87 5.658411 6.508256
## M12 5.836667 0.1982858 87 5.442552 6.230781
##
## Confidence level used: 0.95
##
## 34 Waktu vs Baseline Holm
## kelompok = Kontrol:
## contrast estimate SE df t.ratio p.value
## M4 - M0 0.0000000 0.06943762 87 0.000 1.0000
## M8 - M0 -0.1400000 0.07761121 87 -1.804 0.2241
## M12 - M0 -0.0333333 0.08494382 87 -0.392 1.0000
##
## kelompok = DASH:
## contrast estimate SE df t.ratio p.value
## M4 - M0 -0.2866667 0.06943762 87 -4.128 <0.0001
## M8 - M0 -0.5133333 0.07761121 87 -6.614 <0.0001
## M12 - M0 -0.7900000 0.08494382 87 -9.300 <0.0001
##
## kelompok = DASH+AF:
## contrast estimate SE df t.ratio p.value
## M4 - M0 -0.3633333 0.06943762 87 -5.233 <0.0001
## M8 - M0 -0.7466667 0.07761121 87 -9.621 <0.0001
## M12 - M0 -0.9933333 0.08494382 87 -11.694 <0.0001
##
## P value adjustment: holm method for 3 tests
##
## 35 Antarkelompok Tukey
## waktu = M0:
## contrast estimate SE df t.ratio p.value
## Kontrol - DASH -0.1866667 0.2674094 87 -0.698 0.7653
## Kontrol - (DASH+AF) -0.0500000 0.2674094 87 -0.187 0.9809
## DASH - (DASH+AF) 0.1366667 0.2674094 87 0.511 0.8662
##
## waktu = M4:
## contrast estimate SE df t.ratio p.value
## Kontrol - DASH 0.1000000 0.2832524 87 0.353 0.9337
## Kontrol - (DASH+AF) 0.3133333 0.2832524 87 1.106 0.5129
## DASH - (DASH+AF) 0.2133333 0.2832524 87 0.753 0.7325
##
## waktu = M8:
## contrast estimate SE df t.ratio p.value
## Kontrol - DASH 0.1866667 0.3023387 87 0.617 0.8110
## Kontrol - (DASH+AF) 0.5566667 0.3023387 87 1.841 0.1624
## DASH - (DASH+AF) 0.3700000 0.3023387 87 1.224 0.4424
##
## waktu = M12:
## contrast estimate SE df t.ratio p.value
## Kontrol - DASH 0.5700000 0.2804184 87 2.033 0.1105
## Kontrol - (DASH+AF) 0.9100000 0.2804184 87 3.245 0.0047
## DASH - (DASH+AF) 0.3400000 0.2804184 87 1.212 0.4491
##
## P value adjustment: tukey method for comparing a family of 3 estimates
##
## 36 Kontras Perbedaan Perubahan
## waktu_custom kelompok_pairwise estimate SE df t.ratio p.value
## M12-M0 Kontrol - DASH 0.7566667 0.1201287 87 6.299 <0.0001
## M12-M0 Kontrol - (DASH+AF) 0.9600000 0.1201287 87 7.991 <0.0001
## M12-M0 DASH - (DASH+AF) 0.2033333 0.1201287 87 1.693 0.0941
##
## P value adjustment: holm method for 3 tests
##
## 37 Kontras Tren Linear
## waktu_poly kelompok_pairwise estimate SE df t.ratio p.value p.holm
## 1 linear Kontrol - DASH 2.3566667 0.4004321 87 5.885309 7.260002e-08 1.452000e-07
## 4 linear Kontrol - (DASH+AF) 3.1233333 0.4004321 87 7.799908 1.259894e-11 3.779683e-11
## 7 linear DASH - (DASH+AF) 0.7666667 0.4004321 87 1.914599 5.883019e-02 5.883019e-02
##
## Koreksi perbandingan
## Holm untuk waktu versus baseline berlaku dalam tiap kelompok; Tukey berlaku dalam tiap waktu. Kontras perbedaan perubahan dan tren linear memakai Holm pada masing-masing keluarga kontras.
