Memuat
Memuat
Mediation and Moderation Analysis for Agriculture and Plantations
0 views
Testing the mediating role of technology adoption and the moderating role of farmer group activity in the effect of extension access on farm performance using synthetic data for demonstration. Mediation analysis tests whether the effect of X on Y is transmitted through a mediator M (the bootstrapped indirect effect a*b), while moderation analysis tests whether the size of X's effect depends on a moderator W. It suits theses that explain the mechanism and the conditions under which an effect holds.

Testing the mediating role of technology adoption and the moderating role of farmer group activity in the effect of extension access on farm performance using synthetic data for demonstration.
| Role | Variable | Unit |
|---|---|---|
| Main | Farm performance | score |
| Supporting 1 | Extension access | score |
| Supporting 2 | Technology adoption | score |
| Supporting 3 | Farmer group activity | score |
The indirect effect of Extension access on Farm performance through Technology adoption is 0.138 (95% bootstrap CI 0.072 to 0.207), significant; the direct effect is 0.352 (p < 0.001). The interaction with Farmer group activity is significant (beta = -0.211, p < 0.001): the standardised slope is 0.713 at low and 0.290 at high Farmer group activity.
Computed from illustrative synthetic data, not from a real study.
Mediation analysis tests whether the effect of X on Y is transmitted through a mediator M (the bootstrapped indirect effect a*b), while moderation analysis tests whether the size of X's effect depends on a moderator W. It suits theses that explain the mechanism and the conditions under which an effect holds.
Mediation is present when the bootstrap interval of the indirect effect a*b excludes zero; if the direct effect c' remains significant, mediation is partial. Moderation is present when the X x W interaction is significant; describe the pattern with simple slopes and the Johnson-Neyman region, the range of W where X has a significant effect.
A short starting script with this project's variable names. The delivered version is adapted to your data and fully commented.
# Contoh kode R: Mediation and Moderation Analysis
suppressPackageStartupMessages({
library(ggplot2)
library(patchwork)
library(lavaan)
})
set.seed(42)
# ---- Gaya grafik ----
pal <- c("#1f6feb", "#e8590c", "#2f9e44", "#ae3ec9", "#f59f00")
warna <- list(bg = "#f4f6fa", panel = "#ffffff", teks = "#1f2933", redup = "#5f6b78", grid = "#e3e8ee", bar = "#1d3557")
tata <- 1
theme_set(theme_minimal(base_size = 11) + theme(legend.position = "bottom"))
# ---- Data sintetis: X (variabel bebas), M (mediator), W (moderator), Y (hasil) ----
n <- 200
zx <- rnorm(n); zw <- rnorm(n)
a <- runif(1, 0.35, 0.6) # jalur X -> M
b <- runif(1, 0.25, 0.45) # jalur M -> Y
cp <- runif(1, 0.1, 0.3) # efek langsung X -> Y
h <- runif(1, 0.12, 0.25) * sample(c(-1, 1), 1) # interaksi X x W
zm <- a * zx + rnorm(n, 0, sqrt(1 - a^2))
zy <- cp * zx + b * zm + 0.2 * zw + h * zx * zw + rnorm(n, 0, runif(1, 0.6, 0.8))
zy <- as.numeric(scale(zy))
d <- data.frame(extension_access = round(50 + 10 * zx, 2), technology_adoption = round(30 + 6 * zm, 2),
x3 = round(20 + 5 * zw, 2), farm_performance = round(100 + 15 * zy, 2))
write.csv(d, "data_contoh.csv", row.names = FALSE)
# ---- Mediasi: X -> M -> Y dengan SE bootstrap (variabel dibakukan agar koefisien sebanding) ----
ds <- as.data.frame(scale(d))
model <- "
technology_adoption ~ a*extension_access
farm_performance ~ cp*extension_access + b*technology_adoption
tidak_langsung := a*b
total := cp + a*b
