05 — Show the Right Numbers

# .kjh-yellow[Show the]<br /> .kjh-lblue[Right Numbers] <br /> .kjh-yellow[with `dplyr`]

**Week 05**

]

---

---

# Load our libraries

.SMALL[

```r
library(here)      # manage file paths
library(socviz)    # data and some useful functions
library(tidyverse) # your friend and mine
```
]

---

# Tidyverse components, again

.pull-left.w45[

- .kjh-green[**`library`**]`(tidyverse)`
- `Loading tidyverse: ggplot2`
- `Loading tidyverse: tibble`
- `Loading tidyverse: tidyr`
- `Loading tidyverse: readr`
- `Loading tidyverse: purrr`
- `Loading tidyverse: dplyr`
]

]

---

# Other tidyverse components

--
.left.bottom[.footnote[Not all of these are attached when we do `library(tidyverse)`]]

---
layout: false
class: main-title main-title-inv center middle

---

---

---

---

---

# .huge[.kjh-yellow[Feeding data]<br /> .kjh-lblue[to `ggplot`]]

---

## .middle.huge.squish4[.kjh-orange[Transform and summarize first.]<br />.kjh-lblue[Then send your clean tables to ggplot.]]

---

## .huge.right.bottom.squish4.kjh-yellow[Crosstabulation<br />.kjh-lblue[and beyond]]

---

---

# U.S. General Social Survey data: .kjh-pink[`gss_sm`]

```r
gss_sm  
```

```
## # A tibble: 2,867 × 32
##     year    id ballot   age childs sibs  degree race  sex   region incom…¹ relig
##    <dbl> <dbl> <labe> <dbl>  <dbl> <lab> <fct>  <fct> <fct> <fct>  <fct>   <fct>
##  1  2016     1 1         47      3 2     Bache… White Male  New E… $17000… None 
##  2  2016     2 2         61      0 3     High … White Male  New E… $50000… None 
##  3  2016     3 3         72      2 3     Bache… White Male  New E… $75000… Cath…
##  4  2016     4 1         43      4 3     High … White Fema… New E… $17000… Cath…
##  5  2016     5 3         55      2 2     Gradu… White Fema… New E… $17000… None 
##  6  2016     6 2         53      2 2     Junio… White Fema… New E… $60000… None 
##  7  2016     7 1         50      2 2     High … White Male  New E… $17000… None 
##  8  2016     8 3         23      3 6     High … Other Fema… Middl… $30000… Cath…
##  9  2016     9 1         45      3 5     High … Black Male  Middl… $60000… Prot…
## 10  2016    10 3         71      4 1     Junio… White Male  Middl… $60000… None 
## # … with 2,857 more rows, 20 more variables: marital <fct>, padeg <fct>,
## #   madeg <fct>, partyid <fct>, polviews <fct>, happy <fct>, partners <fct>,
## #   grass <fct>, zodiac <fct>, pres12 <labelled>, wtssall <dbl>,
## #   income_rc <fct>, agegrp <fct>, ageq <fct>, siblings <fct>, kids <fct>,
## #   religion <fct>, bigregion <fct>, partners_rc <fct>, obama <dbl>, and
## #   abbreviated variable name ¹income16
```

- We often want summary tables or graphs of data like this.

---

# Two-way tables: Row percents

|bigregion |Protestant |Catholic |Jewish |None |Other |Total |
|:---------|:----------|:--------|:------|:----|:-----|:-----|
|Northeast |32.4       |33.3     |5.5    |23.0 |5.7   |100.0 |
|Midwest   |47.1       |24.9     |0.4    |22.8 |4.8   |100.0 |
|South     |62.4       |15.4     |1.1    |16.3 |4.8   |100.0 |
|West      |37.7       |24.6     |1.6    |28.5 |7.6   |100.0 |

---

# Two-way tables: Column percents

|bigregion |Protestant |Catholic |Jewish |None  |Other |
|:---------|:----------|:--------|:------|:-----|:-----|
|Northeast |11.5       |25.0     |52.9   |18.1  |17.6  |
|Midwest   |23.7       |26.5     |5.9    |25.4  |20.8  |
|South     |47.4       |24.7     |21.6   |27.5  |31.4  |
|West      |17.4       |23.9     |19.6   |29.1  |30.2  |
|Total     |100.0      |100.0    |100.0  |100.0 |100.0 |

---

# Two-way tables: Full marginals

|bigregion |Protestant |Catholic |Jewish |None |Other |
|:---------|:----------|:--------|:------|:----|:-----|
|Northeast |5.5        |5.7      |0.9    |3.9  |1.0   |
|Midwest   |11.4       |6.0      |0.1    |5.5  |1.2   |
|South     |22.8       |5.6      |0.4    |6.0  |1.8   |
|West      |8.4        |5.4      |0.4    |6.3  |1.7   |

---

# .kjh-yellow[dplyr] lets you work with tibbles

.pull-left-wide[
- Remember, tibbles are tables of data where the columns can be of different types, such as numeric, logical, character, factor, etc.]

--
.pull-left-wide[

- We'll use the pipe operator, .kjh-pink[**`|>`**], to chain together sequences of actions on our tables.

]
---

# .huge.middle.squish4[`dplyr` draws on the logic and language of  .kjh-green[database queries], where the focus is on manipulating tables]

---

---

# Some .kjh-orange[actions] to take on a single table

.pull-left.w80[
- .kjh-orange[**Group**] the data at the level we want, such as “_Religion by Region_” or _“Children by School_”.

- .kjh-orange[**Subset**] either the rows or columns of or table.

