Causal Analysis using R Path Analysis

Readme.txt

This R script provides a step-by-step demonstration and explanation of how to perform a path analysis using R. Path analysis is presented as a special case of structural equation modeling (SEM), utilizing R packages such as lavaan, sem, and semPlot.
A step by step video demonstration can be found at:  https://youtu.be/0eO54Bzmtmc 

R Script

 ### How to Perform a Path Analysis using R

# Path analysis is an extension of regression that models both direct and indirect relationships 
#   among a set of observed variables within a system of structured equations. 
# It allows variables to act as both predictors and outcomes, thus examining complex causal chains, 
#   not just single dependencies. 
# The method breaks down effects into direct effects (immediate influence) and indirect effects 
#   (mediated through other variables), clarifying how causal influence travels through the network. 
# Path analysis is commonly visualized using a path diagram, where arrows represent hypothesized 
#   causal connections, and it serves as a subset of structural equation modeling with only observed 
#   (not latent) variables. 
# Applications span psychology, social sciences, healthcare, marketing, and more, making it 
#    valuable for understanding real-world, multi-step causal processes. 



set.seed(8312025)
n <- 200
attitude <- rnorm(n)
norms <- rnorm(n)
control <- rnorm(n)
# Assume 'attitude' is predicted by 'norms'
attitude <- 0.5*norms + rnorm(n, sd=0.7)
# 'intention' is predicted by all three
intention <- 0.3*attitude + 0.3*norms + 0.2*control + rnorm(n, sd=0.6)

df <- data.frame(attitude, norms, control, intention)
head(df)

##install.packages("lavaan")
library(lavaan)

model <- '
  attitude ~ norms         # norms affects attitude
  intention ~ attitude + norms + control  # intention affected by attitude, norms, control
'
fit <- sem(model, data = df)
summary(fit, fit.measures=TRUE)   # check fit indices and estimates

#Chi-square Test: Test statistic = 0.007, p-value = 0.934. This high p-value indicates that the model fits the data very well (no significant difference between the model and the observed data).
#CFI (Comparative Fit Index) = 1.000, TLI = 1.029: Values above 0.95 indicate excellent fit.
#RMSEA (Root Mean Square Error of Approximation) = 0.000, 90% CI [0.000, 0.052]: Values below 0.06 are regarded as a good fit; your result is near zero (excellent).
#SRMR (Standardized Root Mean Square Residual) = 0.002: Extremely low, indicating minimal difference between observed and predicted co
#All predictors are statistically significant (p < .001); 
#their z-values (>3) reinforce the strength of effects.
#Variances
#.attitude variance = 0.421, .intention variance = 0.437: These are error variances (residuals) for each endogenous variable
#—the unexplained part by predictors.
#Both are significant, meaning not all of the variation 
#in attitude or intention is explained by the model.


###Path Diagram
#install.packages("semPlot")
#install.packages("sem")
library(sem)
library(semPlot)
semPaths(fit, whatLabels="std", layout="tree", edge.label.cex=0.9, title=FALSE)

VideoDemo

A step by step video demonstration can be found at:  https://youtu.be/0eO54Bzmtmc  

A step by step video demonstration can be found at: https://youtu.be/0eO54Bzmtmc