I recently published a work with a dearest colleague of mine on the effects of gamification in collaborative learning on learning outcomes (cognitive and attitudinal learning outcomes). The paper is published in the Journal of Computers in Education in 2024.

The paper can be accessed through this URL: …

For those who are interested in understanding the deeper layer of analysis that we went through. We provided the R scripts that were utilized to conduct the analysis, including the meta-analysis and sub-group analysis. Any queries are welcomed and please address them to taufik.ikhsan.tep@um.ac.id.

Effect size with Cohen’s d

dataset2023a$Effect_Size <- escalc(measure = “SMD”,
m1i = dataset2023a$MTreatment,
sd1i = dataset2023a$SDTreatment,
n1i = dataset2023a$NTreatment,
m2i = dataset2023a$Mcontrol,
sd2i = dataset2023a$SDControl,
n2i = dataset2023a$NControl
)

Effect size with Hedge’s g

dataset2023a$Effect_Size_Hedges <- escalc(measure = “Hedges”,
m1i = dataset2023a$MTreatment,
sd1i = dataset2023a$SDTreatment,
n1i = dataset2023a$NTreatment,
m2i = dataset2023a$Mcontrol,
sd2i = dataset2023a$SDControl,
n2i = dataset2023a$NControl
)

Egger test

egger_test_result <- regtest(meta_analysis_result_hedges)
print(egger_test_result)

Begg test

begg_test_result <- cor.test(meta_analysis_result_hedges$yi, meta_analysis_result_hedges$vi, method = “kendall”)
print(begg_test_result)

str(meta_analysis_result_hedges)

meta_analysis_result_hedges$effect_size <- as.numeric(meta_analysis_result_hedges$effect_size)
meta_analysis_result_hedges$se <- as.numeric(meta_analysis_result_hedges$se)

Assuming you already have Cohen’s d standard errors in a variable named ‘SE_Cohen’

meta_analysis_result_hedges <- rma(yi = dataset2023$Effect_Size_Hedges, sei = dataset2023$SE_Cohen, method = “REML”)

Forrest plot for Hedge’s g

forest(meta_analysis_result_hedges)

Funnel plot for Hedge’s g

funnel(meta_analysis_result_hedges)

Assuming your meta-analysis result is stored in ‘meta_analysis_result_hedges’

hist(meta_analysis_result_hedges$yi, main = “Histogram of Overall Effect Sizes”, xlab = “Effect Size (Hedge’s g)”, col = “darkgrey”, border = “black”)

Add a density curve

lines(density(meta_analysis_result_hedges$yi), col = “darkred”, lwd = 3)

library(ggplot2)
install.packages(“ggplot2”)
ggplot(data.frame(EffectSize = meta_analysis_result_hedges$yi), aes(x = EffectSize)) +
geom_histogram(binwidth = 0.2, fill = “lightblue”, color = “black”, alpha = 0.7) +
geom_density(color = “darkred”, size = 1) +
labs(title = “Histogram with Density Curve”, x = “Effect Size (Hedge’s g)”, y = “Frequency”)

ggplot(data.frame(EffectSize = meta_analysis_result_hedges$yi), aes(x = EffectSize)) +
geom_histogram(binwidth = 0.2, fill = “lightblue”, color = “black”, alpha = 0.7) +
geom_density(aes(y = after_stat(count) * 0.2), color = “darkred”, size = 1) +
labs(title = “Histogram with Density Curve”, x = “Effect Size (Hedge’s g)”, y = “Frequency”)

Assuming ‘dataset2023’ is your dataset

meta_analysis_result <- rma(yi = dataset2023$Effect_Size$yi, sei = dataset2023$Effect_Size$vi, method = “REML”)
summary(meta_analysis_result)
ls()

Forrest plot

forest(meta_analysis_result)
forest(meta_analysis_result, showweights = TRUE, xlim = c(-2, 2), atransf = exp)

Funnel plot

funnel(meta_analysis_result)

Galbraith Plot

galbraith(meta_analysis_result)