3 Analisis Deskriptif

Bagian ini menyajikan statistik deskriptif untuk profil demografi responden, peta sebaran wilayah, dan statistik level item (Mean, SD, Skewness, Kurtosis).

3.1 Statistik Deskriptif Demografi

3.1.1 Jumlah Sampel

Kode

n_total <- nrow(data)
cat("Total Sampel (N):", n_total)

Total Sampel (N): 538

3.1.2 Jenis Kelamin

Kode

print_demo_table(
  data$Gender,
  c("1" = "Laki-laki", "2" = "Perempuan"),
  "Distribusi Jenis Kelamin"
)

Distribusi Jenis Kelamin
Kategori	Frekuensi	Persentase
Laki-laki	289	53.72%
Perempuan	249	46.28%

3.1.3 Usia

Kode

cat("Mean:", round(mean(data$Usia, na.rm = TRUE), 2), "tahun |",
    "SD:", round(sd(data$Usia, na.rm = TRUE), 2), "|",
    "Min:", min(data$Usia, na.rm = TRUE), "| Max:", max(data$Usia, na.rm = TRUE))

Mean: 19.74 tahun | SD: 1.39 | Min: 18 | Max: 24

3.1.4 Semester

Kode

print_demo_table(
    data$Semester,
    c("1" = "Semester 1-2", "2" = "Semester 2-3", "3" = "Semester 3-4",
      "4" = "Semester 5-6", "5" = "Semester 7 atau Lebih"),
    "Distribusi Semester"
)

Distribusi Semester
Kategori	Frekuensi	Persentase
Semester 1-2	174	32.34%
Semester 2-3	64	11.9%
Semester 3-4	157	29.18%
Semester 5-6	70	13.01%
Semester 7 atau Lebih	73	13.57%

3.1.4.1 Semester (Recode 3 Kategori)

Kode

data$Semester_Recode <- ifelse(data$Semester == 1, "Semester Awal (1-2)",
    ifelse(data$Semester %in% c(2, 3), "Semester Menengah (3-4)", "Semester Lanjut (>=5)"))
sem_recode_freq <- table(data$Semester_Recode, useNA = "ifany")
sem_recode_pct  <- prop.table(sem_recode_freq) * 100
knitr::kable(data.frame(
    Kategori = names(sem_recode_freq), Frekuensi = as.numeric(sem_recode_freq),
    Persentase = paste0(round(sem_recode_pct, 2), "%"), check.names = FALSE
), caption = "Semester (Recoded — 3 Kategori)", row.names = FALSE)

Semester (Recoded — 3 Kategori)
Kategori	Frekuensi	Persentase
Semester Awal (1-2)	174	32.34%
Semester Lanjut (>=5)	143	26.58%
Semester Menengah (3-4)	221	41.08%

3.1.5 Area Tempat Tinggal

Kode

print_demo_table(
  data$Area,
  c("1" = "Urban", "2" = "Suburban", "3" = "Rural"),
  "Distribusi Area Tempat Tinggal"
)

Distribusi Area Tempat Tinggal
Kategori	Frekuensi	Persentase
Urban	145	26.95%
Suburban	157	29.18%
Rural	236	43.87%

3.1.6 Tipe Hunian

Kode

print_demo_table(
  data$Hunian,
  c("1" = "Kos/Kontrakan", "2" = "Rumah Orang Tua/Wali", "3" = "Asrama", "4" = "Lainnya"),
  "Distribusi Tipe Hunian"
)

Distribusi Tipe Hunian
Kategori	Frekuensi	Persentase
Kos/Kontrakan	345	64.13%
Rumah Orang Tua/Wali	142	26.39%
Asrama	42	7.81%
Lainnya	9	1.67%

3.1.7 Pengetahuan Lingkungan

Kode

print_demo_table(
  data$Knowl,
  c("0" = "Tidak", "1" = "Ya"),
  "Pengetahuan tentang Lingkungan"
)

Pengetahuan tentang Lingkungan
Kategori	Frekuensi	Persentase
Tidak	333	61.9%
Ya	205	38.1%

3.1.8 Lama Bergabung dengan Organisasi Lingkungan

Kode

print_demo_table(
  data$Bergabung,
  c("1" = "1-2 Tahun", "2" = "2-3 Tahun", "3" = "Lebih dari 3 Tahun"),
  "Lama Bergabung"
)

Lama Bergabung
Kategori	Frekuensi	Persentase
1-2 Tahun	378	70.26%
2-3 Tahun	106	19.7%
Lebih dari 3 Tahun	54	10.04%

