Structural Equation Modeling: Short Dark Triad Survey

Project Overview

This project applies Structural Equation Modeling (SEM) to analyze the Short Dark Triad survey (Jones & Paulhus, 2013), exploring relationships between dark personality traits and additional variables.

Dataset

The dataset for this project is available on Kaggle: Short Dark Triad Dataset

Key Features

• Analysis of Dark Triad personality traits (Machiavellianism, Narcissism, Psychopathy)
• Incorporation of custom variables: risk-taking behavior and gender
• Data cleaning and preliminary analyses
• SEM model construction and evaluation
• Multi-group SEM analysis using gender as the grouping variable

Project Structure

Part 1: Data Preparation and Preliminary Analyses

• Synthetic variable creation
• Missing data check
• Normality assessment:
  • Univariate normality (Shapiro-Wilk's test)
  • Multivariate normality (Mardia's test)

Part 2: Initial SEM Model Construction

• Built the initial structural equation model
• Examined target indices for model fit

Part 3: Multi-group SEM Analysis

• Conducted multi-group SEM using gender as the grouping variable
• Assessed and compared different levels of measurement invariance:
  • Metric invariance
  • Scalar invariance
  • Strict invariance

Requirements

• R version 4.1.0 or later to use R package lavaangui

How to Use

1. Clone the repository
2. Open the RMarkdown file in your preferred R environment
3. Run the code chunks sequentially
4. Alternatively, view the HTML or Markdown (.md) file for a quick overview of the results

Feedback

Constructive feedback is always appreciated. Feel free to share any suggestions or improvements you might have.