- .kjh-orange[**Mutate**] the data. That is, change something at the _current_ level of grouping.  Mutating adds new columns to the table, or changes the content of an existing column. It never changes the number of rows.

- .kjh-orange[**Summarize**] or aggregate the data. That is, make something new at a _higher_ level of grouping. E.g., calculate means or counts by some grouping variable. This will generally result in a smaller, _summary_ table.
]

---

# Each .kjh-orange[action] is implemented by a .kjh-green[function]

--
.pull-left-wide[
- **Group** using  .kjh-green[**`group_by()`**].
]

.pull-left-wide[
- **Subset** has one action for rows and one for columns. We .kjh-green[**`filter()`**] rows and .kjh-green[**`select()`**] columns. 
]
--
.pull-left-wide[
- **Mutate** tables (i.e. add new columns, or re-make existing ones) using .kjh-green[**`mutate()`**].
]
--

.pull-left-wide[
- **Summarize** tables (i.e. perform aggregating calculations) using .kjh-green[**`summarize()`**].
]

---

# .huge[.kjh-lblue[Example:]<br/>.kjh-yellow[The GSS]]

---

# U.S. General Social Survey data: .kjh-pink[`gss_sm`]

```r
gss_sm
```

Notice again how the tibble already tells us a lot.

---

# Summarizing a Table

- Here's what we're going to do:

---

# Summarizing a Table

- We're just taking a look at the relevant columns here. We don't need to narrow it like this to do our summary, though.

```r
gss_sm |> 
  select(id, bigregion, religion) 
```

```
## # A tibble: 2,867 × 3
##       id bigregion religion  
##    <dbl> <fct>     <fct>     
##  1     1 Northeast None      
##  2     2 Northeast None      
##  3     3 Northeast Catholic  
##  4     4 Northeast Catholic  
##  5     5 Northeast None      
##  6     6 Northeast None      
##  7     7 Northeast None      
##  8     8 Northeast Catholic  
##  9     9 Northeast Protestant
## 10    10 Northeast None      
## # … with 2,857 more rows
```

---
count: false
 
# Count up by one column or variable
.panel1-reveal-onetablevar-auto[

```r
*gss_sm
```
]
 
.panel2-reveal-onetablevar-auto[

---
count: false
 
# Count up by one column or variable
.panel1-reveal-onetablevar-auto[

```r
gss_sm %>%
* group_by(bigregion)
```
]
 
.panel2-reveal-onetablevar-auto[

```
## # A tibble: 2,867 × 32
## # Groups:   bigregion [4]
##     year    id ballot   age childs sibs  degree race  sex   region incom…¹ relig
##    <dbl> <dbl> <labe> <dbl>  <dbl> <lab> <fct>  <fct> <fct> <fct>  <fct>   <fct>
##  1  2016     1 1         47      3 2     Bache… White Male  New E… $17000… None 
##  2  2016     2 2         61      0 3     High … White Male  New E… $50000… None 
##  3  2016     3 3         72      2 3     Bache… White Male  New E… $75000… Cath…
##  4  2016     4 1         43      4 3     High … White Fema… New E… $17000… Cath…
##  5  2016     5 3         55      2 2     Gradu… White Fema… New E… $17000… None 
##  6  2016     6 2         53      2 2     Junio… White Fema… New E… $60000… None 
##  7  2016     7 1         50      2 2     High … White Male  New E… $17000… None 
##  8  2016     8 3         23      3 6     High … Other Fema… Middl… $30000… Cath…
##  9  2016     9 1         45      3 5     High … Black Male  Middl… $60000… Prot…
## 10  2016    10 3         71      4 1     Junio… White Male  Middl… $60000… None 
## # … with 2,857 more rows, 20 more variables: marital <fct>, padeg <fct>,
## #   madeg <fct>, partyid <fct>, polviews <fct>, happy <fct>, partners <fct>,
## #   grass <fct>, zodiac <fct>, pres12 <labelled>, wtssall <dbl>,
## #   income_rc <fct>, agegrp <fct>, ageq <fct>, siblings <fct>, kids <fct>,
## #   religion <fct>, bigregion <fct>, partners_rc <fct>, obama <dbl>, and
## #   abbreviated variable name ¹income16
```
]

---
count: false
 
# Count up by one column or variable
.panel1-reveal-onetablevar-auto[

```r
gss_sm %>%
  group_by(bigregion) %>%
* summarize(total = n())
```
]
 
.panel2-reveal-onetablevar-auto[

```
## # A tibble: 4 × 2
##   bigregion total
##   <fct>     <int>
## 1 Northeast   488
## 2 Midwest     695
## 3 South      1052
## 4 West        632
```
]

Grouping changes the _logical_ structure of the tibble. It tells you what subsets of rows the next mutate or summary operation will be carried out within.

