content_rating.Rmd

---
title: "Audience/Critic Rating"
---
Let's see if the movies on my Plex server are even any good.

```{r message=FALSE, echo=FALSE, error=FALSE, warning=FALSE}
# load libraries
library(readr)
library(dplyr)
library(tidyr)
library(ggplot2)
```

To analyse all the movies throughout all the libraries, we first load the existing `.csv` files with the data from Plex. We then combine them into one giant data frame.
```{r warning=FALSE, message=FALSE, error=FALSE}
all_movie_libraries <- list.files(path = "data", pattern = "*.csv", full.names = TRUE) %>%
                              lapply(read.csv) %>%
                              lapply(\(x) mutate(x, across(Audience.Rating, as.double))) %>%
                              bind_rows()
```
We will now filter out all the unused columns and also all the movies which haven't got an audience and/or critic rating.
```{r warning=FALSE, message=FALSE, error=FALSE, echo=FALSE}
rows_before_filtering <- nrow(all_movie_libraries)
```

```{r warning=FALSE, message=FALSE, error=FALSE}
all_movie_libraries <- all_movie_libraries %>%
                        select(title = Title, critic_rating = Rating, audience_rating = Audience.Rating) %>%
                        filter(critic_rating != 'N/A', audience_rating != 'N/A') %>%
                        mutate(critic_rating = as.numeric(critic_rating), audience_rating = as.numeric(audience_rating))
head(all_movie_libraries)
```
This brings us down from `r rows_before_filtering` to `r nrow(all_movie_libraries)` movies. This is now pretty nice to look at but won't work with ggplot2.
For that the data has to be transformed from "wide" to "long". Like that:
```{r warning=FALSE, message=FALSE, error=FALSE}
wide_data <- all_movie_libraries %>%
                gather('critic_rating','audience_rating', key='rating_type', value='rating')
```
This means we now have two rows for each movie:
```{r warning=FALSE, message=FALSE, error=FALSE}
print(wide_data %>% filter(title == 'Animal Farm'))
```
With this ggplot2 can now generate a nice box plot.  
```{r warning=FALSE, message=FALSE, error=FALSE}
wide_data %>%
  ggplot(aes(y = rating, x = rating_type)) +
  geom_boxplot() +
  labs(title = 'Movie Ratings', y = 'Score from 0 to 10', x = 'Rating Type') +
  theme_classic()
```  
  
We can see a couple things here:  
1. That I have a pretty good taste in movies, with the critic rating median being `r median(all_movie_libraries$critic_rating)` and the audience median being `r median(all_movie_libraries$audience_rating)`.  
2. That the critic and audience are generally in agreement.  
3. That there are many out liners on the "not-so-good-a-score" side.

I want to take a closer look at the last point some more code that we can use for an informative table:
```{r warning=FALSE, message=FALSE, error=FALSE}
#Using two functions because I could not get the column name as an argument to work
get_movie_amount_critic <- function(score){
  amount <- nrow(all_movie_libraries %>%
                  filter(critic_rating <= score))
  return(paste0(amount, ' (', get_percent(amount), '%)'))
} 


get_movie_amount_audience <- function(score){
  amount <- nrow(all_movie_libraries %>%
                  filter(audience_rating <= score))
  return(paste0(amount, ' (', get_percent(amount), '%)'))
} 

get_percent <- function(value){
  return(round((value * 100)/nrow(all_movie_libraries), digits=2))
}

lowest_score_row_critic <- all_movie_libraries[which.min(all_movie_libraries$critic_rating),]
lowest_score_row_audience <- all_movie_libraries[which.min(all_movie_libraries$audience_rating),]
```  
|                                    | Critic Rating                                                                                | Audience Rating                                                                                   |
|------------------------------------|----------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------------------------|
| Amount of movies with rating < 5.0 | `r get_movie_amount_critic(5.0)`                                                             | `r get_movie_amount_audience(5.0)`                                                                |
| Amount of movies with rating < 2.5 | `r get_movie_amount_critic(2.5)`                                                             | `r get_movie_amount_audience(2.5)`                                                                |
| Amount of movies with rating < 1.0 | `r get_movie_amount_critic(1.0)`                                                             | `r get_movie_amount_audience(1.0)`                                                                |
| Lowest scoring movie               | `r lowest_score_row_critic$title` with a score of `r lowest_score_row_critic$critic_rating`  | `r lowest_score_row_audience$title` with a score of `r lowest_score_row_audience$audience_rating` |