3.1.9 Provinsi

Kode

print_demo_table(
  data$Provinsi,
  c("1" = "Jawa Barat", "2" = "Jawa Tengah", "3" = "Jawa Timur", 
    "4" = "Banten", "5" = "DIY", "6" = "DKI Jakarta"),
  "Distribusi Provinsi"
)

Distribusi Provinsi
Kategori	Frekuensi	Persentase
Jawa Barat	33	6.13%
Jawa Tengah	263	48.88%
Jawa Timur	51	9.48%
Banten	3	0.56%
DIY	177	32.9%
DKI Jakarta	11	2.04%

3.1.10 Status Anggota

Kode

print_demo_table(
  data$Anggota,
  c("1" = "Pengurus", "2" = "Anggota Biasa"),
  "Status Anggota"
)

Status Anggota
Kategori	Frekuensi	Persentase
Pengurus	176	32.71%
Anggota Biasa	362	67.29%

3.2 Peta Sebaran Responden

Kode peta sebaran responden

library(sf)
library(geodata)
library(ggtext)

data_responden <- data.frame(
    NAME_1 = c("Jawa Barat", "Jawa Tengah", "Jawa Timur", "Banten", "Yogyakarta", "Jakarta Raya"),
    Nama_Dalam = c("Jawa Barat", "Jawa Tengah", "Jawa Timur", "Banten", "DIY", "DKI Jakarta"),
    Label_Luar = c(
        "italic(n)==33~~'(6.1%)'", "italic(n)==263~~'(48.9%)'",
        "italic(n)==51~~'(9.5%)'", "italic(n)==3~~'(0.6%)'",
        "italic(n)==177~~'(32.9%)'", "italic(n)==11~~'(2.0%)'"
    ),
    Frekuensi = c(33, 263, 51, 3, 177, 11),
    stringsAsFactors = FALSE
)

gadm_dir <- file.path(getwd(), "gadm_cache")
if (!dir.exists(gadm_dir)) dir.create(gadm_dir)
jawa_sf <- geodata::gadm("IDN", level = 1, path = gadm_dir) |>
    sf::st_as_sf() |>
    dplyr::filter(NAME_1 %in% data_responden$NAME_1) |>
    dplyr::left_join(data_responden, by = "NAME_1")

suppressWarnings(cents <- sf::st_centroid(jawa_sf))
coords <- sf::st_coordinates(cents)
jawa_sf$cx <- coords[, 1]; jawa_sf$cy <- coords[, 2]

offset <- list(
    "Banten" = c(105.87, -6.85, 105.40, -7.40),
    "Jakarta Raya" = c(106.83, -6.08, 107.00, -5.53),
    "Jawa Barat" = c(108.30, -6.23, 108.30, -5.53),
    "Jawa Tengah" = c(110.10, -6.84, 110.10, -5.53),
    "Yogyakarta" = c(110.45, -8.12, 110.45, -8.70),
    "Jawa Timur" = c(112.73, -7.00, 112.73, -5.53)
)

jawa_sf <- jawa_sf |> dplyr::mutate(
    ax = dplyr::case_when(NAME_1 == "Banten" ~ offset[["Banten"]][1], NAME_1 == "Jakarta Raya" ~ offset[["Jakarta Raya"]][1], NAME_1 == "Jawa Barat" ~ offset[["Jawa Barat"]][1], NAME_1 == "Jawa Tengah" ~ offset[["Jawa Tengah"]][1], NAME_1 == "Yogyakarta" ~ offset[["Yogyakarta"]][1], NAME_1 == "Jawa Timur" ~ offset[["Jawa Timur"]][1]),
    ay = dplyr::case_when(NAME_1 == "Banten" ~ offset[["Banten"]][2], NAME_1 == "Jakarta Raya" ~ offset[["Jakarta Raya"]][2], NAME_1 == "Jawa Barat" ~ offset[["Jawa Barat"]][2], NAME_1 == "Jawa Tengah" ~ offset[["Jawa Tengah"]][2], NAME_1 == "Yogyakarta" ~ offset[["Yogyakarta"]][2], NAME_1 == "Jawa Timur" ~ offset[["Jawa Timur"]][2]),
    lx = dplyr::case_when(NAME_1 == "Banten" ~ offset[["Banten"]][3], NAME_1 == "Jakarta Raya" ~ offset[["Jakarta Raya"]][3], NAME_1 == "Jawa Barat" ~ offset[["Jawa Barat"]][3], NAME_1 == "Jawa Tengah" ~ offset[["Jawa Tengah"]][3], NAME_1 == "Yogyakarta" ~ offset[["Yogyakarta"]][3], NAME_1 == "Jawa Timur" ~ offset[["Jawa Timur"]][3]),
    ly = dplyr::case_when(NAME_1 == "Banten" ~ offset[["Banten"]][4], NAME_1 == "Jakarta Raya" ~ offset[["Jakarta Raya"]][4], NAME_1 == "Jawa Barat" ~ offset[["Jawa Barat"]][4], NAME_1 == "Jawa Tengah" ~ offset[["Jawa Tengah"]][4], NAME_1 == "Yogyakarta" ~ offset[["Yogyakarta"]][4], NAME_1 == "Jawa Timur" ~ offset[["Jawa Timur"]][4]),
    nx = dplyr::case_when(NAME_1 == "Banten" ~ 106.20, NAME_1 == "Jakarta Raya" ~ cx + 0.05, NAME_1 == "Jawa Tengah" ~ cx - 0.25, NAME_1 == "Jawa Timur" ~ cx - 0.10, TRUE ~ cx),
    ny = dplyr::case_when(NAME_1 == "Banten" ~ -6.38, NAME_1 == "Jawa Tengah" ~ cy + 0.15, NAME_1 == "Jawa Timur" ~ cy - 0.10, TRUE ~ cy)
)