---
count: false
 
# Summarize by region and religion
.panel1-reveal-pipe1-auto[

```r
*gss_sm
```
]
 
.panel2-reveal-pipe1-auto[

---
count: false
 
# Summarize by region and religion
.panel1-reveal-pipe1-auto[

```r
gss_sm %>%
* group_by(bigregion, religion)
```
]
 
.panel2-reveal-pipe1-auto[

```
## # A tibble: 2,867 × 32
## # Groups:   bigregion, religion [24]
##     year    id ballot   age childs sibs  degree race  sex   region incom…¹ relig
##    <dbl> <dbl> <labe> <dbl>  <dbl> <lab> <fct>  <fct> <fct> <fct>  <fct>   <fct>
##  1  2016     1 1         47      3 2     Bache… White Male  New E… $17000… None 
##  2  2016     2 2         61      0 3     High … White Male  New E… $50000… None 
##  3  2016     3 3         72      2 3     Bache… White Male  New E… $75000… Cath…
##  4  2016     4 1         43      4 3     High … White Fema… New E… $17000… Cath…
##  5  2016     5 3         55      2 2     Gradu… White Fema… New E… $17000… None 
##  6  2016     6 2         53      2 2     Junio… White Fema… New E… $60000… None 
##  7  2016     7 1         50      2 2     High … White Male  New E… $17000… None 
##  8  2016     8 3         23      3 6     High … Other Fema… Middl… $30000… Cath…
##  9  2016     9 1         45      3 5     High … Black Male  Middl… $60000… Prot…
## 10  2016    10 3         71      4 1     Junio… White Male  Middl… $60000… None 
## # … with 2,857 more rows, 20 more variables: marital <fct>, padeg <fct>,
## #   madeg <fct>, partyid <fct>, polviews <fct>, happy <fct>, partners <fct>,
## #   grass <fct>, zodiac <fct>, pres12 <labelled>, wtssall <dbl>,
## #   income_rc <fct>, agegrp <fct>, ageq <fct>, siblings <fct>, kids <fct>,
## #   religion <fct>, bigregion <fct>, partners_rc <fct>, obama <dbl>, and
## #   abbreviated variable name ¹income16
```
]

---
count: false
 
# Summarize by region and religion
.panel1-reveal-pipe1-auto[

```r
gss_sm %>%
  group_by(bigregion, religion) %>%
* summarize(total = n())
```
]
 
.panel2-reveal-pipe1-auto[

```
## # A tibble: 24 × 3
## # Groups:   bigregion [4]
##    bigregion religion   total
##    <fct>     <fct>      <int>
##  1 Northeast Protestant   158
##  2 Northeast Catholic     162
##  3 Northeast Jewish        27
##  4 Northeast None         112
##  5 Northeast Other         28
##  6 Northeast <NA>           1
##  7 Midwest   Protestant   325
##  8 Midwest   Catholic     172
##  9 Midwest   Jewish         3
## 10 Midwest   None         157
## # … with 14 more rows
```
]

---
count: false
 
# Summarize by region and religion
.panel1-reveal-pipe1-auto[

```r
gss_sm %>%
  group_by(bigregion, religion) %>%
  summarize(total = n()) %>%
* mutate(freq = total / sum(total),
*          pct = round((freq*100), 1))
```
]
 
.panel2-reveal-pipe1-auto[

```
## # A tibble: 24 × 5
## # Groups:   bigregion [4]
##    bigregion religion   total    freq   pct
##    <fct>     <fct>      <int>   <dbl> <dbl>
##  1 Northeast Protestant   158 0.324    32.4
##  2 Northeast Catholic     162 0.332    33.2
##  3 Northeast Jewish        27 0.0553    5.5
##  4 Northeast None         112 0.230    23  
##  5 Northeast Other         28 0.0574    5.7
##  6 Northeast <NA>           1 0.00205   0.2
##  7 Midwest   Protestant   325 0.468    46.8
##  8 Midwest   Catholic     172 0.247    24.7
##  9 Midwest   Jewish         3 0.00432   0.4
## 10 Midwest   None         157 0.226    22.6
## # … with 14 more rows
```
]

---
count: false
 
# Summarize by region and religion
.panel1-reveal-pipe1-auto[

```r
gss_sm %>%
  group_by(bigregion, religion) %>%
  summarize(total = n()) %>%
  mutate(freq = total / sum(total),
           pct = round((freq*100), 1))
```
]
 
.panel2-reveal-pipe1-auto[

---

# Pipelines carry assumptions forward

```r
gss_sm |> 
* group_by(bigregion, religion) |>
  summarize(total = n()) |> 
  mutate(freq = total / sum(total),
           pct = round((freq*100), 1))
```

Groups are carried forward till summarized over, or explicitly ungrouped with .kjh-green[`ungroup()`].

Summary calculations are done on the innermost group, which then "disappears". (Notice how .kjh-orange[`religion`] is no longer a group in the output, but .kjh-orange[`bigregion`] is.)

---

# Pipelines carry assumptions forward

```r
gss_sm |> 
  group_by(bigregion, religion) |> 
  summarize(total = n()) |> 
  mutate(freq = total / sum(total),
*        pct = round((freq*100), 1))
```

.kjh-green[**`mutate()`**] is quite clever. See how we can immediately use .kjh-orange[**`freq`**] to calculate .kjh-orange[**`pct`**], even though we are creating them both in the same .kjh-green[**`mutate()`**] expression.

---

# Convenience functions

```r
gss_sm |> 
* group_by(bigregion, religion) |>
* summarize(total = n()) |>
  mutate(freq = total / sum(total),
           pct = round((freq*100), 1)) 
```

We're going to be doing this .kjh-green[**`group_by()`**] ... .kjh-green[**`n()`**] step a lot. Some shorthand for it would be useful.

