Week 8: Strings, dates and times, joining and binding

class: center, middle, inverse, title-slide

# Week 8: Strings, dates and times, joining and binding
## PUBPOL 750 Data Analysis for Public Policy I
### Justin Savoie
### MPP-DS McMaster
### 2022-07-08

---

class: inverse, center, middle

# Strings, dates and times, joining and binding

---

# Data wrangling ?

- We've discussed select(), filter(), group(), mutate() etc.
- A lot of data wrangling is about fixing messy strings, dates; joining data from different sources (from different datasets).

---

# Plan for today

- Joining and binding
- Dealing with character strings
- Dealing with dates
- Two examples + two questions I had on Project 1

---

# Binding (columsn or rows) means appending new lines or new columns

## Binding rows with `bind_rows()`

```r
df1 <- tibble(age=c(25,29,22),name=c("Justin","Ian","Rosaline"))
df2 <- tibble(age=c(55,30,12),name=c("Hillary","Anna","Luis"))
(df <- bind_rows(df1,df2))
```

```
## # A tibble: 6 × 2
##     age name    
##   <dbl> <chr>   
## 1    25 Justin  
## 2    29 Ian     
## 3    22 Rosaline
## 4    55 Hillary 
## 5    30 Anna    
## 6    12 Luis
```

---

```r
df1 <- read_csv("mydata1.csv")
df2 <- read_csv("mydata2.csv")
df <- bind_rows(df1,df2)
```

- Sometimes, you can use a loop to read in numerous csvs. We'll cover this next week. Think of data for 125 riding.
- Need to make sure columns names are the same.

---

## Binding columns with `bind_cols()`

- `bind_cols()` is often used if you want to stack variables where the individuals are the same

```r
df1 <- tibble(age=c(25,29,22),name=c("Justin","Ian","Rosaline"))
df2 <- tibble(income=c(55,30,12),name=c("Justin","Ian","Rosaline"))
(df <- bind_cols(df1,df2 %>% select(-name)))
```

```
## # A tibble: 3 × 3
##     age name     income
##   <dbl> <chr>     <dbl>
## 1    25 Justin       55
## 2    29 Ian          30
## 3    22 Rosaline     12
```

```r
(df <- bind_cols(df1,df2))
```

```
## New names:
## • `name` -> `name...2`
## • `name` -> `name...4`
```

```
## # A tibble: 3 × 4
##     age name...2 income name...4
##   <dbl> <chr>     <dbl> <chr>   
## 1    25 Justin       55 Justin  
## 2    29 Ian          30 Ian     
## 3    22 Rosaline     12 Rosaline
```

---

```r
(df <- bind_rows(df1,df2 %>% select(-name)))
```

```
## # A tibble: 6 × 3
##     age name     income
##   <dbl> <chr>     <dbl>
## 1    25 Justin       NA
## 2    29 Ian          NA
## 3    22 Rosaline     NA
## 4    NA <NA>         55
## 5    NA <NA>         30
## 6    NA <NA>         12
```

```r
bind_rows(tibble(x=c(1,2),y=c(2,3)),
          tibble(y=c(9,8),x=c(10,11)))
```

```
## # A tibble: 4 × 2
##       x     y
##   <dbl> <dbl>
## 1     1     2
## 2     2     3
## 3    10     9
## 4    11     8
```

---

# Joins

Instead of "stacking" like binds, joins "joins" on one or more columns. If you have riding names in two different datasets, you can join them.

- `inner_join()`: includes all rows in x and y.

- `left_join()`: includes all rows in x.

- `right_join()`: includes all rows in y.

- `full_join()`: includes all rows in x or y.

---

```r
left_join(df1,df2)
```