Statistical Thinking

## What words or phrases do you think of when you hear the word .orange["Harvard"]?

### This being a .orange[data] class, I'd like to collect some data related to .orange["statistical thinking."]

### Go to [bit.ly/stat-100-think](https://bit.ly/stat-100-think) to provide the words or phrases you think of when you hear .orange["statistical thinking."]

---

background-image: url("img/DAW.png")
background-position: left
background-size: 50%
class: middle, center,

## .blue[Statistical Thinking]

<br>

### .purple[Kelly McConville]

#### .purple[ Stat 100 | Week 1 | Fall 2022]

]

---

class:

## Getting Started in Stat 100

#### Step 1: Getting Started Module in Canvas

---

## Announcements

* Lecture slide decks will always be posted and linked to a Canvas Module the day before lecture.
    + These are HTML files but you can print them to PDF.
    + Looking into printed options.
* No section, no lecture quiz, no p-set this week.
* Only I will be running office hours this week at the following times:
    + Today: 1:30 - 2:30pm Science Center 316.04
    + Thursday: 10 - 11am Science Center 316.04
* No lecture or office hours next Monday!
* Be on the look-out for the section preference form.

---

## Week 1 + Week 2 Goals

**Week 1 Lecture**

* Statistical thinking
* Introduction to data
* Hand-drawn data visualizations
* Hop into the RStudio Server using FAS OnDemand

]

**Week 2 Lecture**

* Discuss course structure
* Getting up and running in `RStudio`
* Working with `RMarkdown` documents

]

---

## But first, let me quickly introduce myself...

---

### Let's start with my path to Harvard...

---

## Research Interests

### Survey statistics and collaborate with

---

## Research Interests

### Where survey statistics meets data science

---

### Advising Undergraduate Forestry Data Science Research

---

class: middle
background-image: url("img/seedlings.jpg")
background-position: left
background-size: contain

* I **love** teaching stats and coding.

]

* But, learning stats and coding is **hard**.

]

* With the **right scaffolding**, **good strategies**, and **sustained effort**, you can excel at both!

]

* And mistakes are part of the learning process.  They don't imply that you are bad at stats.

]

---

## The Rest of the Teaching Team

]

<img src="img/team.png" width="95%" style="display: block; margin: auto;" />
]

---

background-image: url("img/structures.001.jpeg")
background-position: contain
background-size: 65%

## Stat 100 Tech & Materials

---

## Stat 100 is about developing our .orange[statistical thinking] skills.

### What is .orange[statistical thinking]?

### It is not the same as mathematical thinking.

### Let's discover what .orange[statistical thinking] is through some examples.

---

## Data in Stat 100

Will use a wide-range of **real** and **relevant** data examples

]

]

---

## Data in Stat 100

]

]

* I understand that some of these topics have likely had profound impacts on your lives.

* We will focus class time on the key course objectives but will use these current topics to empower ourselves and to see how we can productively participate with data.

---

## Example: Visualizing COVID Prevalence

---

### Example: Visualizing COVID Prevalence

* In May of 2020, the Georgia Department of Public Health posted the following graph:

* At a quick first glance, what story does the Georgia Department of Public Health graph appear to be telling?

* What is misleading about the Georgia Department of Public Health graph?  How could we fix this issue?

---

### Example: Visualizing COVID Prevalence

* After public outcry, the Georgia Department of Public Health said they made a mistake and posted the following updated graph:

* How do your conclusions about COVID-19 cases in Georgia change when now interpreting this new graph?

---

### Example: Visualizing COVID Prevalence

* Alberto Cairo, a journalist and designer, created the second graph of the Georgia COVID-19 data:

* A key principle of data visualization is to “help the viewer make meaningful comparisons”.

* What comparisons are made easy by the lefthand graph?  What about by the righthand graph?

* From these graphs, can we get an accurate estimate of the COVID prevalence in these Georgian counties over this two week period?

---

### Example: Visualizing COVID Prevalence

* The [Massachusetts Water Resources Authority (MWRA) graph](https://www.mwra.com/biobot/biobotdata.htm) tracks the presence of COVID-19 in the Boston-area wastewater.

* What are the pros of using wastewater over nasal swabs to assess COVID prevalence?  What are the cons?

* One more note: The graph also incorporates **uncertainty measures**, a key statistical thinking idea that we will learn more about later in the semester!

---

## What is "Statistical Thinking?"

---

## Statistical Thinking

* Understanding the importance of **context**.

]

&#8594; Context explained the Monday jumps in the COVID counts.

]

* How we **encode** information in a graph should be driven by our research question.

]

&#8594; **Design choices** impact the conclusions the viewer draws.

]

* How the data were **collected** impacts the conclusions we can make.

]

&#8594; Voluntary COVID test results don't likely provide good estimates of COVID prevalence.

]

* Often we are using a **sample** of data to say something about a larger group.  In this case, we should measure how certain our estimates are!

]

&#8594; We will learn to **compute** and **interpret** certainty estimates like those in the wastewater graph later in the course!

]

Developing our statistical thinking skills will allow us to soundly **extract knowledge from data**!