---

# .kjh-pink[Three options] for counting up rows

- .SMALL.squish3[Do it yourself with .kjh-green[**`n()`**]]

.SMALL[

```r
gss_sm |> 
* group_by(bigregion, religion) |>
* summarize(n = n())
```

```
## # A tibble: 24 × 3
## # Groups:   bigregion [4]
##    bigregion religion       n
##    <fct>     <fct>      <int>
##  1 Northeast Protestant   158
##  2 Northeast Catholic     162
##  3 Northeast Jewish        27
##  4 Northeast None         112
##  5 Northeast Other         28
##  6 Northeast <NA>           1
##  7 Midwest   Protestant   325
##  8 Midwest   Catholic     172
##  9 Midwest   Jewish         3
## 10 Midwest   None         157
## # … with 14 more rows
```
]

- .small.squish3[_The result is **grouped**._]
]

- .SMALL.squish3[Use .kjh-green[**`tally()`**]]

.SMALL[

```r
gss_sm |> 
  group_by(bigregion, religion) |> 
* tally()
```

- .small.squish3[_The result is **grouped**._]
]

- .SMALL.squish3[use .kjh-green[**`count()`**]]

.SMALL[

```r
gss_sm |> 
* count(bigregion, religion)
```

```
## # A tibble: 24 × 3
##    bigregion religion       n
##    <fct>     <fct>      <int>
##  1 Northeast Protestant   158
##  2 Northeast Catholic     162
##  3 Northeast Jewish        27
##  4 Northeast None         112
##  5 Northeast Other         28
##  6 Northeast <NA>           1
##  7 Midwest   Protestant   325
##  8 Midwest   Catholic     172
##  9 Midwest   Jewish         3
## 10 Midwest   None         157
## # … with 14 more rows
```
]

- .small.squish3[_One step; the result is **not grouped**._]
]

---

# Pipelined tables can be quickly checked

```r
## Calculate pct religion within region?
rel_by_region <- gss_sm |> 
  count(bigregion, religion) |> 
  mutate(pct = round((n/sum(n))*100, 1))

rel_by_region
```

```
## # A tibble: 24 × 4
##    bigregion religion       n   pct
##    <fct>     <fct>      <int> <dbl>
##  1 Northeast Protestant   158   5.5
##  2 Northeast Catholic     162   5.7
##  3 Northeast Jewish        27   0.9
##  4 Northeast None         112   3.9
##  5 Northeast Other         28   1  
##  6 Northeast <NA>           1   0  
##  7 Midwest   Protestant   325  11.3
##  8 Midwest   Catholic     172   6  
##  9 Midwest   Jewish         3   0.1
## 10 Midwest   None         157   5.5
## # … with 14 more rows
```

Hm, did I sum over right group?

]

--
.pull-right[

```r
## Each region should sum to ~100
rel_by_region |> 
  group_by(bigregion) |> 
  summarize(total = sum(pct)) 
```

```
## # A tibble: 4 × 2
##   bigregion total
##   <fct>     <dbl>
## 1 Northeast  17  
## 2 Midwest    24.3
## 3 South      36.7
## 4 West       22
```

No! What has gone wrong here?

]

---

# Pipelined tables can be quickly checked

```r
rel_by_region <- gss_sm |> 
* count(bigregion, religion) |>
  mutate(pct = round((n/sum(n))*100, 1)) 
```

.SMALL.squish3[.kjh-green[**`count()`**] returns ungrouped results, so there are no groups carry forward to the .kjh-green[**`mutate()`**] step.]

```r
rel_by_region |> 
  summarize(total = sum(pct))
```

```
## # A tibble: 1 × 1
##   total
##   <dbl>
## 1   100
```

.SMALL.squish3[With .kjh-green[**`count()`**], the .kjh-orange[**`pct`**] values in this case are the marginals for the whole table.]

]

--
.pull-right[

```r
rel_by_region <- gss_sm |> 
* group_by(bigregion, religion) |>
* tally() |>
  mutate(pct = round((n/sum(n))*100, 1)) 
```

```r
# Check
rel_by_region |> 
  group_by(bigregion) |> 
  summarize(total = sum(pct))
```

```
## # A tibble: 4 × 2
##   bigregion total
##   <fct>     <dbl>
## 1 Northeast 100  
## 2 Midwest    99.9
## 3 South     100  
## 4 West      100.
```

.SMALL.squish3[.kjh-green[**`group_by()`**] and .kjh-green[**`tally()`**] both return a grouped result. We get some rounding error because we used .kjh-green[**`round()`**] after summing originally.]
]

---

# Two lessons

## 1: Check your tables!

-     Pipelines feed their content forward, so you need to make sure your results are not incorrect.

- Often, complex tables and graphs can be disturbingly plausible even when wrong.

- So, figure out what the result should be and test it!

- Starting with simple or toy cases can help with this process.

---

# Two lessons

## 2: Inspect your pipes!

- Understand pipelines by running them forward or peeling them back a step at a time.

- This is a _very_ effective way to understand your own and other people's code.

---

# Pass your pipeline on to a .kjh-yellow[table]

```r
gss_sm |> 
  count(bigregion, religion) |> 
* pivot_wider(names_from = bigregion, values_from = n) |>
  kable()  
```

|religion   | Northeast| Midwest| South| West|
|:----------|---------:|-------:|-----:|----:|
|Protestant |       158|     325|   650|  238|
|Catholic   |       162|     172|   160|  155|
|Jewish     |        27|       3|    11|   10|
|None       |       112|     157|   170|  180|
|Other      |        28|      33|    50|   48|
|NA         |         1|       5|    11|    1|
]

---

# Pass your pipeline on to a .kjh-yellow[graph]

.SMALL[

```r
gss_sm |> 
  group_by(bigregion, religion) |> 
  tally() |> 
  mutate(pct = round((n/sum(n))*100, 1)) |> 
  drop_na() |> 
* ggplot(mapping = aes(x = pct, y = reorder(religion, -pct), fill = religion)) +
* geom_col() +
    labs(x = "Percent", y = NULL) +
    guides(fill = "none") + 
    facet_wrap(~ bigregion, nrow = 1)
```

<img src="05-slides_files/figure-html/05-work-with-dplyr-and-geoms-23-1.png" width="1080" style="display: block; margin: auto;" />
]

---

## .middle.huge.squish4[.kjh-orange[Use `dplyr` pipelines to create summary tables.]<br />.kjh-lblue[Then send your clean tables to `ggplot`.]]