##
## 38 Tabel ANOVA GG
## Effect df MSE F pes p.value
## 1 kelompok 2, 87 4.42 1.29 .029 .280
## 2 waktu 2.42, 210.37 0.17 48.14 *** .356 <.001
## 3 kelompok:waktu 4.84, 210.37 0.17 10.13 *** .189 <.001
##
## Interpretasi interaksi
## Mixed ANOVA dengan koreksi Greenhouse-Geisser menghasilkan interaksi kelompok x waktu: F(4.84, 210.37) = 10.131; p < 0,001. Pola perubahan kadar asam urat berbeda secara statistik antar kelompok. Arah dan besarnya perbedaan dinilai dari rerata serta kontras perubahan M12-M0.## boundary (singular) fit: see help('isSingular')##
## 39 LMM Intersep
## Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
## Formula: asam_urat ~ kelompok * waktu + (1 | id)
## Data: dat_long
## Control: kontrol
##
## REML criterion at convergence: 651.6
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.70490 -0.54912 -0.01823 0.52179 2.99938
##
## Random effects:
## Groups Name Variance Std.Dev.
## id (Intercept) 1.0720 1.035
## Residual 0.1347 0.367
## Number of obs: 360, groups: id, 90
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 6.536667 0.110839 86.999996 58.974 < 2e-16 ***
## kelompok1 0.200000 0.156751 86.999996 1.276 0.20538
## kelompok2 0.032500 0.156751 86.999996 0.207 0.83623
## waktu1 0.322222 0.033500 261.000001 9.619 < 2e-16 ***
## waktu2 0.105556 0.033500 261.000001 3.151 0.00182 **
## waktu3 -0.144444 0.033500 261.000001 -4.312 2.30e-05 ***
## kelompok1:waktu1 -0.278889 0.047376 261.000001 -5.887 1.21e-08 ***
## kelompok2:waktu1 0.075278 0.047376 261.000001 1.589 0.11329
## kelompok1:waktu2 -0.062222 0.047376 261.000001 -1.313 0.19021
## kelompok2:waktu2 0.005278 0.047376 261.000001 0.111 0.91138
## kelompok1:waktu3 0.047778 0.047376 261.000001 1.008 0.31416
## kelompok2:waktu3 0.028611 0.047376 261.000001 0.604 0.54643
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) klmpk1 klmpk2 waktu1 waktu2 waktu3 klm1:1 klm2:1 klm1:2 klm2:2 klm1:3
## kelompok1 0.000
## kelompok2 0.000 -0.500
## waktu1 0.000 0.000 0.000
## waktu2 0.000 0.000 0.000 -0.333
## waktu3 0.000 0.000 0.000 -0.333 -0.333
## klmpk1:wkt1 0.000 0.000 0.000 0.000 0.000 0.000
## klmpk2:wkt1 0.000 0.000 0.000 0.000 0.000 0.000 -0.500
## klmpk1:wkt2 0.000 0.000 0.000 0.000 0.000 0.000 -0.333 0.167
## klmpk2:wkt2 0.000 0.000 0.000 0.000 0.000 0.000 0.167 -0.333 -0.500
## klmpk1:wkt3 0.000 0.000 0.000 0.000 0.000 0.000 -0.333 0.167 -0.333 0.167
## klmpk2:wkt3 0.000 0.000 0.000 0.000 0.000 0.000 0.167 -0.333 0.167 -0.333 -0.500
##
## 40 LMM Intersep Slope
## Linear mixed model fit by REML. t-tests use Satterthwaite's method ['lmerModLmerTest']
## Formula: asam_urat ~ kelompok * waktu + (1 + minggu | id)
## Data: dat_long
## Control: kontrol
##
## REML criterion at convergence: 651.4
##
## Scaled residuals:
## Min 1Q Median 3Q Max
## -2.69257 -0.55505 -0.03363 0.51123 3.05558
##
## Random effects:
## Groups Name Variance Std.Dev. Corr
## id (Intercept) 1.048e+00 1.023765
## minggu 3.757e-06 0.001938 1.00
## Residual 1.346e-01 0.366839
## Number of obs: 360, groups: id, 90
##
## Fixed effects:
## Estimate Std. Error df t value Pr(>|t|)
## (Intercept) 6.536667 0.110839 