---
title: "Unite Examples"
output:
  html_notebook: default
  pdf_document: default
  word_document: default
---

# Some Examples of spread, gather, unite, separate.

See the [R Studio Data Wrangling Cheatsheet](https://www.rstudio.com/wp-content/uploads/2015/02/data-wrangling-cheatsheet.pdf).

See the [R Studio ggplot Cheatsheet](https://www.rstudio.com/wp-content/uploads/2015/03/ggplot2-cheatsheet.pdf).

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


### The examples from the cheatsheet.

```{r}
mydata <- data_frame(
  a = 1:3, 
  b = 4:6
)

mydata
```

Arrange the data.

```{r}
arrange(mydata, desc(a)) 
```

Rename the data.

```{r}
mydata <- rename(mydata, x = a, y = b)

mydata

arrange(mydata, desc(x)) 
```

```{r}
mydata %>% mutate(x.prop = x/sum(x), x.cum.prop = cumsum(x)/sum(x), y.prop = y/sum(y), y.cum.prop = cumsum(y)/sum(y)) %>%
  select(x, x.prop, x.cum.prop, y, y.prop, y.cum.prop)
```

# Example, page 27, Problem 2.2, Ott 3rd Edition

```{r}
imports <- data_frame(
  Year = c(1979:1986),
  Import = c(17518,15491,19898,16663,17061,26171,23650,19650)
)

imports

imports.wide <- spread(imports, key = Year, value = Import)
imports.wide

imports.narrow <- gather(imports.wide, key = "Year.New", value = Import.New)
imports.narrow

imports.narrow <- imports.narrow %>% mutate(Year.New = as.integer(Year.New))
imports.narrow

imports %>% ggplot(aes(x=Year, y=Import)) +
  geom_col()

imports %>% ggplot(aes(x=Year, y=Import)) +
  geom_line()

```

# Example, page 28, Problem 2.4, Ott 3rd Edition

```{r}


GNP.1985 <- data_frame(
  Year = c(1985,1985,1985,1985),
  Quarter = c("I","II","III","IV"),
  GNP = c(3910,3961,4017,4067),
  DPI = c(2505,2532,2503,2533)
)
GNP.1985

GNP.1986 <- data_frame(
  Year = c(1986,1986,1986,1986),
  Quarter = c("I","II","III","IV"),
  GNP = c(4137,4203,4266,4308),
  DPI = c(2536,2555,2579,2589)
)
GNP.1986

GNP <- bind_rows(GNP.1985, GNP.1986)
GNP

GNP.wide <- GNP %>% select(Year, Quarter, GNP) %>%
  spread(key = Quarter, value = GNP ) 
GNP.wide

GNP.narrow <- GNP.wide %>%
  gather(key = Quarter, value = GNP, I,II,III, IV ) %>%
  arrange(Year)
GNP.narrow

GNP %>% ggplot(aes(x= factor(Year), y = GNP, fill = Quarter)) +
  geom_bar(stat = "identity", position = "dodge") +
  labs(x = "Year")
```

# Example, page 30, Problem 2.14, Ott 3rd Edition


```{r}


SAT <- data_frame(
  GT = c("Male, Math", "Female, Math", "Male, Verbal", "Female, Verbal"),
  "Year 1967" = c(514,467,463,486),
  "Year 1970" = c(509,465,459,461),
  "Year 1975" = c(495,449,437,431),
  "Year 1980" = c(491,443,428,420),
  "Year 1983" = c(493,445,430,420)
)
SAT

SAT.wide <- SAT %>% rename("1967" = "Year 1967", "1970" = "Year 1970", "1975" = "Year 1975", "1980" = "Year 1980", "1983" = "Year 1983")
SAT.wide

SAT.narrow <- SAT.wide %>% gather( key = Year, value = Score, "1967", "1970", "1975", "1980", "1983" )
SAT.narrow

SAT.narrow2 <- SAT.narrow %>% separate(GT, c("Gender", "Type"))
SAT.narrow2

SAT.narrow3 <- SAT.narrow2 %>% unite(GT, c("Gender", "Type"), sep="_")
SAT.narrow3

SAT.narrow %>% ggplot(aes(x=Year, y=Score, color=GT) ) +
  geom_line(aes(group = GT)) 

SAT.narrow %>% ggplot( aes( x=factor(Year), y=Score, fill=factor(GT) ) ) +
  geom_col()

SAT.narrow %>% ggplot( aes( x=factor(Year), y=Score, fill=factor(GT) ) ) +
  geom_col(position = "dodge")
```