Communicating data science results effectively 📰

---

layout: true
  
<div class="my-footer">
<span>
<a href="https://DataScience4Psych.github.io/DataScience4Psych/" target="_blank">Data Science for Psychologists</a>
</span>
</div>

---

```
## These packages have more recent versions available.
## It is recommended to update all of them.
## Which would you like to update?
## 
## 1: All                          
## 2: CRAN packages only           
## 3: None                         
## 4: rlang (1.1.3 -> 1.1.4) [CRAN]
## 5: cli   (3.6.2 -> 3.6.3) [CRAN]
## 
## rlang (1.1.3 -> 1.1.4) [CRAN]
## cli   (3.6.2 -> 3.6.3) [CRAN]
## package 'rlang' successfully unpacked and MD5 sums checked
## package 'cli' successfully unpacked and MD5 sums checked
## 
## The downloaded binary packages are in
## 	C:\Users\smaso\AppData\Local\Temp\RtmpgtoHp3\downloaded_packages
## ── R CMD build ──────────────────────────────────────────────────
##   
  
  
✔  checking for file 'C:\Users\smaso\AppData\Local\Temp\RtmpgtoHp3\remotes81e05aa1117\mine-cetinkaya-rundel-nycsquirrels18-abae394/DESCRIPTION'
## 
  
  
  
─  preparing 'nycsquirrels18':
##    checking DESCRIPTION meta-information ...
  
   checking DESCRIPTION meta-information ... 
  
✔  checking DESCRIPTION meta-information
## 
  
  
  
─  checking for LF line-endings in source and make files and shell scripts
## 
  
─  checking for empty or unneeded directories
## 
  
─  building 'nycsquirrels18_0.0.0.9000.tar.gz'
## 
  
   
## 
```

# Communicating data science results effectively

---

# Five core activities of data science

1. Stating and refining the question
1. Exploring the data
1. Building formal statistical models
1. Interpreting the results
1. Communicating the results

.center[
<img src="img/data-science.png" width="60%" height="100%" style="display: block; margin: auto;" />
]
.footnote[
Roger D. Peng and Elizabeth Matsui. "The Art of Data Science." A Guide for Anyone Who Works with Data. Skybrude Consulting, LLC (2015).
]

---

# Stating and refining the question

---

## Six types of questions

1. **Descriptive:** summarize a characteristic of a set of data
1. **Exploratory:** analyze to see if there are patterns, trends, or relationships between variables (hypothesis generating)
1. **Inferential:** analyze patterns, trends, or relationships in representative data from a population
1. **Predictive:** make predictions for individuals or groups of individuals
1. **Causal:** whether changing one factor will change another factor, on average, in a population
1. **Mechanistic:** explore "how" as opposed to whether

.footnote[
Jeffery T. Leek and Roger D. Peng. "What is the question?." Science 347.6228 (2015): 1314-1315.
]

---

## Ex: COVID-19 and Vitamin D

1. **Descriptive:** frequency of hospitalizations due to COVID-19 in a set of data collected from a group of individuals
--

1. **Exploratory:** examine relationships between a range of dietary factors and COVID-19 hospitalizations
--

1. **Inferential:** examine whether any relationship between taking Vitamin D supplements and COVID-19 hospitalizations found in the sample hold for the population at large

--
1. **Predictive:** what types of people will take Vitamin D supplements during the next year

--
1. **Causal:** whether people with COVID-19 who were randomly assigned to take Vitamin D supplements or those who were not are hospitalized

--
1. **Mechanistic:** how increased vitamin D intake leads to a reduction in the number of viral illnesses

---

## Questions to data science problems

- Do you have appropriate data to answer your question?
- Do you have information on confounding variables?
- Was the data you're working with collected in a way that introduces bias?

--
.pull-left[
.question[
- Suppose I want to estimate the average number of children in households in Winston-Salem. 
- I conduct a survey at an elementary school in Winston-Salem and ask those students: 
  - how many children, including themselves, live in their house.
- Then, I take the average of the responses. 
]
]

--
.pull-right[
.question[
- Is this a biased or an unbiased estimate of the number of children in households in Winston-Salem? 
- If biased, will the value be an overestimate or underestimate?
]
]
---

# Exploratory data analysis

---

## Checklist

- Formulate your question
- Read in your data
- Check the dimensions
- Look at the top and the bottom of your data
- Validate with at least one external data source
- Make a plot
- Try the easy solution first

---

## Formulate your question

- Consider scope:
  - Are air pollution levels higher on the east coast than on the west coast?
  - Are hourly ozone levels on average higher in New York City than they are in Los Angeles?
  - Do counties in the eastern United States have higher ozone levels than counties in the western United States?
- Most importantly: "Do I have the right data to answer this question?"