86.999934 58.974 < 2e-16 ***
## kelompok1 0.200000 0.156751 86.999935 1.276 0.20538
## kelompok2 0.032500 0.156751 86.999935 0.207 0.83623
## waktu1 0.322222 0.033510 260.429159 9.616 < 2e-16 ***
## waktu2 0.105556 0.033490 261.064188 3.152 0.00181 **
## waktu3 -0.144444 0.033490 261.064188 -4.313 2.29e-05 ***
## kelompok1:waktu1 -0.278889 0.047390 260.429159 -5.885 1.22e-08 ***
## kelompok2:waktu1 0.075278 0.047390 260.429159 1.588 0.11340
## kelompok1:waktu2 -0.062222 0.047362 261.064188 -1.314 0.19008
## kelompok2:waktu2 0.005278 0.047362 261.064188 0.111 0.91136
## kelompok1:waktu3 0.047778 0.047362 261.064188 1.009 0.31402
## kelompok2:waktu3 0.028611 0.047362 261.064188 0.604 0.54631
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Correlation of Fixed Effects:
## (Intr) klmpk1 klmpk2 waktu1 waktu2 waktu3 klm1:1 klm2:1 klm1:2 klm2:2 klm1:3
## kelompok1 0.000
## kelompok2 0.000 -0.500
## waktu1 -0.036 0.000 0.000
## waktu2 -0.012 0.000 0.000 -0.333
## waktu3 0.012 0.000 0.000 -0.334 -0.333
## klmpk1:wkt1 0.000 -0.036 0.018 0.000 0.000 0.000
## klmpk2:wkt1 0.000 0.018 -0.036 0.000 0.000 0.000 -0.500
## klmpk1:wkt2 0.000 -0.012 0.006 0.000 0.000 0.000 -0.333 0.166
## klmpk2:wkt2 0.000 0.006 -0.012 0.000 0.000 0.000 0.166 -0.333 -0.500
## klmpk1:wkt3 0.000 0.012 -0.006 0.000 0.000 0.000 -0.334 0.167 -0.333 0.167
## klmpk2:wkt3 0.000 -0.006 0.012 0.000 0.000 0.000 0.167 -0.334 0.167 -0.333 -0.500
## optimizer (bobyqa) convergence code: 0 (OK)
## boundary (singular) fit: see help('isSingular')
##
##
## 41 Perbandingan LMM REML
## Data: dat_long
## Models:
## lmm1: asam_urat ~ kelompok * waktu + (1 | id)
## lmm2: asam_urat ~ kelompok * waktu + (1 + minggu | id)
## npar AIC BIC logLik -2*log(L) Chisq Df Pr(>Chisq)
## lmm1 14 679.57 733.97 -325.78 651.57
## lmm2 16 683.38 745.55 -325.69 651.38 0.189 2 0.9098
##
## Perbandingan model
## Kedua model mempunyai fixed effects yang sama. Perbandingan REML mengikuti materi; nilai p likelihood-ratio bersifat pendekatan karena varians acak diuji pada batas ruang parameter.
##
## 42 Singularitas LMM
## model singular
## 1 Intersep FALSE
## 2 Intersep dan slope TRUE
##
## 43 LMM Kenward Roger
## Type III Analysis of Variance Table with Kenward-Roger's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## kelompok 0.3473 0.1736 2 87.00 1.2903 0.2804
## waktu 19.4072 6.4691 3 185.37 47.8514 < 2.2e-16 ***
## kelompok:waktu 8.1681 1.3614 6 206.40 10.0587 9.554e-10 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## 44 ICC
## # Intraclass Correlation Coefficient
##
## Adjusted ICC: 0.888
## Unadjusted ICC: 0.815
##
## 45 EMM LMM
## kelompok = Kontrol:
## waktu emmean SE df lower.CL upper.CL
## M0 6.780000 0.1985502 93.20 6.385730 7.174270
## M4 6.780000 0.1998832 102.10 6.383537 7.176463
## M8 6.640000 0.2012174 101.88 6.240881 7.039119
## M12 6.746667 0.2025527 92.95 6.344434 7.148899
##
## kelompok = DASH:
## waktu emmean SE df lower.CL upper.CL
## M0 6.966667 0.1985502 93.20 6.572397 7.360937
## M4 6.680000 0.1998832 102.10 6.283537 7.076463
## M8 6.453333 0.2012174 101.88 6.054214 6.852453
## M12 6.176667 0.2025527 92.95 5.774434 