---
title: "Using the Student and School Data" 
output: rmarkdown::html_vignette
vignette: >
  %\VignetteIndexEntry{learningtower_school}
  %\VignetteEncoding{UTF-8}
  %\VignetteEngine{knitr::rmarkdown}
editor_options: 
  chunk_output_type: console
---

```{r setup, include = FALSE}
knitr::opts_chunk$set(
echo = TRUE,
collapse = TRUE,
comment = "#>",
warning = FALSE,
message = FALSE,
error = FALSE,
outwidth = "100%",
fig.width = 6,
fig.height = 4,
fig.align = "center")
```

# Introduction

The goal of `learningtower` is to provide a user-friendly R package to provide easy access to a subset of variables from PISA data collected from the [OECD](https://www.oecd.org/en/about/programmes/pisa/pisa-data.html). Version `r utils::packageVersion("learningtower")` of this package provides the data for the years `r learningtower:::data_range()`. The survey data is published every three years. This is an excellent real world dataset for data exploring, data visualizing and statistical computations.

This vignette documents how to access the data, and shows a few ways of integrating the data.

# Using both the `student` and `school` data

The size of the full `student` is too big to fit inside the package. Hence, in our package, we provide a random subset of the student data, stored as `student_subset_yyyy` data objects (where `yyyy` denotes the specific year of the study). These subset data can be used to understanding the data structure before using the full dataset which is available for download.

In the `student_subset_2018` and `school` data, there are three common columns, `school_id`, `country` and `year`. It should be noted that `school_id` is only meaningful within a country within a specific year; meaning that when we join the two data, we need to use the keys `c("school_id", "country", "year")`.

## Using the student subset data and school data

```{r}
library(dplyr)
library(ggplot2)
library(forcats)
library(learningtower)

#loading the student subset data 
data(student_subset_2018)

#loading the school data
data(school)

#loading the country data
data(countrycode)

selected_countries = c("AUS", "FIN", "JPN", "USA", "NZL", "ESP")

#joining the student, school dataset
school_student_subset_2018 <- left_join(
  student_subset_2018, 
  school, 
  by = c("school_id", "country", "year")) 

#check the count of public and private schools in the a few randomly selected countries
school_student_subset_2018 |> 
  dplyr::filter(country %in% selected_countries) |>
  group_by(country, public_private) |> 
  tally() |> 
  dplyr::mutate(percent = n/sum(n)) |> 
  dplyr::ungroup() |> 
  left_join(countrycode, by = "country") |> 
  ggplot(aes(x = percent,
             y = country_name,
             fill = public_private)) +
  geom_col(position = position_stack()) +
  scale_x_continuous(labels = scales::percent) +
  scale_fill_manual(values = c("#FF7F0EFF", "#1F77B4FF")) +
  labs(title = "Distribution of public and private schools in the year 2018",
       y = "",
       x = "Percentage of schools",
       fill = "")
```

- The graph assists us in understanding the distribution of public and private schools in few countries based on the datasets. Taking a closer look at the above plot, we can infer that most countries have more public schools than private schools. Interestingly, Spain had a nearly equal mix of public and private schools in the year 2018.

- Similarly, we may derive additional intriguing patterns and analysis by considering the other variables in the school dataset.

```{r, echo=FALSE}
student_data_2018 <- load_student("2018")
data(school)

data(countrycode)

school_student_2018 <- left_join(
  student_data_2018, 
  school,
  by = c("school_id", "country", "year"))

school_student_2018 |>
  dplyr::filter(country %in% selected_countries) |> 
  group_by(country) |> 
  summarise(avg_fund_gov = mean(fund_gov, na.rm = TRUE)) |> 
  arrange(avg_fund_gov) |> 
  mutate(country = fct_reorder(country, avg_fund_gov)) |> 
  left_join(countrycode, by = "country") |> 
  mutate(country_name = fct_reorder(country_name, avg_fund_gov)) |> 
  ggplot(aes(x=country_name, y=avg_fund_gov)) +
  geom_segment(aes(xend=country_name, yend=0)) +
  geom_point(size=4, color="orange") +
  coord_flip() +
  theme_bw() +
  labs(x = "", 
       y = "Average percentage of government funding", 
       title = "Funding for schools in the year 2018 from government")
```

- The above figure shows the average percentage of overall financing in various schools for a random sample of countries. We conclude that countries such as Finland and the United States received the most funding from their governments, whilst Qatar received the least funding.

- In addition, to perform  a detail analysis on the school and entire student data it can be downloaded for the desired years using the `load_student` function available in this package. 

- Similarly, you may import student data for any chosen year and experiment with PISA scores growth or additional analysis of these datasets with their other elements that assist contributor comprehend the data. Refer to our articles [here](https://kevinwang09.github.io/learningtower/articles/exploring_time.html) for additional interesting analyses and plots.