---
title: "Stat. 651: ggplot2"
author: "Prof. Eric A. Suess"
format:
  html:
    self-contained: true
---

# ggplot2 examples

```{r message=FALSE}
library(tidyverse)
library(mdsr)
library(mosaicData)
```

# CIACounties

Make the base plot **g** and then add different layers on to it.

```{r}
head(CIACountries)

# base plot g

g <- CIACountries |>  ggplot(aes(y = gdp, x = educ)) 

g + geom_point()

g + geom_point(size = 3)

g + geom_point(aes(color = net_users), size = 3)

# no geom_point used for the next picture

g + geom_text( aes(label = country, color = net_users), size = 3 )

g + geom_point( aes(color = net_users, size = roadways) )

```

Change the scales

```{r}
g + geom_point(aes(color = net_users, size = roadways)) +
  coord_trans( y = "log10")

g + 
  geom_point(aes(color = net_users, size = roadways)) + 
  scale_y_continuous(
    name = "Gross Domestic Product", 
    trans = "log10",
    labels = scales::comma
  )

```


# Faceting

```{r}
g + geom_point(alpha = 0.9, aes(size = roadways)) +
  coord_trans(y = "log10") +
  facet_wrap( ~ net_users, nrow = 1) +
  theme(legend.position = "top")

g + geom_point(alpha = 0.9, aes(size = roadways)) +
  coord_trans(y = "log10") +
  scale_y_continuous(name = "Gross Domestic Product", trans = "log10") +
  facet_wrap( ~ net_users, nrow = 1) +
  theme(legend.position = "top")
```


# Export the data and try in Tableau

```{r}
getwd()

write_csv(CIACountries, "CIACountries.csv")
```


# MedicareCharges

Check out the [MEPS](https://meps.ahrq.gov/mepsweb/) website for more real data.

```{r}
# head(MedicareCharges) # This now causes an error, remove the grouping.

? MedicareCharges

ChargesNJ <- MedicareCharges |> 
  filter(stateProvider == "NJ")
```

```{r}
p <- ggplot(
  data = ChargesNJ, 
  aes(x = reorder(drg, mean_charge), y = mean_charge)
) +
  geom_col(fill = "gray") +
  ylab("Statewide Average Charges ($)") + 
  xlab("Medical Procedure (DRG)") +
  theme(axis.text.x = element_text(angle = 90, hjust = 1, size = rel(0.5)))
p

```

Now add the overall data to the plot to compare with NJ.

```{r}
p <- p + geom_point(data = MedicareCharges, size = 1, alpha = 0.3) 
p
```

# SAT

Here is the link to the [College Board SAT](https://research.collegeboard.org/programs/sat/data) website.

```{r}
g <- SAT_2010 |> ggplot(aes(x = math))

g + geom_histogram()

g + geom_histogram(binwidth = 10) + labs(x = "Average math SAT score")

g + geom_density(adjust = 0.3)
```

```{r}
ggplot(
  data = head(SAT_2010, 10), 
  aes(x = reorder(state, math), y = math)
) +
  geom_col() +
  labs(x = "State", y = "Average math SAT score")
```

# HELPPrct

Here is the link to the [NSDUH](https://www.samhsa.gov/data/data-we-collect/nsduh-national-survey-drug-use-and-health) website.

```{r}
ggplot(data = mosaicData::HELPrct, aes(x = homeless)) + 
  geom_bar(aes(fill = substance), position = "fill") +
  scale_fill_brewer(palette = "Spectral") + 
  coord_flip()
```


Scatterplot with tend lines

```{r}
g <- ggplot(
  data = SAT_2010, 
  aes(x = expenditure, y = math)
) + 
  geom_point()
g

g <- g + geom_smooth(method="lm", se = 0) +
  xlab("Average expenditure per student ($100)") +
  ylab("Average score on math SAT")
g
```

Add the trend line within groups representing rate of taking the test.

```{r}
SAT_2010 <- SAT_2010|> 
  mutate(
    SAT_rate = cut(
      sat_pct, 
      breaks = c(0, 30, 60, 100), 
      labels = c("low", "medium", "high")
    )
  )

g <- g %+% SAT_2010
g

g + aes(color = SAT_rate)

g + facet_wrap( ~ SAT_rate)
```

# NHANES

Here is the link to the [NHANES](https://www.cdc.gov/nchs/nhanes/index.htm) website.

```{r}
library(NHANES)

head(NHANES)
```

Take a sample first and then make the plot.  

```{r}
library(NHANES)
ggplot(
  data = slice_sample(NHANES, n = 1000), 
  aes(x = Age, y = Height, color = fct_relevel(Gender, "male"))
) + 
  geom_point() + 
  geom_smooth() + 
  xlab("Age (years)") + 
  ylab("Height (cm)") +
  labs(color = "Gender")
```

Here is an alternative plot using all the data.  This is hexbin plot.

```{r}
NHANES |> ggplot(aes(x = Age, y = Height, color = Gender)) +
  geom_hex() +
  geom_smooth() +
  xlab("Age (years)") +
  ylab("Height (cm)")
```

```{r}
library(mosaic)

head(NHANES)

NHANES2 <- NHANES |> select(AgeDecade, BMI_WHO) 
head(NHANES2)

NHANES2_table <- table(NHANES2)
NHANES2_table

mosaicplot(NHANES2_table, color = TRUE)

```


# Weather 

```{r}
library(macleish)

head(whately_2015)

whately_2015 |> ggplot(aes(x = when, y=temperature)) +
  geom_line(color = "darkgrey") +
  geom_smooth() +
  xlab(NULL) +
  ylab("Tempurature (degrees Fahrenheit)")
```

Here is the link to the [choroplethr](https://arilamstein.com/open-source/) website.

```{r}
library(choroplethr)
library(choroplethrMaps)

	
data(df_pop_state)
state_choropleth(df_pop_state, 
                 title  = "US 2012 State Population Estimates", 
                 legend = "Population")

data(df_pop_county)
county_choropleth(df_pop_county, 
                  title  = "US 2012 County Population Estimates", 
                  legend = "Population")

data(df_pop_country)
country_choropleth(df_pop_country, "2012 World Bank Populate Estimates")

```


Review Table 3.3 on page 47 for the different kinds of plots that can be made for different kinds of x, y variables.


Continue with the Extended example: Historical baby names on page 48.