---

## .huge.right.bottom.squish4[.kjh-lblue[Facets] .kjh-yellow[are often<br />better than] .kjh-lblue[Guides]]

---

---

# Let's put that table in an object

```r
rel_by_region <- gss_sm |> 
  group_by(bigregion, religion) |> 
  tally() |> 
  mutate(pct = round((n/sum(n))*100, 1)) |> 
  drop_na()

head(rel_by_region)
```

```
## # A tibble: 6 × 4
## # Groups:   bigregion [2]
##   bigregion religion       n   pct
##   <fct>     <fct>      <int> <dbl>
## 1 Northeast Protestant   158  32.4
## 2 Northeast Catholic     162  33.2
## 3 Northeast Jewish        27   5.5
## 4 Northeast None         112  23  
## 5 Northeast Other         28   5.7
## 6 Midwest   Protestant   325  46.8
```

---

# We might write ...

```r
p <- ggplot(data = rel_by_region, 
                mapping = aes(x = bigregion, 
                              y = pct, 
                              fill = religion))
p_out <- p + geom_col(position = "dodge") +
    labs(x = "Region",
         y = "Percent", 
         fill = "Religion") 
```

---

# We might write ...

---

# Is this an effective graph? .kjh-red[Not really!]

---

# Try .kjh-lblue[faceting] instead

```r
p <- ggplot(data = rel_by_region, 
*               mapping = aes(x = pct,
*                             y = reorder(religion, -pct),
                              fill = religion))
p_out_facet <- p + geom_col() +
  guides(fill = "none") + 
  facet_wrap(~ bigregion, nrow = 1) +
  labs(x = "Percent",
       y = NULL) 
```

- Putting categories on the y-axis is a very useful trick.

- Faceting reduces the number of guides the viewer needs to consult.

---

# Try .kjh-lblue[faceting] instead

### .kjh-green[Try putting categories on the y-axis. (And reorder them by x.)]

### .kjh-lblue[Try faceting variables instead of mapping them to color or shape.]

### .kjh-pink[Try to minimize the need for guides and legends.]

---

## .huge.right.bottom.squish4[.kjh-yellow[Two kinds of] .kjh-lblue[facet]]

---

# Facet Children vs Age, by Race

```r
p <-  ggplot(data = gss_sm,
             mapping = aes(x = age, y = childs))

p + geom_point(alpha = 0.2) + 
  geom_smooth() +
  facet_wrap(~ race)
```

---

# We can facet by more than one variable

```r
p <-  ggplot(data = gss_sm,
             mapping = aes(x = age, y = childs))

p + geom_point(alpha = 0.2) + 
  geom_smooth() +
* facet_wrap(~ sex + race)
```

---

# We can arrange .kjh-green[facet_wrap()] quite freely

```r
p <-  ggplot(data = gss_sm,
             mapping = aes(x = age, y = childs))

p + geom_point(alpha = 0.2) + 
  geom_smooth() +
* facet_wrap(~ sex + race, nrow = 1)
```

---

# .kjh-green[facet_grid()] is more like a true crosstab

```r
p + geom_point(alpha = 0.2) + 
  geom_smooth() +
* facet_grid(sex ~ race)
```

---

# Both can be exended to multi-way views

```r
p_out <- p + geom_point(alpha = 0.2) + 
  geom_smooth() +
* facet_grid(bigregion ~ race + sex)
```

---
layout: false

---

# .huge[.kjh-lblue[What we've] .kjh-yellow[built-up]]

---

# Core Grammar

---

# Grouped data; faceting

- Along with a few peeks at scale transformations, guide adjustments, and theme adjustment

---

# .kjh-lblue[`dplyr`] and Pipelining

###  The elements of filtering and summarizing

```r
gss_sm |> 
  group_by(bigregion, religion) |> 
  tally() |> 
  mutate(freq = n / sum(n),
         pct = round((freq*100), 1)) 
```

```
## # A tibble: 24 × 5
## # Groups:   bigregion [4]
##    bigregion religion       n    freq   pct
##    <fct>     <fct>      <int>   <dbl> <dbl>
##  1 Northeast Protestant   158 0.324    32.4
##  2 Northeast Catholic     162 0.332    33.2
##  3 Northeast Jewish        27 0.0553    5.5
##  4 Northeast None         112 0.230    23  
##  5 Northeast Other         28 0.0574    5.7
##  6 Northeast <NA>           1 0.00205   0.2
##  7 Midwest   Protestant   325 0.468    46.8
##  8 Midwest   Catholic     172 0.247    24.7
##  9 Midwest   Jewish         3 0.00432   0.4
## 10 Midwest   None         157 0.226    22.6
## # … with 14 more rows
```

