BAIS 462 Assignment 6

IMDB Top 250 TV Shows

IMDB provides ratings and reviews for movies, TV shows, and celebrity content. It is a very popular source when searching through these categories. They also create various charts and rankings for these categories by using the rating from IMDB’s users. I will be scraping the “Top 250 TV Shows” chart. This will help me to answer my question of:

How does year released, number of episodes, and type of show influence the overall ranking of a show?

Download Data

After scraping this data and creating the dataframe, I created a CSV file that can be downloaded in order to follow this analysis. The link to download this file is below:

https://myxavier-my.sharepoint.com/:x:/g/personal/farfsings2_xavier_edu/EZPkukP5QIZNncNwbj15AvkBZ-li7ufBHev3S5EybblGJA?download=1

The packages required for this analysis are:

# Loading Packages
library(tidyverse)
library(rvest)
library(lubridate)
library(httr)

Data Dictionary

Show Rank
Show Name
Show Release Date
Number of Episodes
Show Rating
Show Type

Analysis

Distribution of Release Dates

This first visual I created was a histogram to show the distribution of shows in the top 250 chart based on their release date. This will help better understand the second visual, which shows the average rating based on release year.

The graph shows that most of the shows found in the top 250 chart have been released recently. This makes sense, as the longer a show has been out, the less popular it becomes. The highest frequency is seen around 2019. This may have to do with the pandemic, as many people spent more time watching tv at home and most likely found shows they really enjoyed during this time.

Average Rating Based on Release Year

The second visual shows an analysis of how the year a series was released impacts its average rating on IMDB. This will also show trends that have occurred throughout the years and help to show when the most popular shows were released.

The results of this line graph show that the highest average ratings are seen in series released in the 1970’s. After this peak, the average ratings are fairly low compared to this. There was another spike in ratings seen around 2008, but the ratings have still been low. As someone that prefers to watch shows rather than movies, I feel that the recent releases have not been as good as shows released when I was younger. So, this data makes sense in my opinion.

Average Ranking Based on Number of Episodes

The third visual shows the correlation between the number of episodes in a series and its average ranking. This will also provide a correlation coefficient to help determine the strength of the relationship between these two variables.

Correlation Coefficient = 0.06

Based on the graph and its correlation coefficient, there is a somewhat strong correlation (0.6) between the number of episodes in a series and its ranking. This shows that viewers tend to like when a series has more episodes compared to those that do not. This leads to my next analysis, which is based off the relationship between show ranking and series type (mini vs regular).

Average Show Rating Based on Series Type

This bar chart shows that the average show rating for the different show types, mini series versus normal series, is very similar. Prior to this analysis, I thought that mini series would have lower rankings, since people would want more episodes in a series that they are watching. However, this is not the case since they are so close in average ranking.

Relationship Between Show Rating, Number of Episodes, and Show Rank

This bubble graph shows the relationship between these three variables to help show how show rank is impacted. In the graph, the x-axis shows the show rating, the y-axis shows the number of episodes, and the size of the bubbles represents the show rank (1-250).

The results of this graph show that the top ranked shows do not have very many episodes. The highest ranked shows have less than 150 episodes. Many of the results ares seen with rankings between 8.5-8.75. The majority of these shows have less than 300 episodes. Based off this graph, it seems that the highest rated shows do not have many episodes.

Conclusion

After performing this analysis, it seems that a show’s rating and number of episodes have the most impact on its rank in the IMDB charts. The other variables, such as show type and release date, do not have as much of an impact on the rankings. However, it was interesting to see the years that most of the shows in the top 250 were released. There are definitely more variables that go into the rankings and viewer reviews for shows, so it would be interesting to collect more data to add to this data set and see what else impacts the rankings.

