RespawnMetrics

Advanced gaming psychology analytics revealing the complex relationship between gaming behavior and mental health through machine learning and statistical analysis

0 Participants
0% ML Accuracy
0.000 Correlation
0 Variables
Explore Analytics View Research

Research Methods & Analysis

Comprehensive data science approach using advanced statistical techniques and machine learning

🗄️

SQL Database Analysis

Comprehensive SQLite database with multiple table joins analyzing gaming behavior, personality metrics, and mental health outcomes across participant demographics.

Database Structure:

  • Comprehensive gaming data with 24 variables
  • Gaming wellbeing data (1,500 sessions)
  • Steam games metadata integration
  • Complex SQL joins for multi-dimensional analysis
📊

Statistical Testing

Rigorous statistical analysis including ANOVA, correlation matrices, chi-square tests, and effect size calculations to identify significant patterns.

Statistical Methods:

  • ANOVA: F=385.48, p<0.001, η²=0.39
  • Pearson correlations with 95% CI
  • Chi-square: χ²=140.57, p<0.001
  • Cramér's V effect size: 0.242
🤖

Machine Learning Models

Random Forest classification and regression models achieving high accuracy in predicting gaming relationships and wellness outcomes.

Model Performance:

  • Classification accuracy: 80.6%
  • Regression R²: 0.578, RMSE: 13.17
  • Feature importance analysis
  • Cross-validation with train/test split
🧠

Psychological Classification

Novel multi-dimensional classification system identifying gaming relationship patterns beyond traditional addiction models using Big Five personality traits.

Classification Categories:

  • Healthy Enthusiasts: 15.9% (191 participants)
  • Moderate Gamers: 27.0% (324 participants)
  • Problematic Patterns: 57.1% (685 participants)
  • Custom wellness scoring algorithm
📈

Data Visualization

Comprehensive visualization suite including correlation heatmaps, distribution analysis, box plots, and scatter plots for pattern identification.

Visualization Types:

  • Correlation matrix heatmaps
  • Distribution analysis with normality tests
  • Box plots by demographic groups
  • Interactive scatter plot regression
⚙️

Feature Engineering

Custom functions with type hints and docstrings for gaming wellness scoring, behavior categorization, and predictive feature creation.

Engineering Functions:

  • Gaming wellness score calculation
  • Behavior pattern categorization
  • Predictive feature matrix creation
  • Big Five personality wellness composite

Live Research Dashboard

Explore the complete RespawnMetrics analysis with interactive visualizations

Loading RespawnMetrics Dashboard...

Initializing interactive analytics platform

Key Research Findings

Significant insights from comprehensive analysis of 1,200 participants

-0.658
Gaming-Wellness Correlation
Strong negative correlation between daily gaming hours and overall wellness scores (p < 0.001)
57.1%
Problematic Patterns
Percentage of participants showing concerning gaming behavior requiring intervention
0.578
ML Prediction R²
Random Forest model explains 57.8% of wellness score variance with RMSE of 13.17
3.76
Average Hours/Day
Mean daily gaming hours across all study participants (range: 0.35 - 15.00 hours)
385.48
ANOVA F-Statistic
Highly significant difference in gaming hours between wellness groups (η² = 0.39)
0.242
Cramér's V Effect
Medium effect size for association between gaming relationship and age group (χ² = 140.57)