line plot.Rmd

---
title: "Line plots"
author: "Grace Kim"
date: "12/8/2019"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
library(tidyverse)
library(readxl)
library(janitor)
library(scales)

```

```{r}
#############  Line plot data   ##############

# 2015, 2005, 1995, 1985 Religion data for Korea

x2015 <- read_xlsx("Korea/raw-data/2015_Religions_Korea.xlsx") %>% 
  clean_names()
x2005 <- read_xlsx("Korea/raw-data/2005_Religions_Korea.xlsx") %>% 
  clean_names()
x1995 <- read_xlsx("Korea/raw-data/1995_Religions_Korea.xlsx") %>% 
  clean_names()
x1985 <- read_xlsx("Korea/raw-data/1985_Religions_Korea.xlsx") %>% 
  clean_names()

# Fixes merged column problem from excel for the dataset

x2015 <- x2015 %>% fill(gender) %>% fill(korean_region)
x2005 <- x2005 %>% fill(gender) %>% fill(korean_region)
x1995 <- x1995 %>% fill(gender) %>% fill(korean_region)
x1985 <- x1985 %>% fill(gender) %>% fill(korean_region)


# Data cleaning, need to only get the regions of all the total populations of Korea of
# 2015, only the gender and age totals, and we only selected the specific religions
# we wanted to view on the line plot, christianity and buddhism and religious totals
# and non religious totals

totals2015 <- x2015 %>% 
  filter(korean_region == "All of Korea") %>% 
  filter(gender == "Korea - Total") %>% 
  filter(age == "Total") %>% 
  select(total,religious_total,christianity_protestant,
         christianity_catholic,buddhism,no_religion) %>%
  mutate(year = "2015")
  
# Data cleaning, need to only get the regions of all the total populations of Korea of
# 2005, only the gender and age totals, and we only selected the specific religions
# we wanted to view on the line plot, christianity and buddhism and religious totals
# and non religious totals

totals2005 <- x2005 %>% 
  filter(korean_region == "All of Korea") %>% 
  filter(gender == "Korea - Total") %>% 
  filter(age == "Total") %>% 
  select(total,religious_total,christianity_protestant,
         christianity_catholic,buddhism,no_religion) %>%
  mutate(year = "2005")

# Data cleaning, need to only get the regions of all the total populations of Korea of
# 1995, only the gender and age totals, and we only selected the specific religions
# we wanted to view on the line plot, christianity and buddhism and religious totals
# and non religious totals

totals1995 <- x1995 %>% 
  filter(korean_region == "All of Korea") %>% 
  filter(gender == "Korea - Total") %>% 
  filter(age == "Total") %>% 
  select(total,religious_total,christianity_protestant,
         christianity_catholic,buddhism,no_religion) %>%
  mutate(year = "1995")

# Data cleaning, need to only get the regions of all the total populations of Korea of
# 1985, only the gender and age totals, and we only selected the specific religions
# we wanted to view on the line plot, christianity and buddhism and religious totals
# and non religious totals

totals1985 <- x1985 %>% 
  filter(korean_region == "All of Korea") %>% 
  filter(gender == "Total") %>% 
  filter(age == "Total") %>% 
  select(total,religious_total,christianity_protestant,
         christianity_catholic,buddhism,no_religion) %>%
  mutate(year = "1985")

# Combined the four different years datas so that we could have it all in one set

combined2 <- rbind(totals1985, totals1995, totals2005, totals2015) %>% 
  mutate(freq = 6)

# Converted to a data frame, duplicated each row 6 times so that we could make a column
# that contains all of the various religion data for grouped line bar plots

combined <- as.data.frame(combined2[rep(row.names(combined2),
                                       combined2$freq), 1:7])

# Made several rows that could contain numbers for the four different religous population
# numbers, and the total population numbers and non religious population numbers
# had a identifier row called religion and numbers in population
# needed a case_when function because we had so many different possibilities

combined <- combined %>% 
  mutate(religion = case_when(row_number() %% 6 == 1 ~ "Total Population",
                              row_number() %% 6 == 2 ~ "All Religions",
                              row_number() %% 6 == 3 ~ "Christianity Protestant",
                              row_number() %% 6 == 4 ~ "Christianity Catholic",
                              row_number() %% 6 == 5 ~ "Buddhism",
                              row_number() %% 6 == 0 ~ "No Religion",
                              TRUE ~ "no")) %>% 
  mutate(population = case_when(row_number() %% 6 == 1 ~ total,
                                row_number() %% 6 == 2 ~ religious_total,
                                row_number() %% 6 == 3 ~ christianity_protestant,
                                row_number() %% 6 == 4 ~ christianity_catholic,
                                row_number() %% 6 == 5 ~ buddhism,
                                row_number() %% 6 == 0 ~ no_religion,
                                TRUE ~ "no"))

# line plot for the year by data change of the various line plots

combined %>% 
  ggplot(aes(x=year, 
             y=as.numeric(population), 
             group=religion, 
             color=religion, size = 2)) +
  geom_line() + 
  scale_y_continuous(limits = c(0,50000000), labels = comma)+
  labs(x = "Year", y = "Population")
```