About the Data

This analysis was provided by a public data set via Kaggle (https://www.kaggle.com/datasets/grouplens/movielens-20m-dataset/data?select=link.csv). The original source of data comes from the website MovieLens (https://movielens.org/), a non-commercial, personalized movie recommendation-based site which includes data such as movie information, user ratings, and genres. Insights determined in this analysis are non-definitive, as they are only representative of available data gathered from MovieLens.

Movie Count

Movies in this data set span from years 1891-2015, with total movies being n = 27,278. However, some movies do not have a rating, so in later calculations measuring user rating, n = 26744. Movies released in 2009 had the highest total, n = 1,114. The 2000s had the largest number of movies released by decade, n = 7,880.

Genre

There are 20 distinct genres, including a ‘no genres listed’ for movies with no reported genre. As seen below, Drama had the largest quantity of movies released (n = 13,344), followed by Comedy (n = 9,374), Thriller (n = 4,178), then Romance (n = 4,127). While excluding genres ‘no genre’ and IMAX, the lowest count of movies were Film-Noir, Western, and Animation (n = 330, n = 676, n = 1,027, respectively).

User Ratings

MovieLens has a movie rating feature for users on a scale of 1-5. The number of distinctive users was n = 201,801, the overall number of ratings being n = 20,000,263. Thus, according to this data set, users rated on average 99 movies per user. The span of for ratings are from years 1995-2015. As seen below, the two graphs showing user ratings and distinct user ratings follow a similar pattern, with spikes in ratings counts in years 1996, 1998, 2005. In the third visual, user ratings were compared with movie releases by decade. Average ratings remained in the 3.5-4 score range, showing no correlation between movie release date and average user rating. In the fourth visual, movies with highest rating count are shown, with Shawshank Redemption (n = 40,569).

Top Movies (Analysis)

The table below shows movies with the highest average rating. The rating column represents the average rating (1-5), with Shawshank Redemption at the highest rank with a score of 4.54. In order for more accurate results, ratings were calculated using movies with more than 100 ratings, then were converted to z-scores (m = 3.45, sd = 0.522).

Top 10 Movies by Average Rating
title rating z_score
Shawshank Redemption The 4.54 0.84
Godfather The 4.46 0.76
Band of Brothers 4.44 0.74
City of God 4.43 0.74
Lives of Others The 4.44 0.74
Usual Suspects The 4.44 0.74
Schindler’s List 4.40 0.71
Dark Knight The 4.41 0.71
Spirited Away 4.39 0.70
Seven Samurai 4.38 0.69