"
fit <- sem(model, data = ds, se = "bootstrap", bootstrap = 500)
pe <- parameterEstimates(fit, boot.ci.type = "perc")
print(pe[pe$op %in% c("~", ":="), c("lhs", "op", "rhs", "label", "est", "se", "pvalue", "ci.lower", "ci.upper")])
ambil <- function(lb) pe[pe$label == lb | pe$lhs == lb, ][1, ]
e_a <- ambil("a"); e_b <- ambil("b"); e_c <- ambil("cp"); e_i <- ambil("tidak_langsung"); e_t <- ambil("total")
bb <- lavInspect(fit, "boot") # sampel bootstrap koefisien
ab_boot <- bb[, "a"] * bb[, "b"]
# ---- Moderasi: regresi dengan interaksi (prediktor dipusatkan) ----
dm <- data.frame(yy = d$farm_performance, xp = d$extension_access - mean(d$extension_access), wp = d$x3 - mean(d$x3))
mod <- lm(yy ~ xp * wp, data = dm)
print(summary(mod))
sw <- sd(d$x3)
V <- vcov(mod)
lereng <- function(w) { # simple slope X pada nilai W tertentu (terpusat)
est <- coef(mod)[["xp"]] + coef(mod)[["xp:wp"]] * w
se <- sqrt(V["xp", "xp"] + w^2 * V["xp:wp", "xp:wp"] + 2 * w * V["xp", "xp:wp"])
c(est = est, se = se, p = 2 * pt(-abs(est / se), df.residual(mod)))
}
ss <- t(sapply(c(-sw, 0, sw), lereng)); rownames(ss) <- c("W -1 SD", "W mean", "W +1 SD")
print(round(ss, 4))
# Johnson-Neyman: nilai W saat simple slope X berubah signifikan
wg <- seq(min(dm$wp), max(dm$wp), length.out = 200)
jn <- as.data.frame(t(sapply(wg, lereng))); jn$w <- wg + mean(d$x3)
p_int <- summary(mod)$coefficients["xp:wp", 4]
b_int <- coef(lm(scale(yy) ~ scale(xp) * scale(wp), data = dm))[4] # koefisien interaksi terstandar
sig_pct <- 100 * mean(jn$p < 0.05)
# ---- Grafik ----
bintang <- function(p) ifelse(p < 0.001, "***", ifelse(p < 0.01, "**", ifelse(p < 0.05, "*", " ns")))
bungkus <- function(s) paste(head(strwrap(s, 14), 2), collapse = "\n") # nama maks. 2 baris
singkat <- function(s, w = 18) { # nama pendek untuk legenda
k <- strsplit(s, " ")[[1]]
if (nchar(s) > w) k <- k[!tolower(k) %in% c("of", "to", "the", "and", "for", "in", "on", "with")]
i <- length(k)
while (nchar(paste(k, collapse = " ")) > w && i >= 1) {
if (nchar(k[i]) > 5) k[i] <- paste0(substr(k[i], 1, 4), ".")
i <- i - 1
}
substr(paste(k, collapse = " "), 1, w)
}
nd <- data.frame(x = c(0, 1, 2), y = c(0, 1, 0), lab = c(bungkus("extension_access"), bungkus("technology_adoption"), bungkus("farm_performance")))
pan <- data.frame(x = c(0.2, 1.2, 0.36), y = c(0.2, 0.8, 0), xend = c(0.8, 1.8, 1.64), yend = c(0.8, 0.2, 0),
lab = c(sprintf("a = %.2f%s", e_a$est, bintang(e_a$pvalue)), sprintf("b = %.2f%s", e_b$est, bintang(e_b$pvalue)),
sprintf("c' = %.2f%s", e_c$est, bintang(e_c$pvalue))),
lx = c(0.28, 1.72, 1), ly = c(0.62, 0.62, -0.1))
p1 <- ggplot() +
geom_segment(data = pan, aes(x, y, xend = xend, yend = yend), colour = warna$redup, linewidth = 0.9,
arrow = arrow(length = unit(0.13, "inches"), type = "closed")) +
geom_text(data = pan, aes(lx, ly, label = lab), colour = warna$teks, size = 3.2, fontface = "bold") +
geom_tile(data = nd, aes(x, y), width = 0.8, height = 0.38, fill = pal[1:3]) +
geom_text(data = nd, aes(x, y, label = lab), colour = warna$panel, fontface = "bold", size = 2.8, lineheight = 0.9) +
annotate("text", x = 1, y = -0.34, colour = warna$teks, size = 3,
label = sprintf("Indirect a*b = %.3f [%.3f, %.3f]", e_i$est, e_i$ci.lower, e_i$ci.upper)) +
coord_cartesian(xlim = c(-0.45, 2.45), ylim = c(-0.42, 1.22)) +
theme_void() + theme(plot.background = element_rect(fill = warna$panel, colour = NA),
plot.title = element_text(face = "bold", colour = warna$teks, margin = margin(4, 0, 4, 6))) +
labs(title = "Mediation model (standardised)")
xs <- seq(min(dm$xp), max(dm$xp), length.out = 50)
dsl <- do.call(rbind, lapply(1:3, function(i) data.frame(
