Graphic(s) with the PSID

library(tidyverse)
library(readxl)
library(broom)
library(ggthemes)

# Load the dataset
psid <- read_xlsx("C:/Dropbox/DRPH/P9489 Application of Epi Research Methods II/PSID/PSID_2.17.2020.xlsx")

# create table "psidn" for plotting - use pipes
psidn <- psid %>%
  
  # Select Health Status variables from 2013, 2015, and 2017
  transmute(`2013` = ER55244, `2015` = ER62366, `2017` = ER68420)  %>%
  
  # Pivot longer  
  pivot_longer(c(`2013`,`2015`,`2017`), names_to = "year", values_to = "health_status")  %>%

  # Keep health status values (1=Excellent, 2=Very good, 3=Good, 4=Fair, 5=Poor)
  filter(health_status %in%  c(1:5))  %>%
  
  # Summarize frequencies by grouping by year and health status
  group_by(year=factor(year), health_status=factor(health_status, levels = c(1,2,3,4,5), labels = c("Excellent", "Very good", "Good", "Fair", "Poor"))) %>%
  summarise(count = n())

  # if you want to check that mappings are correct use the following statement:
  # table(psidn$year, psidn$health_status)

# Alternative 1 - use Tufte Theme for grouped bar chart (frequencies)
ggplot(psidn, aes(year, count, fill = health_status)) + 
  geom_bar(stat="identity", position = "dodge") + 
  theme_tufte() + 
  theme(legend.title = element_blank()) +
  labs(title="Self Reported Health Status: 2013-2017",
            caption="Panel Study of Income Dynamics",
            x="Year",
            y="Frequency")

# Alternative 2 - use Tufte Theme for percent stacked bar chart (population percentages)
  ggplot(psidn, aes(year, count, fill = health_status )) + 
  geom_bar(stat="identity", position = "fill") + 
  theme_tufte() + 
  theme(legend.title = element_blank()) +
  labs(title="Self Reported Health Status: 2013-2017",
            caption="Panel Study of Income Dynamics",
            x="Year",
            y="Population Proportions")

 
# Alternative 3 - use standard colors for connected points for each group (showing tendency)   
  ggplot(psidn, aes(year, count, group = health_status, color=health_status)) + 
  geom_point() + geom_line() + 
  theme(legend.title = element_blank()) +
  labs(title="Self Reported Health Status: 2013-2017",
            caption="Panel Study of Income Dynamics",
            x="Year",
            y="Frequency")