--- title: "BAIS 462 Assignment 6" editor: visual format: html: code-tools: TRUE # Allow the code tools option showing in the output. self-contained: TRUE # Embeds all components into a single HTML file. execute: warning: FALSE # Prevent all warnings from displaying in the output. message: FALSE # Prevent all messages from displaying in the output. --- ## IMDB Top 250 TV Shows IMDB provides ratings and reviews for movies, TV shows, and celebrity content. It is a very popular source when searching through these categories. They also create various charts and rankings for these categories by using the rating from IMDB's users. I will be scraping the "Top 250 TV Shows" chart. This will help me to answer my question of: How does year released, number of episodes, and type of show influence the overall ranking of a show? ## Download Data After scraping this data and creating the dataframe, I created a CSV file that can be downloaded in order to follow this analysis. The link to download this file is below: <https://myxavier-my.sharepoint.com/:x:/g/personal/farfsings2_xavier_edu/EZPkukP5QIZNncNwbj15AvkBZ-li7ufBHev3S5EybblGJA?download=1> The packages required for this analysis are: ```{r, echo=TRUE, results='hide'} # Loading Packages library(tidyverse) library(rvest) library(lubridate) library(httr) ``` ```{r, echo=FALSE, include=FALSE} imdb_top_250 <- read_csv("imdb_top_250.csv") ``` ## Data Dictionary - Show Rank - Show Name - Show Release Date - Number of Episodes - Show Rating - Show Type ## Analysis ### Distribution of Release Dates This first visual I created was a histogram to show the distribution of shows in the top 250 chart based on their release date. This will help better understand the second visual, which shows the average rating based on release year. ```{r, echo=FALSE} ggplot(imdb_top_250, aes(x = show_release)) + geom_histogram(fill = "red", color = "black", bins = 20) + labs(title = "Distribution of Release Year", x = "Release Year", y = "Frequency") ``` The graph shows that most of the shows found in the top 250 chart have been released recently. This makes sense, as the longer a show has been out, the less popular it becomes. The highest frequency is seen around 2019. This may have to do with the pandemic, as many people spent more time watching tv at home and most likely found shows they really enjoyed during this time. ### Average Rating Based on Release Year The second visual shows an analysis of how the year a series was released impacts its average rating on IMDB. This will also show trends that have occurred throughout the years and help to show when the most popular shows were released. ```{r, echo=FALSE} ggplot(imdb_top_250, aes(x = show_release, y = show_rating)) + geom_line(stat = "summary", fun = "mean") + labs(x = "Release Year", y = "Average Show Rating") + ggtitle("Average Show Rating by Release Year") + theme(axis.text.x = element_text(angle = 45, hjust = 1)) ``` The results of this line graph show that the highest average ratings are seen in series released in the 1970's. After this peak, the average ratings are fairly low compared to this. There was another spike in ratings seen around 2008, but the ratings have still been low. As someone that prefers to watch shows rather than movies, I feel that the recent releases have not been as good as shows released when I was younger. So, this data makes sense in my opinion. ### Average Ranking Based on Number of Episodes The third visual shows the correlation between the number of episodes in a series and its average ranking. This will also provide a correlation coefficient to help determine the strength of the relationship between these two variables. ```{r, echo=FALSE} # Calculate correlation coefficient correlation <- round(cor(log10(imdb_top_250$show_episodes), imdb_top_250$show_rank, use = "pairwise.complete.obs"), 2) cat("Correlation Coefficient =", correlation, "\n") ggplot(imdb_top_250, aes(x = show_episodes, y = show_rank)) + geom_point(stat = "summary", fun = "mean") + geom_smooth(method = "lm", se = FALSE) + scale_x_log10() + labs(x = "Number of Episodes", y = "Average Show Rank") + ggtitle("Average Show Rank by Number of Episodes") ``` Based on the graph and its correlation coefficient, there is a somewhat strong correlation (0.6) between the number of episodes in a series and its ranking. This shows that viewers tend to like when a series has more episodes compared to those that do not. This leads to my next analysis, which is based off the relationship between show ranking and series type (mini vs regular). ### Average Show Rating Based on Series Type ```{r, echo=FALSE, results='hide'} numeric_data <- imdb_top_250[, sapply(imdb_top_250, is.numeric)] correlation_matrix <- cor(numeric_data) correlation_table <- as.data.frame(correlation_matrix) summary(imdb_top_250) ``` ```{r, echo=FALSE} ggplot(imdb_top_250, aes(x = show_type, y = show_rating)) + geom_bar(stat = "summary", fun.y = "mean") + scale_y_continuous(breaks = seq(0, 10, by = 0.5), labels = seq(0, 10, by = 0.5)) + labs(x = "Show Type", y = "Average Show Rank") + ggtitle("Average Show Rating by Show Type") ``` This bar chart shows that the average show rating for the different show types, mini series versus normal series, is very similar. Prior to this analysis, I thought that mini series would have lower rankings, since people would want more episodes in a series that they are watching. However, this is not the case since they are so close in average ranking. ### Relationship Between Show Rating, Number of Episodes, and Show Rank This bubble graph shows the relationship between these three variables to help show how show rank is impacted. In the graph, the x-axis shows the show rating, the y-axis shows the number of episodes, and the size of the bubbles represents the show rank (1-250). ```{r, echo=FALSE} # Create a bubble chart ggplot(data = imdb_top_250, aes(x = show_rating, y = show_episodes, size = show_rank)) + geom_point(alpha = 0.6, color = "blue") + scale_size_continuous(range = c(2, 10)) + labs(title = "Relationship Between Show Rating, Number of Episodes, and Show Rank", x = "Show Rating", y = "Number of Episodes", size = "Show Rank") ``` The results of this graph show that the top ranked shows do not have very many episodes. The highest ranked shows have less than 150 episodes. Many of the results ares seen with rankings between 8.5-8.75. The majority of these shows have less than 300 episodes. Based off this graph, it seems that the highest rated shows do not have many episodes. ## Conclusion After performing this analysis, it seems that a show's rating and number of episodes have the most impact on its rank in the IMDB charts. The other variables, such as show type and release date, do not have as much of an impact on the rankings. However, it was interesting to see the years that most of the shows in the top 250 were released. There are definitely more variables that go into the rankings and viewer reviews for shows, so it would be interesting to collect more data to add to this data set and see what else impacts the rankings.