ggplot(jawa_sf) +
    geom_sf(aes(fill = Frekuensi), color = "#777777", linewidth = 0.30) +
    scale_fill_distiller(palette = "YlGnBu", direction = 1, name = "Jumlah Responden (n)",
        breaks = c(3, 50, 100, 177, 263), labels = scales::comma, limits = c(0, 270),
        guide = guide_colorbar(title.position = "top", title.hjust = 0.5,
                               barwidth = 17, barheight = 0.6)) +
    geom_text(aes(x = nx, y = ny, label = Nama_Dalam), size = 3.0, fontface = "bold", color = "#0a1628") +
    geom_segment(aes(x = ax, y = ay, xend = lx, yend = ly), color = "#888888", linewidth = 0.28) +
    geom_label(aes(x = lx, y = ly, label = Label_Luar), parse = TRUE, size = 3.0,
               fill = "white", color = "#1a1a2e", alpha = 0.95, linewidth = 0.20,
               label.r = unit(0.10, "lines"), label.padding = unit(0.25, "lines")) +
    labs(caption = "<i>Catatan.</i> N = 538. Peta menggunakan shapefile GADM v4.1.") +
    coord_sf(xlim = c(104.0, 115.5), ylim = c(-9.2, -5.0), expand = FALSE) +
    theme_minimal(base_size = 11) +
    theme(plot.caption = ggtext::element_markdown(size = 9, hjust = 0, color = "#666666"),
          legend.position = "bottom", panel.grid = element_blank(),
          axis.text = element_blank(), axis.ticks = element_blank(), axis.title = element_blank(),
          panel.background = element_rect(fill = "#EAF3FB", color = NA))

3.3 Statistik Deskriptif Level Item

Statistik deskriptif semua item P (PEB), C (CCS), dan O (OL) untuk data screening dan pengecekan normalitas univariat.

Kode

item_cols <- names(data)[grep("^P\\d+$|^C\\d+$|^O\\d*$", names(data))]
item_desc <- psych::describe(data[, item_cols], type = 2)
item_stats <- data.frame(
    Item = item_cols, N = item_desc$n,
    Mean = round(item_desc$mean, 3), SD = round(item_desc$sd, 3),
    Min = item_desc$min, Max = item_desc$max,
    Skewness = round(item_desc$skew, 3), Kurtosis = round(item_desc$kurtosis, 3),
    check.names = FALSE
)
knitr::kable(item_stats, caption = "Statistik Deskriptif Level Item", row.names = FALSE)