---

## Read in your data

- Place your data in a folder called `data`
- Read it into R with `read_csv()` or friends (`read_delim()`, `read_excel()`, etc.)

``` r
library(readxl)
fav_food <- read_excel("data/favorite-food.xlsx")
fav_food
```

```
## # A tibble: 5 × 4
##   `Student ID` `Full Name`      favorite.food      mealPlan      
##          <dbl> <chr>            <chr>              <chr>         
## 1            1 Sunil Huffmann   Strawberry yoghurt Lunch only    
## 2            2 Barclay Lynn     French fries       Lunch only    
## 3            3 Jayendra Lyne    Peaches            Breakfast and…
## 4            4 Leon Rossini     Anchovies          Lunch only    
## 5            5 Chidiegwu Dunkel Pizza              Breakfast and…
```

---

## `clean_names()`

If the variable names are malformatted, use `janitor::clean_names()`

``` r
library(janitor)
fav_food %>% clean_names()  
```

```
## # A tibble: 5 × 4
##   student_id full_name        favorite_food      meal_plan       
##        <dbl> <chr>            <chr>              <chr>           
## 1          1 Sunil Huffmann   Strawberry yoghurt Lunch only      
## 2          2 Barclay Lynn     French fries       Lunch only      
## 3          3 Jayendra Lyne    Peaches            Breakfast and l…
## 4          4 Leon Rossini     Anchovies          Lunch only      
## 5          5 Chidiegwu Dunkel Pizza              Breakfast and l…
```
]

---

# Wrapping Up...
---

# Case study: NYC Squirrels!

---

## NYC Squirrels!

- [The Squirrel Census](https://www.thesquirrelcensus.com/) is a multimedia science, design, and storytelling project focusing on the Eastern gray (*Sciurus carolinensis*). They count squirrels and present their findings to the public.
- This table contains squirrel data for each of the 3,023 sightings, including location coordinates, age, primary and secondary fur color, elevation, activities, communications, and interactions between squirrels and with humans.

``` r
#install_github("mine-cetinkaya-rundel/nycsquirrels18")
library(nycsquirrels18)
```

---

## Locate the codebook

[mine-cetinkaya-rundel.github.io/nycsquirrels18/reference/squirrels.html](https://mine-cetinkaya-rundel.github.io/nycsquirrels18/reference/squirrels.html)

## Check the dimensions

``` r
dim(squirrels)
```

```
## [1] 3023   35
```

---

## Look at the top...

``` r
squirrels %>% head()
```

```
## # A tibble: 6 × 35
##    long   lat unique_squirrel_id hectare shift date      
##   <dbl> <dbl> <chr>              <chr>   <chr> <date>    
## 1 -74.0  40.8 13A-PM-1014-04     13A     PM    2018-10-14
## 2 -74.0  40.8 15F-PM-1010-06     15F     PM    2018-10-10
## 3 -74.0  40.8 19C-PM-1018-02     19C     PM    2018-10-18
## 4 -74.0  40.8 21B-AM-1019-04     21B     AM    2018-10-19
## 5 -74.0  40.8 23A-AM-1018-02     23A     AM    2018-10-18
## 6 -74.0  40.8 38H-PM-1012-01     38H     PM    2018-10-12
## # ℹ 29 more variables: hectare_squirrel_number <dbl>, age <chr>,
## #   primary_fur_color <chr>, highlight_fur_color <chr>,
## #   combination_of_primary_and_highlight_color <chr>,
## #   color_notes <chr>, location <chr>,
## #   above_ground_sighter_measurement <chr>,
## #   specific_location <chr>, running <lgl>, chasing <lgl>,
## #   climbing <lgl>, eating <lgl>, foraging <lgl>, …
```
]

---

## ...and the bottom

``` r
squirrels %>% tail()
```

```
## # A tibble: 6 × 35
##    long   lat unique_squirrel_id hectare shift date      
##   <dbl> <dbl> <chr>              <chr>   <chr> <date>    
## 1 -74.0  40.8 6D-PM-1020-01      06D     PM    2018-10-20
## 2 -74.0  40.8 21H-PM-1018-01     21H     PM    2018-10-18
## 3 -74.0  40.8 31D-PM-1006-02     31D     PM    2018-10-06
## 4 -74.0  40.8 37B-AM-1018-04     37B     AM    2018-10-18
## 5 -74.0  40.8 21C-PM-1006-01     21C     PM    2018-10-06
## 6 -74.0  40.8 7G-PM-1018-04      07G     PM    2018-10-18
## # ℹ 29 more variables: hectare_squirrel_number <dbl>, age <chr>,
## #   primary_fur_color <chr>, highlight_fur_color <chr>,
## #   combination_of_primary_and_highlight_color <chr>,
## #   color_notes <chr>, location <chr>,
## #   above_ground_sighter_measurement <chr>,
## #   specific_location <chr>, running <lgl>, chasing <lgl>,
## #   climbing <lgl>, eating <lgl>, foraging <lgl>, …
```
]