---

## .huge.right.bottom.squish4[.kjh-yellow[Some data on]<br />.kjh-lblue[Organ Donation]]

---
# .kjh-pink[`organdata`] is in the .kjh-lblue[`socviz`] package

```r
organdata
```

```
## # A tibble: 238 × 21
##    country  year       donors   pop pop_d…¹   gdp gdp_lag health healt…² pubhe…³
##    <chr>    <date>      <dbl> <int>   <dbl> <int>   <int>  <dbl>   <dbl>   <dbl>
##  1 Austral… NA          NA    17065   0.220 16774   16591   1300    1224     4.8
##  2 Austral… 1991-01-01  12.1  17284   0.223 17171   16774   1379    1300     5.4
##  3 Austral… 1992-01-01  12.4  17495   0.226 17914   17171   1455    1379     5.4
##  4 Austral… 1993-01-01  12.5  17667   0.228 18883   17914   1540    1455     5.4
##  5 Austral… 1994-01-01  10.2  17855   0.231 19849   18883   1626    1540     5.4
##  6 Austral… 1995-01-01  10.2  18072   0.233 21079   19849   1737    1626     5.5
##  7 Austral… 1996-01-01  10.6  18311   0.237 21923   21079   1846    1737     5.6
##  8 Austral… 1997-01-01  10.3  18518   0.239 22961   21923   1948    1846     5.7
##  9 Austral… 1998-01-01  10.5  18711   0.242 24148   22961   2077    1948     5.9
## 10 Austral… 1999-01-01   8.67 18926   0.244 25445   24148   2231    2077     6.1
## # … with 228 more rows, 11 more variables: roads <dbl>, cerebvas <int>,
## #   assault <int>, external <int>, txp_pop <dbl>, world <chr>, opt <chr>,
## #   consent_law <chr>, consent_practice <chr>, consistent <chr>, ccode <chr>,
## #   and abbreviated variable names ¹pop_dens, ²health_lag, ³pubhealth
```

---

# First looks

```r
p <- ggplot(data = organdata,
            mapping = aes(x = year, y = donors))
p + geom_point()
```

---

# First looks

```r
p <- ggplot(data = organdata,
            mapping = aes(x = year, y = donors))
p + geom_line() 
```

---

# First looks

```r
p <- ggplot(data = organdata,
            mapping = aes(x = year, y = donors))
p + geom_line(aes(group = country)) 
```

---

# First looks

```r
p <- ggplot(data = organdata,
            mapping = aes(x = year, y = donors))
p + geom_line() + 
  facet_wrap(~ country, nrow = 3)
```

---

# First looks

```r
p <- ggplot(data = organdata,
            mapping = aes(x = year, y = donors))
p + geom_line() + 
  facet_wrap(~ reorder(country, donors, na.rm = TRUE), nrow = 3)
```

---

# First looks

```r
p <- ggplot(data = organdata,
            mapping = aes(x = year, y = donors))
p + geom_line() + 
  facet_wrap(~ reorder(country, -donors, na.rm = TRUE), nrow = 3)
```

---

## .huge.right.bottom.squish4[.kjh-yellow[Summarize better]<br /> .kjh-lblue[with **`dplyr`**]]

---

# Summarize a bunch of variables

```r
by_country <- organdata |>  
  group_by(consent_law, country)  |> 
    summarize(donors_mean= mean(donors, na.rm = TRUE),
              donors_sd = sd(donors, na.rm = TRUE),
              gdp_mean = mean(gdp, na.rm = TRUE),
              health_mean = mean(health, na.rm = TRUE),
              roads_mean = mean(roads, na.rm = TRUE),
              cerebvas_mean = mean(cerebvas, na.rm = TRUE))

head(by_country)
```

```
## # A tibble: 6 × 8
## # Groups:   consent_law [1]
##   consent_law country     donors_mean donors_sd gdp_mean healt…¹ roads…² cereb…³
##   <chr>       <chr>             <dbl>     <dbl>    <dbl>   <dbl>   <dbl>   <dbl>
## 1 Informed    Australia          10.6     1.14    22179.   1958.   105.     558.
## 2 Informed    Canada             14.0     0.751   23711.   2272.   109.     422.
## 3 Informed    Denmark            13.1     1.47    23722.   2054.   102.     641.
## 4 Informed    Germany            13.0     0.611   22163.   2349.   113.     707.
## 5 Informed    Ireland            19.8     2.48    20824.   1480.   118.     705.
## 6 Informed    Netherlands        13.7     1.55    23013.   1993.    76.1    585.
## # … with abbreviated variable names ¹health_mean, ²roads_mean, ³cerebvas_mean
```

- .medium[This works, but there's so much repetition! It's an open invitation to make mistakes copying and pasting.]

---
layout: false
class: main-title main-title-inv

# .middle.squish4.huge[.kjh-lblue[DRY:]] <br /> .middle.squish4.large.kjh-orange[Don't Repeat Yourself]

---

---

# Use .kjh-green[`across()`] and .kjh-green[`where()`] instead

```r
  by_country <- organdata |> 
    group_by(consent_law, country) |>
*     summarize(across(where(is.numeric),
                       list(mean = ~ mean(.x, na.rm = TRUE), 
                            sd = ~ sd(.x, na.rm = TRUE))))
head(by_country)              
```

```
## # A tibble: 6 × 28
## # Groups:   consent_law [1]
##   consen…¹ country donor…² donor…³ pop_m…⁴ pop_sd pop_d…⁵ pop_d…⁶ gdp_m…⁷ gdp_sd
##   <chr>    <chr>     <dbl>   <dbl>   <dbl>  <dbl>   <dbl>   <dbl>   <dbl>  <dbl>
## 1 Informed Austra…    10.6   1.14   18318.  831.    0.237  0.0107  22179.  3959.
## 2 Informed Canada     14.0   0.751  29608. 1193.    0.297  0.0120  23711.  3966.
## 3 Informed Denmark    13.1   1.47    5257.   80.6  12.2    0.187   23722.  3896.
## 4 Informed Germany    13.0   0.611  80255. 5158.   22.5    1.44    22163.  2501.
## 5 Informed Ireland    19.8   2.48    3674.  132.    5.23   0.187   20824.  6670.
## 6 Informed Nether…    13.7   1.55   15548.  373.   37.4    0.898   23013.  3770.
## # … with 18 more variables: gdp_lag_mean <dbl>, gdp_lag_sd <dbl>,
## #   health_mean <dbl>, health_sd <dbl>, health_lag_mean <dbl>,
## #   health_lag_sd <dbl>, pubhealth_mean <dbl>, pubhealth_sd <dbl>,
## #   roads_mean <dbl>, roads_sd <dbl>, cerebvas_mean <dbl>, cerebvas_sd <dbl>,
## #   assault_mean <dbl>, assault_sd <dbl>, external_mean <dbl>,
## #   external_sd <dbl>, txp_pop_mean <dbl>, txp_pop_sd <dbl>, and abbreviated
## #   variable names ¹consent_law, ²donors_mean, ³donors_sd, ⁴pop_mean, …
```