Statistik Deskriptif Level Item
Item	N	Mean	SD	Min	Max	Skewness	Kurtosis
C1	538	3.307	1.484	1	7	0.157	-0.580
C2	538	3.273	1.606	1	7	0.391	-0.798
C3	538	2.115	1.254	1	7	1.122	0.634
C4	538	2.745	1.439	1	7	0.659	-0.174
C5	538	3.052	1.563	1	7	0.483	-0.473
C6	538	2.375	1.271	1	7	0.665	-0.259
C7	538	2.684	1.145	1	6	0.133	-1.062
C8	538	2.364	1.495	1	7	0.988	0.192
C9	538	2.688	1.591	1	7	0.788	-0.226
C10	538	2.946	1.547	1	7	0.545	-0.403
C11	538	2.463	1.547	1	7	1.029	0.323
C12	538	2.063	1.100	1	5	0.776	-0.634
P1	538	4.249	0.988	1	5	-1.259	0.915
P2	538	3.794	1.145	1	5	-0.665	-0.352
P3	538	3.487	1.187	1	5	-0.395	-0.684
P4	538	3.779	0.994	1	5	-0.265	-0.747
P5	538	4.424	0.845	1	5	-1.377	1.191
P6	538	4.145	0.879	1	5	-0.700	-0.273
P7	538	3.301	1.175	1	5	-0.284	-0.519
P8	538	3.916	1.026	1	5	-0.517	-0.506
P9	538	4.288	0.870	1	5	-0.900	-0.062
P10	538	3.561	1.054	1	5	-0.229	-0.356
P11	538	3.375	0.899	1	5	0.114	0.006
P12	538	3.593	0.852	1	5	0.034	-0.147
P13	538	3.416	0.885	1	5	0.101	0.031
P14	538	3.582	0.858	1	5	0.011	-0.076
P15	538	3.738	0.854	1	5	0.063	-0.732
P16	538	3.059	1.194	1	5	-0.128	-0.719
P17	538	3.517	0.969	1	5	-0.300	-0.080
P18	538	2.455	1.284	1	5	0.481	-0.817
P19	538	2.786	1.330	1	5	0.159	-1.071
P20	538	3.753	0.953	1	5	-0.459	0.095
O1	538	3.831	0.830	1	5	-0.225	-0.208
O2	538	3.848	0.823	1	5	-0.194	-0.347
O3	538	3.822	0.860	1	5	-0.354	0.002
O4	538	3.619	0.873	1	5	-0.157	-0.177
O5	538	3.894	0.813	1	5	-0.200	-0.376
O6	538	4.132	0.805	1	5	-0.437	-0.694
O7	538	4.024	0.826	1	5	-0.443	-0.131
O8	538	3.862	0.859	1	5	-0.281	-0.242
O9	538	4.030	0.797	1	5	-0.297	-0.550
O10	538	3.879	0.831	1	5	-0.161	-0.705
O11	538	3.978	0.818	1	5	-0.286	-0.535
O12	538	3.875	0.838	1	5	-0.201	-0.558
O13	538	3.974	0.807	1	5	-0.209	-0.749
O14	538	3.853	0.834	1	5	-0.280	-0.176
O15	538	3.987	0.773	1	5	-0.123	-0.828

Catatan Metodologis

Skewness & Kurtosis: Type 2 (psych::describe), konsisten dengan JASP/SPSS
Kurtosis = Excess Kurtosis (distribusi normal = 0)
Threshold (West et al., 1995): Skewness > |2.0|, Kurtosis > |7.0|

3.3.1 Visualisasi Deskriptif Item — Dot Plot

Kode plot deskriptif item

# ---- Bangun data frame plot dari item_stats ----
item_plot_df <- item_stats |>
    mutate(
        prefix = gsub("[0-9]", "", Item),
        Midpoint = ifelse(prefix == "C", 4, 3),
        MaxSkala = ifelse(prefix == "C", 7, 5),
        Skala = case_when(
            prefix == "C" ~ "CCS",
            prefix == "O" ~ "OL",
            prefix == "P" ~ "PEB"
        ),
        item_num = as.integer(gsub("[A-Za-z]", "", Item))
    ) |>
    arrange(prefix, Mean) |>
    mutate(
        Skala   = factor(Skala, levels = c("CCS", "OL", "PEB")),
        Item    = factor(Item, levels = unique(Item)),
        label_x = Mean + SD + 0.08
    )

# Data zona (di bawah / di atas midpoint)
zone_df <- item_plot_df |> distinct(Skala, Midpoint, MaxSkala)

# Batas x-axis (memaksa 1–MaxSkala per panel via geom_blank)
ref_items <- item_plot_df |>
    group_by(Skala) |>
    slice(1) |>
    ungroup() |>
    select(Skala, Item)

limits_df <- bind_rows(
    zone_df |> select(Skala) |> mutate(x_lim = 1),
    zone_df |> transmute(Skala, x_lim = MaxSkala)
) |>
    left_join(ref_items, by = "Skala")

# Anotasi: rentang mean + interpretasi per panel
range_annot <- item_plot_df |>
    summarise(
        lo = min(Mean),
        hi = max(Mean),
        Midpoint = first(Midpoint),
        MaxSkala = first(MaxSkala),
        .by = Skala
    ) |>
    mutate(
        skala_label = case_when(Skala == "CCS" ~ "Skala 1\u20137", TRUE ~ "Skala 1\u20135"),
        x_center    = (1 + MaxSkala) / 2,
        label       = sprintf("%s, Mean: %.2f\u2013%.2f", skala_label, lo, hi)
    )