x = xs + mean(d$extension_access), y = predict(mod, data.frame(xp = xs, wp = c(-sw, 0, sw)[i])),
tingkat = c("Low W (-1 SD)", "Mean W", "High W (+1 SD)")[i])))
dsl$tingkat <- factor(dsl$tingkat, levels = c("Low W (-1 SD)", "Mean W", "High W (+1 SD)"))
p2 <- ggplot() +
geom_point(data = d, aes(extension_access, farm_performance), colour = warna$redup, alpha = 0.25, size = 1.1) +
geom_line(data = dsl, aes(x, y, colour = tingkat), linewidth = 1.1) +
scale_colour_manual(values = pal[c(4, 3, 5)], name = singkat("x3")) +
labs(title = "Simple slopes at low, mean and high W", x = "extension_access", y = "farm_performance")
p3 <- ggplot(data.frame(ab = ab_boot), aes(ab)) +
geom_histogram(bins = 30, fill = pal[2], colour = warna$panel, alpha = 0.85) +
geom_vline(xintercept = c(e_i$ci.lower, e_i$ci.upper), linetype = "dashed", colour = warna$teks) +
geom_vline(xintercept = 0, colour = warna$redup) +
labs(title = "Bootstrap distribution of a*b (500)", x = "Indirect effect (std.)", y = "Count")
p4 <- ggplot(jn, aes(w, est)) +
geom_ribbon(aes(ymin = est - 1.96 * se, ymax = est + 1.96 * se), fill = pal[1], alpha = 0.2) +
geom_hline(yintercept = 0, colour = warna$redup) +
geom_line(aes(colour = p < 0.05, group = 1), linewidth = 1.1) +
scale_colour_manual(values = c(`TRUE` = pal[1], `FALSE` = pal[2]), labels = c(`TRUE` = "p < .05", `FALSE` = "n.s."),
name = NULL, drop = FALSE) +
labs(title = "Johnson-Neyman: slope of X across W", x = "x3", y = "Simple slope of X")
p_hasil <- if (tata == 1) (p1 | p2) / (p3 | p4) else (p1 | p3) / (p2 | p4)
print(p_hasil)
The original project file is available. It is not just the picture above: you get the working R project. It can be downloaded after payment, and we send it by WhatsApp or email.
ZIP: analisis.R + data_contoh.csv + BACA-DULU.txt
Choose how much help you need. Each higher package includes everything in the lower ones. Prices exclude 11% VAT.
Package 1
US$25
excl. 11% VAT
Duration: 5 working days
Output: R files + result figures
Deliverables
Package 2 · most chosen
US$69
excl. 11% VAT
Duration: 9 working days
Output: Code + analysis + interpretation notes
Includes all of Package 1, plus
Package 3
US$150
excl. 11% VAT
Duration: 14-17 working days
Output: Code + analysis + guidance until ready to present
Includes all of Packages 1 & 2, plus
Yes. The original project file is a ZIP containing the complete analisis.R script, a data_contoh.csv sample dataset and a BACA-DULU.txt guide. It can be downloaded after payment, and we send it by WhatsApp or email. For your own data (Package 2 and 3) the script is adapted to your dataset.
Yes. Mediation and Moderation Analysis is commonly used at this level in Agriculture and Plantations. We adapt the analysis to your research question, your own data and your supervisor's requirements.
No. The sheet is computed from illustrative synthetic data so you can see the kind of output R Studio gives. For your order we run the same method on your own data.
That is the point of the packages. Every script is commented, and Package 3 includes two coaching sessions plus practice questions so you can explain the method and the results at your seminar and defence.
Each project has three packages: Package 1 (code and explanation) US$25, Package 2 (your data analysis and interpretation) US$69 and Package 3 (full guidance) US$150, excluding 11% VAT. Larger datasets, extra methods or tight deadlines adjust the quote after a short consultation.
The result sheet uses illustrative synthetic data. R is a programming language of the R Foundation project; RStudio and Posit are trademarks of Posit PBC.
Back to R Studio portfolio