---

# Use .kjh-green[`across()`] and .kjh-green[`where()`] instead

```r
by_country <- organdata |> 
  group_by(consent_law, country) |>
*   summarize(across(where(is.numeric),
                       list(mean = ~ mean(.x, na.rm = TRUE), 
                            sd = ~ sd(.x, na.rm = TRUE))), 
*             .groups = "drop")
head(by_country)              
```

```
## # A tibble: 6 × 28
##   consen…¹ country donor…² donor…³ pop_m…⁴ pop_sd pop_d…⁵ pop_d…⁶ gdp_m…⁷ gdp_sd
##   <chr>    <chr>     <dbl>   <dbl>   <dbl>  <dbl>   <dbl>   <dbl>   <dbl>  <dbl>
## 1 Informed Austra…    10.6   1.14   18318.  831.    0.237  0.0107  22179.  3959.
## 2 Informed Canada     14.0   0.751  29608. 1193.    0.297  0.0120  23711.  3966.
## 3 Informed Denmark    13.1   1.47    5257.   80.6  12.2    0.187   23722.  3896.
## 4 Informed Germany    13.0   0.611  80255. 5158.   22.5    1.44    22163.  2501.
## 5 Informed Ireland    19.8   2.48    3674.  132.    5.23   0.187   20824.  6670.
## 6 Informed Nether…    13.7   1.55   15548.  373.   37.4    0.898   23013.  3770.
## # … with 18 more variables: gdp_lag_mean <dbl>, gdp_lag_sd <dbl>,
## #   health_mean <dbl>, health_sd <dbl>, health_lag_mean <dbl>,
## #   health_lag_sd <dbl>, pubhealth_mean <dbl>, pubhealth_sd <dbl>,
## #   roads_mean <dbl>, roads_sd <dbl>, cerebvas_mean <dbl>, cerebvas_sd <dbl>,
## #   assault_mean <dbl>, assault_sd <dbl>, external_mean <dbl>,
## #   external_sd <dbl>, txp_pop_mean <dbl>, txp_pop_sd <dbl>, and abbreviated
## #   variable names ¹consent_law, ²donors_mean, ³donors_sd, ⁴pop_mean, …
```

---

# Plot our summary data

.pull-left.w45[

```r
by_country |> 
  ggplot(mapping = 
           aes(x = donors_mean, 
               y = reorder(country, donors_mean),
               color = consent_law)) + 
  geom_point(size=3) +
  labs(x = "Donor Procurement Rate",
       y = NULL, 
       color = "Consent Law")
```
]

.pull-right.w55[
<img src="05-slides_files/figure-html/codefig-consent1-1.png" width="768" style="display: block; margin: auto;" />

]

---

# What about faceting it instead?

.pull-left.w45[

```r
by_country |> 
  ggplot(mapping = 
           aes(x = donors_mean, 
               y = reorder(country, donors_mean),
               color = consent_law)) + 
  geom_point(size=3) +
  guides(color = "none") +
* facet_wrap(~ consent_law) +
  labs(x = "Donor Procurement Rate",
       y = NULL, 
       color = "Consent Law")
```

.pull-left.w80[The problem is that countries can only be in one Consent Law category.]

]

.pull-right.w55[
<img src="05-slides_files/figure-html/codefig-consent2-1.png" width="768" style="display: block; margin: auto;" />
]

---

# What about faceting it instead?

.pull-left.w45[

```r
by_country |> 
  ggplot(mapping = 
           aes(x = donors_mean, 
               y = reorder(country, donors_mean),
               color = consent_law)) + 
  geom_point(size=3) +
  guides(color = "none") +
* facet_wrap(~ consent_law, ncol = 1) +
  labs(x = "Donor Procurement Rate",
       y = NULL, 
       color = "Consent Law")
```

.pull-left.w80[Restricting to one column doesn't fix it.]

]

.pull-right.w55[
<img src="05-slides_files/figure-html/codefig-consent2a-1.png" width="480" style="display: block; margin: auto;" />
]

---

# Allow the y-scale to vary

.pull-left.w45[

```r
by_country |> 
  ggplot(mapping = 
           aes(x = donors_mean, 
               y = reorder(country, donors_mean),
               color = consent_law)) + 
  geom_point(size=3) +
  guides(color = "none") +
  facet_wrap(~ consent_law, 
             ncol = 1,
*            scales = "free_y") +
  labs(x = "Donor Procurement Rate",
       y = NULL, 
       color = "Consent Law")
```

.pull.left.w90[Normally the point of a facet is to preserve comparability between panels by not allowing the scales to vary. But for categorical measures it can be useful to allow this.]

