36-315-Lab4

class: inverse, center, middle

# 36-315: Statistical Graphics and Visualization
## Lab 4

Meghan Hall <br> Department of Statistics & Data Science <br> Carnegie Mellon University <br> June 3, 2021

---
layout: true
<div class="my-footer"><span>cmu-36315.netlify.app</span></div>
---

# Today

<br>

.large[Creating line graphs]
<br>
.medium[pivoting data]
<br>
.medium[labeling with `ggrepel`]
<br>
.medium[using functions from `lubridate`]
<br>
<br>
.large[Instructions]
<br>
.medium[HW: describe the graphs]
<br>
.medium[HW/labs: add plot title and relevant axis titles (include units!)]

---

# `pivot_longer`

**need to pivot when your data is at a different observation level than necessary for analysis/visualization**<br>
data here is by track, we need observations per week

<table class="table" style="font-size: 16px; width: auto !important; margin-left: auto; margin-right: auto;">
 <thead>
  <tr>
   <th style="text-align:left;font-weight: bold;color: white !important;background-color: #bb0000 !important;"> artist </th>
   <th style="text-align:left;font-weight: bold;color: white !important;background-color: #bb0000 !important;"> track </th>
   <th style="text-align:right;font-weight: bold;color: white !important;background-color: #bb0000 !important;"> wk1 </th>
   <th style="text-align:right;font-weight: bold;color: white !important;background-color: #bb0000 !important;"> wk2 </th>
   <th style="text-align:right;font-weight: bold;color: white !important;background-color: #bb0000 !important;"> wk3 </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;"> Dixie Chicks, The </td>
   <td style="text-align:left;"> Cold Day In July </td>
   <td style="text-align:right;"> 80 </td>
   <td style="text-align:right;"> 79 </td>
   <td style="text-align:right;"> 76 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Dixie Chicks, The </td>
   <td style="text-align:left;"> Cowboy Take Me Away </td>
   <td style="text-align:right;"> 79 </td>
   <td style="text-align:right;"> 72 </td>
   <td style="text-align:right;"> 70 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Dixie Chicks, The </td>
   <td style="text-align:left;"> Goodbye Earl </td>
   <td style="text-align:right;"> 40 </td>
   <td style="text-align:right;"> 29 </td>
   <td style="text-align:right;"> 24 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Dixie Chicks, The </td>
   <td style="text-align:left;"> Without You </td>
   <td style="text-align:right;"> 80 </td>
   <td style="text-align:right;"> 70 </td>
   <td style="text-align:right;"> 63 </td>
  </tr>
</tbody>
</table>

---

# `pivot_longer`

**need to pivot when your data is at a different observation level than necessary for analysis/visualization**<br>
data here is by track, we need observations per week

```r
billboard %>% 
  filter(artist == "Dixie Chicks, The") %>% 
  select(artist, track, wk1:wk3) %>% 
* pivot_longer(wk1:wk3, names_to = "week", values_to = "ranking")
```

<table class="table" style="font-size: 16px; width: auto !important; margin-left: auto; margin-right: auto;">
 <thead>
  <tr>
   <th style="text-align:left;font-weight: bold;color: white !important;background-color: #bb0000 !important;"> artist </th>
   <th style="text-align:left;font-weight: bold;color: white !important;background-color: #bb0000 !important;"> track </th>
   <th style="text-align:left;font-weight: bold;color: white !important;background-color: #bb0000 !important;"> week </th>
   <th style="text-align:right;font-weight: bold;color: white !important;background-color: #bb0000 !important;"> ranking </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;"> Dixie Chicks, The </td>
   <td style="text-align:left;"> Cold Day In July </td>
   <td style="text-align:left;"> wk1 </td>
   <td style="text-align:right;"> 80 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Dixie Chicks, The </td>
   <td style="text-align:left;"> Cold Day In July </td>
   <td style="text-align:left;"> wk2 </td>
   <td style="text-align:right;"> 79 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Dixie Chicks, The </td>
   <td style="text-align:left;"> Cold Day In July </td>
   <td style="text-align:left;"> wk3 </td>
   <td style="text-align:right;"> 76 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Dixie Chicks, The </td>
   <td style="text-align:left;"> Cowboy Take Me Away </td>
   <td style="text-align:left;"> wk1 </td>
   <td style="text-align:right;"> 79 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Dixie Chicks, The </td>
   <td style="text-align:left;"> Cowboy Take Me Away </td>
   <td style="text-align:left;"> wk2 </td>
   <td style="text-align:right;"> 72 </td>
  </tr>
  <tr>
   <td style="text-align:left;"> Dixie Chicks, The </td>
   <td style="text-align:left;"> Cowboy Take Me Away </td>
   <td style="text-align:left;"> wk3 </td>
   <td style="text-align:right;"> 70 </td>
  </tr>
</tbody>
</table>

---

# Options for relabeling

**`fct_recode` is useful when you have a factor and ordering is important**

```r
# from Wednesday's lecture
mutate(qround = fct_recode(qround, "Qualifying 1" = "q1",
                           "Qualifying 2" = "q2", 
                           "Qualifying 3" = "q3")
```

**`case_when` is generally easier when you have a standard categorical variable**

```r
# example from mpg data set
mutate(drv = case_when(drv == "f" ~ "front-wheel drive",
                       drv == "r" ~ "rear-wheel drive",
                       drv == "4" ~ "four-wheel drive"))
```

---

# Dealing with dates

**functions from `lubridate`**

```r
# from Wednesday's lecture
mutate(time_format = ms(time),
       seconds = seconds(time_format))

# from today's lab
mutate(date = ym(paste(year, month)))
```

**special scale when axis is a date**

```r
scale_x_date(date_breaks = "1 year",
             date_labels = "%m-%Y")
```

---

# Upcoming

<br>

.large[Lab assignment due 11:30am EDT Friday!]
<br>
.medium[Ask questions on Piazza if they don't get answered here]

<br>

.large[Lecture 6 on Friday June 3]
<br>
.medium[Scatter plots]

<br>

.large[Homework 2 due 11:30am EDT Tuesday]
<br>
.medium[Posted now]