# ---- Plot ----
p_item_means <- ggplot(item_plot_df, aes(x = Mean, y = Item)) +
    # Zona: di bawah midpoint (merah muda lembut)
    geom_rect(
        data = zone_df,
        aes(xmin = 1, xmax = Midpoint, ymin = -Inf, ymax = Inf),
        inherit.aes = FALSE, fill = "#FFF0EE"
    ) +
    # Zona: di atas midpoint (biru muda lembut)
    geom_rect(
        data = zone_df,
        aes(xmin = Midpoint, xmax = MaxSkala, ymin = -Inf, ymax = Inf),
        inherit.aes = FALSE, fill = "#EDF4FC"
    ) +
    # Paksa batas sumbu x sesuai rentang skala asli
    geom_blank(data = limits_df, aes(x = x_lim, y = Item), inherit.aes = FALSE) +
    # Garis midpoint
    geom_vline(
        data = zone_df, aes(xintercept = Midpoint),
        linetype = "longdash", color = "grey40", linewidth = 0.5
    ) +
    # Error bar \u00b11 SD
    geom_errorbarh(aes(xmin = Mean - SD, xmax = Mean + SD),
        height = 0.25, linewidth = 0.3, color = "grey55"
    ) +
    # Titik mean — warna berdasarkan posisi terhadap midpoint
    geom_point(aes(fill = Mean < Midpoint),
        shape = 21, size = 2.3,
        color = "grey30", stroke = 0.3
    ) +
    scale_fill_manual(
        values = c("TRUE" = "#E74C3C", "FALSE" = "#2980B9"),
        labels = c(
            "TRUE" = "Di bawah midpoint",
            "FALSE" = "Di atas midpoint"
        ),
        name = "Mean"
    ) +
    # Label nilai mean — di kanan ujung error bar (tidak menimpa titik/garis)
    geom_text(aes(x = label_x, label = sprintf("%.2f", Mean)),
        hjust = 0, vjust = 0.4, size = 2.2, color = "grey15",
        fontface = "bold"
    ) +
    # Anotasi skala + rentang mean (di luar panel, bawah x-axis)
    geom_text(
        data = range_annot,
        aes(x = x_center, y = -Inf, label = label),
        inherit.aes = FALSE, vjust = 3.8, hjust = 0.5,
        size = 2.8, color = "grey30", fontface = "bold"
    ) +
    coord_cartesian(clip = "off") +
    # Facet: 3 kolom (CCS | OL | PEB)
    facet_wrap(~Skala, scales = "free", ncol = 3) +
    # Integer breaks — tidak ada desimal di sumbu x
    scale_x_continuous(
        breaks = function(lims) seq(ceiling(lims[1]), floor(lims[2])),
        expand = expansion(mult = c(0.02, 0.12))
    ) +
    labs(
        title = "Item-Level Descriptive Statistics",
        x = NULL,
        y = NULL,
        caption = paste0(
            "N = ", nrow(data), ".  ",
            "Zona merah muda = di bawah midpoint (rendah); ",
            "zona biru muda = di atas midpoint (tinggi).\n",
            "Garis putus-putus = titik tengah skala (midpoint). ",
            "Garis horizontal = \u00b11 SD."
        )
    ) +
    theme_bw(base_size = 10) +
    theme(
        plot.title = element_text(face = "bold", size = 13, hjust = 0.5),
        plot.caption = element_text(
            size = 7.5, color = "grey45", hjust = 0,
            margin = margin(t = 8), lineheight = 1.2
        ),
        plot.margin = margin(10, 12, 35, 10),
        axis.text.y = element_text(size = 9, face = "bold"),
        axis.text.x = element_text(size = 9, face = "bold"),
        strip.text = element_text(face = "bold", size = 12),
        strip.background = element_rect(fill = "grey93", color = "grey70"),
        panel.grid.major.y = element_blank(),
        panel.grid.major.x = element_line(color = "grey92", linewidth = 0.25),
        panel.grid.minor = element_blank(),
        panel.spacing = unit(1.0, "lines"),
        legend.position = "bottom",
        legend.direction = "horizontal",
        legend.title = element_text(face = "bold", size = 9),
        legend.text = element_text(size = 9),
        legend.key.size = unit(0.5, "cm"),
        legend.margin = margin(t = 5, b = 0),
        legend.box.margin = margin(t = 0)
    )

print(p_item_means)