]

.pull-right.w55[
<img src="05-slides_files/figure-html/codefig-consent3-1.png" width="768" style="display: block; margin: auto;" />
]

---

# Again, these methods are general

.pull-left.w50[

```r
by_country |> 
  ggplot(mapping = 
           aes(x = donors_mean, 
               y = reorder(country, donors_mean),
               color = consent_law)) + 
* geom_pointrange(mapping =
*                   aes(xmin = donors_mean - donors_sd,
*                       xmax = donors_mean + donors_sd)) +
  guides(color = "none") +
  facet_wrap(~ consent_law, 
             ncol = 1,
             scales = "free_y") +  
  labs(x = "Donor Procurement Rate",
       y = NULL, 
       color = "Consent Law")
```
]

.pull-right.w50[
<img src="05-slides_files/figure-html/codefig-consent4-1.png" width="768" style="display: block; margin: auto;" />
]

---

# .kjh-green[`across()`] and .kjh-green[`where()`] again

```r
gss_sm |> 
  select(madeg, padeg)
```

```
## # A tibble: 2,867 × 2
##    madeg          padeg         
##    <fct>          <fct>         
##  1 High School    Graduate      
##  2 High School    Lt High School
##  3 Lt High School High School   
##  4 High School    <NA>          
##  5 High School    Bachelor      
##  6 High School    <NA>          
##  7 High School    High School   
##  8 Lt High School Lt High School
##  9 Lt High School Lt High School
## 10 High School    High School   
## # … with 2,857 more rows
```

---

# .kjh-green[`across()`] and .kjh-green[`where()`] again

```r
gss_sm |> 
  group_by(sex, padeg) |>
  summarize(across(where(is.numeric),
                     list(mean = ~ mean(.x, na.rm = TRUE), 
                          sd = ~ sd(.x, na.rm = TRUE)))) |> 
  select(sex, padeg, contains(c("age", "childs", "sibs"))) 
```

```
## # A tibble: 12 × 8
## # Groups:   sex [2]
##    sex    padeg          age_mean age_sd childs_mean childs_sd sibs_mean sibs_sd
##    <fct>  <fct>             <dbl>  <dbl>       <dbl>     <dbl>     <dbl>   <dbl>
##  1 Male   Lt High School     57.8   16.8        2.54      2.06      4.86    3.46
##  2 Male   High School        46.7   16.7        1.54      1.52      3.14    2.76
##  3 Male   Junior College     39.9   16.9        1.07      1.44      3.30    2.87
##  4 Male   Bachelor           43.3   14.6        1.27      1.35      2.54    2.41
##  5 Male   Graduate           39.9   14.8        1.01      1.35      2.36    1.88
##  6 Male   <NA>               47.1   17.1        1.75      1.67      3.84    3.21
##  7 Female Lt High School     58.5   18.0        2.46      1.72      4.74    3.43
##  8 Female High School        48.5   17.4        1.76      1.48      3.12    2.82
##  9 Female Junior College     39.2   11.6        1.46      1.43      3.19    2.00
## 10 Female Bachelor           44.8   15.4        1.32      1.35      2.88    2.62
## 11 Female Graduate           43.5   13.8        1.42      1.26      2.33    1.50
## 12 Female <NA>               47.4   17.8        2.08      1.69      4.65    3.93
```

---

# .kjh-green[`across()`] and .kjh-green[`where()`] again

```r
gss_sm |> 
  select(padeg, madeg, contains(c("age", "childs", "sibs"))) |> 
  group_by(padeg, madeg) |>
  summarize(across(where(is.numeric),
                     list(
                       mean = ~ mean(.x, na.rm = TRUE), 
                       sd = ~ sd(.x, na.rm = TRUE)
                       ))) |> 
  drop_na() |> 
  ggplot(mapping = aes(x = childs_mean,
                       xmin = childs_mean - childs_sd,
                       xmax = childs_mean + childs_sd,
                       y = madeg)) + 
  geom_pointrange() + 
  facet_wrap(~ padeg, ncol = 5)
```

---

# The .kjh-lblue[`nycdogs`] package

.pull-left.w60[

```r
library(nycdogs)
nyc_license
```

```
## # A tibble: 493,072 × 9
##    anima…¹ anima…² anima…³ breed…⁴ borough zip_c…⁵ license_…⁶ license_…⁷ extra…⁸
##    <chr>   <chr>     <dbl> <chr>   <chr>     <int> <date>     <date>       <dbl>
##  1 Paige   F          2014 Pit Bu… Manhat…   10035 2014-09-12 2017-09-12    2016
##  2 Yogi    M          2010 Boxer   Bronx     10465 2014-09-12 2017-10-02    2016
##  3 Ali     M          2014 Basenji Manhat…   10013 2014-09-12 2019-09-12    2016
##  4 Queen   F          2013 Akita … Manhat…   10013 2014-09-12 2017-09-12    2016
##  5 Lola    F          2009 Maltese Manhat…   10028 2014-09-12 2017-10-09    2016
##  6 Ian     M          2006 Unknown Manhat…   10013 2014-09-12 2019-10-30    2016
##  7 Buddy   M          2008 Unknown Manhat…   10025 2014-09-12 2017-10-20    2016
##  8 Chewba… F          2012 Labrad… Manhat…   10013 2014-09-12 2019-10-01    2016
##  9 Heidi-… F          2007 Dachsh… Brookl…   11215 2014-09-13 2017-04-16    2016
## 10 Massimo M          2009 Bull D… Brookl…   11201 2014-09-13 2017-09-17    2016
## # … with 493,062 more rows, and abbreviated variable names ¹animal_name,
## #   ²animal_gender, ³animal_birth_year, ⁴breed_rc, ⁵zip_code,
## #   ⁶license_issued_date, ⁷license_expired_date, ⁸extract_year
```
]

.pull-right.w40[
.center[![:scale 80%](img/07_nycdogs_hex.png)]
]

---

# Dogs of New York

```r
nyc_license
```

---