---

## What are data?

---

* The dictionary definition:

> "data: factual information (such as measurements or statistics) used as a basis for reasoning, discussion, or calculation" -- Merriam-Webster

* Wikipedia:

> "Data are characteristics or information, usually numerical, that are collected through observation. In a more technical sense, data are a set of values of qualitative or quantitative variables about one or more persons or objects, while a datum (singular of data) is a single value of a single variable."

---

* Our textbook definition:

> "Data comes to us in a variety of formats, from pictures to text to numbers." -- ModernDive

* Data Feminism:

> "... by the time that information becomes data, it's already been classified in some way. Data after all, is information made *tractable*." -- D'Ignazio and Klein

---

## Data Frames

<table class="table table-responsive table-bordered table-striped" style="font-size: 12px; margin-left: auto; margin-right: auto;">
 <thead>
  <tr>
   <th style="text-align:left;"> UserID </th>
   <th style="text-align:right;"> Tree_Height </th>
   <th style="text-align:left;"> Common_Name </th>
   <th style="text-align:left;"> Park </th>
   <th style="text-align:right;"> DBH </th>
   <th style="text-align:left;"> Species_Factoid </th>
  </tr>
 </thead>
<tbody>
  <tr>
   <td style="text-align:left;"> 1 </td>
   <td style="text-align:right;"> 105 </td>
   <td style="text-align:left;"> Douglas-Fir </td>
   <td style="text-align:left;"> Gammans Park </td>
   <td style="text-align:right;"> 37.4 </td>
   <td style="text-align:left;"> Bracts on cones look like a mouse's feet and tail. </td>
  </tr>
  <tr>
   <td style="text-align:left;"> 2 </td>
   <td style="text-align:right;"> 94 </td>
   <td style="text-align:left;"> Douglas-Fir </td>
   <td style="text-align:left;"> Gammans Park </td>
   <td style="text-align:right;"> 32.5 </td>
   <td style="text-align:left;"> Bracts on cones look like a mouse's feet and tail. </td>
  </tr>
  <tr>
   <td style="text-align:left;"> 3 </td>
   <td style="text-align:right;"> 23 </td>
   <td style="text-align:left;"> Lavalle Hawthorn </td>
   <td style="text-align:left;"> Gammans Park </td>
   <td style="text-align:right;"> 9.7 </td>
   <td style="text-align:left;"> Like most hawthorns, the tree has stout thorns up to 2" long. </td>
  </tr>
  <tr>
   <td style="text-align:left;"> 4 </td>
   <td style="text-align:right;"> 28 </td>
   <td style="text-align:left;"> Northern Red Oak </td>
   <td style="text-align:left;"> Gammans Park </td>
   <td style="text-align:right;"> 10.3 </td>
   <td style="text-align:left;"> Acorns take two years to mature and are an important food source for wildlife. </td>
  </tr>
  <tr>
   <td style="text-align:left;"> 5 </td>
   <td style="text-align:right;"> 102 </td>
   <td style="text-align:left;"> Douglas-Fir </td>
   <td style="text-align:left;"> Gammans Park </td>
   <td style="text-align:right;"> 33.2 </td>
   <td style="text-align:left;"> Bracts on cones look like a mouse's feet and tail. </td>
  </tr>
  <tr>
   <td style="text-align:left;"> 6 </td>
   <td style="text-align:right;"> 95 </td>
   <td style="text-align:left;"> Douglas-Fir </td>
   <td style="text-align:left;"> Gammans Park </td>
   <td style="text-align:right;"> 32.1 </td>
   <td style="text-align:left;"> Bracts on cones look like a mouse's feet and tail. </td>
  </tr>
</tbody>
</table>

Data in spreadsheet-like format where:

* Rows = Observations/cases

* Columns = Variables

---

## Data Frames

Rows = Observations/cases

**What are the cases?  What does each row represent?**

---

## Data Frames

Columns = Variables

**Variables**: Describe characteristics of the observations

* **Quantitative**: Numerical in nature

* **Categorical**: Values are categories

* **Identification**: Uniquely identify each case

---

## Data Frames

**Important to understand what each variable represents and the units of measurement.**

Example questions:

* For categorical variables, what are the categories? Do those categories adequately represent the data represented by that variable?

* For quantitative variables, what values are possible? Were the data rounded or binned?  Are those values actually encoding categories?

---

## Goal: Over the next week, collect data on your world so that you can visualize it on P-Set 1.

---

## Hand-Drawn Data Viz

* Two key aspects of data visualization:
    + Determining how you want to display the data.
    + Figuring out how to tell the computer to do that mapping.

* Hand-drawn data visualizations allow us to focus on the first part and with full control over the creative process!

---

## Hand-Drawn Data Viz Examples

* [Dear Data](http://www.dear-data.com/theproject)

> "Each week, and for a year, we collected and measured a particular type of data about our lives, used this data to make a drawing on a postcard-sized sheet of paper, and then dropped the postcard in an English “postbox” (Stefanie) or an American “mailbox” (Giorgia)!"

---

### Dear Data Examples