Movie Analysis: What Makes a Great Movie?

Project Overview

This project analyses IMDb's top 1000 movies dataset to understand what factors contribute to high movie ratings and how audience perceptions compare to critic evaluations.

Research Questions

1. Do IMDb scores differ from Metacritic? - Explored differences between audience and critic ratings
2. How have movie ratings changed over time? - Analyzed rating trends across different decades
3. What factors best predict movie ratings? - Identified variables with strongest correlation to high ratings

Key Findings

• Audience vs. Critic Divergence: Statistical testing (p<0.05) confirmed significant differences between IMDb ratings and Metacritic scores
• Rating Trends: IMDb ratings are more stable over time while Metacritic scores show greater fluctuation
• Predictive Factors: Runtime, director experience, and release year have small but significant effects on ratings
• Limited Explanatory Power: Our multiple regression model explains ~14% of rating variance, suggesting movie success is influenced by many complex factors

Methodology

• Data cleaning and preparation
• Statistical tests including t-tests
• Single and multiple linear regression
• Feature engineering (certificate categorisation, star power metrics)
• Time series analysis
• Data visualisation (density plots, violin plots, regression plots)

Technologies Used

• Python (pandas, seaborn, matplotlib, numpy, scipy.stats, statsmodels)

Getting Started

git clone https://github.com/Doctor2007/movie_analysis.git
cd movie_analysis
pip install -r requirements.txt
jupyter notebook main.ipynb