---

## Validate with at least one external data source

```
## # A tibble: 3,023 × 2
##     long   lat
##    <dbl> <dbl>
##  1 -74.0  40.8
##  2 -74.0  40.8
##  3 -74.0  40.8
##  4 -74.0  40.8
##  5 -74.0  40.8
##  6 -74.0  40.8
##  7 -74.0  40.8
##  8 -74.0  40.8
##  9 -74.0  40.8
## 10 -74.0  40.8
## 11 -74.0  40.8
## 12 -74.0  40.8
## 13 -74.0  40.8
## 14 -74.0  40.8
## 15 -74.0  40.8
## # ℹ 3,008 more rows
```
]
]
.pull-right-wide[
<img src="img/central-park-coords.png" width="100%" style="display: block; margin: auto;" />
]

---

## Make a plot

``` r
ggplot(squirrels, aes(x = long, y = lat)) +
  geom_point(alpha = 0.2)
```
]
]
.pull-left-wide[
<img src="d15_goodtalk_files/figure-html/unnamed-chunk-11-1.png" width="90%" style="display: block; margin: auto;" />
]

.pull-left-narrow[
**Hypothesis:** Squirrel sightings will have a higher density on the perimeter than inside the park.
]
---

## Try the easy solution first

``` r
squirrels <- squirrels %>%
  separate(hectare, 
           into = c("NS", "EW"), 
           sep = 2, remove = FALSE) %>%
  mutate(where = 
        if_else(NS %in% c("01", "42") | 
                 EW %in% c("A", "I"),
                      "perimeter",
                      "inside"))

ggplot(squirrels, aes(x = long, y = lat,
                      color = where)) +
  geom_point(alpha = 0.2)
```
]
]
.pull-left-wide[
<img src="d15_goodtalk_files/figure-html/unnamed-chunk-12-1.png" width="90%" style="display: block; margin: auto;" />
]
					   
---

## Then go deeper...

``` r
hectare_counts <- squirrels %>%
  group_by(hectare) %>%
  summarize(n = n())

hectare_centroids <- squirrels %>%
  group_by(hectare) %>%
  summarize(
    centroid_x = mean(long),
    centroid_y = mean(lat))

squirrels %>%
  left_join(hectare_counts,
            by = "hectare") %>%
  left_join(hectare_centroids,
            by = "hectare") %>%
  ggplot(aes(x = centroid_x, 
             y = centroid_y, 
             color = n)) +
  geom_hex()
```

```
## Warning: The following aesthetics were dropped during statistical
## transformation: colour.
## ℹ This can happen when ggplot fails to infer the correct
##   grouping structure in the data.
## ℹ Did you forget to specify a `group` aesthetic or to convert a
##   numerical variable into a factor?
```
]		
]
.pull-right[

<img src="d15_goodtalk_files/figure-html/unnamed-chunk-13-1.png" width="100%" style="display: block; margin: auto;" />
]
---

``` r
hectare_counts <- squirrels %>%
  group_by(hectare) %>%
  summarize(n = n())

hectare_centroids <- squirrels %>%
  group_by(hectare) %>%
  summarize(
    centroid_x = mean(long),
    centroid_y = mean(lat)
  )

squirrels %>%
  left_join(hectare_counts, by = "hectare") %>%
  left_join(hectare_centroids, by = "hectare") %>%
  ggplot(aes(x = centroid_x, y = centroid_y, color = n)) +
  geom_hex()
```

---

## The squirrel is staring at me!

``` r
squirrels %>%
  filter(str_detect(other_interactions, "star")) %>%
  select(shift, age, other_interactions)
```

```
## # A tibble: 11 × 3
##    shift age   other_interactions                                
##    <chr> <chr> <chr>                                             
##  1 AM    Adult staring at us                                     
##  2 PM    Adult he took 2 steps then turned and stared at me      
##  3 PM    Adult stared                                            
##  4 PM    Adult stared                                            
##  5 PM    Adult stared                                            
##  6 PM    Adult stared & then went back up tree—then ran to diffe…
##  7 PM    Adult stared at me                                      
##  8 AM    Adult approaches (saw me & came forward),runs from (sta…
##  9 AM    Adult started climbing down to me                       
## 10 PM    Adult stared at me                                      
## # ℹ 1 more row
```
]

---

## Communicating for your audience
.pull-left-narrow[
- Avoid: 
  - jargon, 
  - uninterpreted results, 
  - lengthy output
- Pay attention to: 
  - organization, 
  - presentation, 
  - flow
]

.pull-right-wide[
- Don't forget about: 
  - code style, 
  - coding best practices, 
  - meaningful commits
- Be open to: 
  - suggestions, 
    - feedback, 
    - taking (calculated) risks
]

---
# Sources

- Mine Çetinkaya-Rundel's Data Science in a Box ([link](https://datasciencebox.org/))
- Jeffery T. Leek and Roger D. Peng. "What is the question?." Science 347.6228 (2015): 1314-1315.
- Roger D. Peng and Elizabeth Matsui. "The Art of Data Science." A Guide for Anyone Who Works with Data. Skybrude Consulting, LLC (2015).
---

# Wrapping Up...