6.578899
##
## kelompok = DASH+AF:
## waktu emmean SE df lower.CL upper.CL
## M0 6.830000 0.1985502 93.20 6.435730 7.224270
## M4 6.466667 0.1998832 102.10 6.070204 6.863129
## M8 6.083333 0.2012174 101.88 5.684214 6.482453
## M12 5.836667 0.2025527 92.95 5.434434 6.238899
##
## Degrees-of-freedom method: kenward-roger
## Confidence level used: 0.95
##
## Demonstrasi data hilang
## Sebanyak 30 pengukuran pascabaseline dihapus secara acak hanya pada salinan demonstrasi. Analisis utama tetap menggunakan seluruh data yang dilampirkan; demonstrasi ini mengikuti bagian LMM data hilang dalam materi.## boundary (singular) fit: see help('isSingular')##
## 46 LMM Data Hilang KR
## Type III Analysis of Variance Table with Kenward-Roger's method
## Sum Sq Mean Sq NumDF DenDF F value Pr(>F)
## kelompok 0.3146 0.1573 2 87.001 1.2176 0.3009
## waktu 18.4655 6.1552 3 167.832 47.4175 < 2.2e-16 ***
## kelompok:waktu 7.0931 1.1822 6 185.520 9.0966 1.007e-08 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## 47 Ringkasan Data Hilang
## pengukuran_hilang peserta_terdampak peserta_lengkap pengukuran_dianalisis_LMM
## 1 30 25 65 330
##
## 48 Informasi Sesi
## R version 4.6.1 (2026-06-24 ucrt)
## Platform: x86_64-w64-mingw32/x64
## Running under: Windows 11 x64 (build 26200)
##
## Matrix products: default
## LAPACK version 3.12.1
##
## locale:
## [1] LC_COLLATE=English_Indonesia.utf8 LC_CTYPE=English_Indonesia.utf8 LC_MONETARY=English_Indonesia.utf8
## [4] LC_NUMERIC=C LC_TIME=English_Indonesia.utf8
##
## time zone: Asia/Makassar
## tzcode source: internal
##
## attached base packages:
## [1] stats graphics grDevices utils datasets methods base
##
## other attached packages:
## [1] ggplot2_4.0.3 tidyr_1.3.2 dplyr_1.2.1
##
## loaded via a namespace (and not attached):
## [1] gtable_0.3.6 xfun_0.60 bslib_0.12.0 bayestestR_0.19.0 insight_1.5.4
## [6] rstatix_1.1.0 lattice_0.22-9 numDeriv_2016.8-1.1 vctrs_0.7.3 tools_4.6.1
## [11] Rdpack_2.6.6 generics_0.1.4 pbkrtest_0.5.5 datawizard_1.4.0 parallel_4.6.1
## [16] tibble_3.3.1 pkgconfig_2.0.3 Matrix_1.7-5 WRS2_1.1-7 RColorBrewer_1.1-3
## [21] S7_0.2.2 afex_1.5-1 lifecycle_1.0.5 compiler_4.6.1 farver_2.1.2
## [26] stringr_1.6.0 lmerTest_3.2-1 carData_3.0-6 htmltools_0.5.9 sass_0.4.10
## [31] yaml_2.3.12 Formula_1.2-6 ggpubr_1.0.0 pillar_1.11.1 car_3.1-5
## [36] nloptr_2.2.1 jquerylib_0.1.4 MASS_7.3-65 cachem_1.1.0 reformulas_0.4.4
## [41] boot_1.3-32 abind_1.4-8 nlme_3.1-169 tidyselect_1.2.1 digest_0.6.39
## [46] performance_0.18.2 mvtnorm_1.4-2 stringi_1.8.9 reshape2_1.4.5 purrr_1.2.2
## [51] labeling_0.4.3 splines_4.6.1 fastmap_1.2.0 grid_4.6.1 cli_3.6.6
## [56] magrittr_2.0.5 base64enc_0.1-6 utf8_1.2.6 broom_1.0.13 withr_3.0.3
## [61] scales_1.4.0 backports_1.5.1 estimability_2.0.0 rmarkdown_2.32 emmeans_2.0.4
## [66] lme4_2.0-6 ggsignif_0.6.4 evaluate_1.0.5 knitr_1.52 parameters_0.29.3
## [71] rbibutils_2.4.1 rlang_1.3.0 Rcpp_1.1.2 glue_1.8.1 reshape_0.8.10
## [76] rstudioapi_0.19.0 minqa_1.2.8 jsonlite_2.0.0 effectsize_1.0.3 R6_2.6.1
## [81] plyr_1.8.9
##
## Hasil analisis: D:\PELAJARAN KULIAH S2\New folder\Hasil_Analisis_Asam_Urat