Data used in this assignment was found at https://data.oecd.org/healthres/pharmaceutical-spending.htm#indicator-chart. This data was verified as being open and available to public use by referring to the terms and conditions section of the website. Such data was sourced from the above website and was not modified, beyond sorting and changing some labels to make them more visually appealing to the viewer. Citation: OECD (2020), Pharmaceutical spending (indicator). doi: 10.1787/998febf6-en (Accessed on 28 November 2020)
library(tidyverse)
library(dplyr)

datainput <- read_csv("DP_LIVE_29112020020227678.csv", )

datainput1 <- select(datainput, 'LOCATION', 'TIME', 'Value')

datainput3 <- datainput1[datainput1$'LOCATION' == "BEL"|datainput1$'LOCATION' == "ITA"|datainput1$'LOCATION' == "GRC",] 


datainput2 <- datainput3 %>% rename(Percent = Value, Year = TIME, Country = LOCATION)


datainput2$Country <- ifelse(datainput2$Country %in% ("BEL"), "Belgium", datainput2$Country)
datainput2$Country <- ifelse(datainput2$Country %in% ("ITA"), "Italy", datainput2$Country)
datainput2$Country <- ifelse(datainput2$Country %in% ("GRC"), "Greece", datainput2$Country)

datainput6 <- datainput2 %>%
  mutate(Year = as.integer(Year))

ggplot(data = datainput6, aes(x = Year, y = Percent, color = Country)) +
  geom_line()+ geom_point()+
ggtitle("% of Medical Expenses Going Towards Pharmaceuticals")

The following shows how graphs can potentially be misleading to someone quickly viewing them. In the first graph, you can see it looks like a minor percentage change from year to year, while the second graph shows what looks to be a significant decline, even though its only about a 2% decline. Same data, but completely different views.
greece <- datainput2[datainput2$'Country' == "Greece",] 

ggplot(data = greece, aes(x = Year, y = Percent)) +
  geom_line()+ geom_point()+
  scale_y_continuous(expand = c(0, 0), limits = c(0, NA))+
ggtitle("% of Medical Expenses Going Towards Pharmaceuticals in Greece")

ggplot(data = greece, aes(x = Year, y = Percent)) +
  geom_line()+ geom_point()+
ggtitle("Potentially Misleading % of Medical Expenses Going Towards Pharmaceuticals in Greece")

datatable1 <- datainput2 %>% rename("Pharm %"  = Percent)
datatable <- datatable1 %>%
  filter(Year == 2016| Year == 2017|Year ==2018)
library(knitr)
kable(datatable, align = "lcc", caption = '__Pharmaceutical spend as a percentage of total healthcare costs for 2016-2018__')
Pharmaceutical spend as a percentage of total healthcare costs for 2016-2018
Country Year Pharm %
Belgium 2016 14.329
Belgium 2017 14.155
Belgium 2018 14.584
Greece 2016 26.700
Greece 2017 27.597
Greece 2018 26.238
Italy 2016 17.647
Italy 2017 17.609
Italy 2018 17.904

What I found most interesting is that the countries in Europe that I selected do not have any major correlation to each other’s spending behaviors. I had a lot of issues with this project. I initially had another data set on education levels in Europe, but could not figure out how to properly graph the data. My assumption was that the percentage column was a chr factor, and I could not figure out how to convert that column into a number or integer. I attempted various renditions of as.integer to no aval. Additionally, on this data, I had a similar problem with the year data being in number form and while I was successful in converting year to an integer the graph still added a decimal place.