Welcome Back!
How are you doing?
Please fill out today’s survey in
#reporting-ii-2023
Intro to Dataviz
Today
- Project-related and homework-related announcements
- Exploratory Data Visualization
(visualizing data for yourself)
- Data Visualization
(visualizing data for the reader)
- Guest Speakers: Visual Journalists Ryan Best and Humera Lodhi from FiveThirtyEight
Project Announcements
(Milestones and Deadlines)
- Milestone 0: PITCH APPROVED
- Does the pitch/hypothesis need refinement? -
Milestone 1: DATASETS NAILED DOWN, METHODOLOGY DEFINED AND BACKGROUND RESEARCH HAS BEGUN
If you're building a dataset in Milestone 1 you've started collecting the data, ,you've decided on the variables, you probably have some rows of data already. -
Milestone 2: REPORTING IS WELL UNDERWAY, STRUCTURE OF THE STORY IS STARTING TO FORM
OCT 30: Outline Due -
Milestone 3: REPORTING IS DONE, PRIMARY DATA ANALYSIS IS DONE (it doesn't have to be pretty, but you need to have arrived at some conclusions)
Nov 20: First Draft Due -
Milestone 4: READER-FACING ASSETS ARE DONE
Dec 6: Project Presentation - Milestone 5: FINAL EDITS
Dec 11: Final Draft Due
Important Dates
-
OCT 30
Outline Due
-
Nov 20
First Draft Due
-
Dec 6
Project Presentations
- Dec 11 - Final Draft Due
Homework Announcements
Reminder: Don't forget to respond to ⛔️, ❓and 🤯
Your response can be a rewrite of that section or a reflection or discussion in the comment to demonstrate that you understand thoroughly what the issue was and how to prevent it from happening again.
Feedback Status
- [ x ] Diagnostic Assignment Feedback
- [ x ] Reverse engineer a story
- [ x ] pre-pitch brainstorm
- [ x ] selected pitches
in progress - [ ] Data Assignment 2 feedback
- [ ] Pitch Feedback (for pitches that were not selected)
- [ ] OSINT/FOIA Feedback
Learning Objectives
-
Communicate stories in data effectively for your audience with charts and tables.
-
Know when and how to apply statistical treatments to data and how to communicate your methodological choices to your audience.
- Transparently and effectively communicate the uncertainty embedded in quantitative analysis.
Exploratory Data Viz
(visualizing data for yourself)
A quick polly about your exploratory data visualization homework...
1977
slide credit: Nikhil Garg https://orie5355.github.io/Fall_2021/static_files/lectures/Lecture2_WhatIsData.pdf
slide credit: Nikhil Garg https://orie5355.github.io/Fall_2021/static_files/lectures/Lecture2_WhatIsData.pdf
Acquiring Data
- What is this data?
- What does this data measure?
- What constructs will it help me understand?
- Is there other data that might measure something else but help me get at the same constructs I'm interested in analyzing.
- Are there other datasets I can merge with this one?
Acquiring Data
Understanding the Data
- how was it put together?
- what is each row?
- what is each column?
- what kinds of unique values are in each column?
- what are the caveats?
Acquiring Data
Understanding Data
Cleaning Data
⚠️ if you make any methodological choices when cleaning the data, explore the implications of each choice you made - each choice will follow you all the way to whatever conclusions you draw.
Acquiring Data
Understanding Data
Cleaning Data
Summarizing Data
- How is the data distributed?
- What are the ranges (max/min)
- What are the central tendencies?
- Sanity checks: what do I expect to see? Do I see that?
Acquiring Data
Understanding Data
Cleaning Data
Summarizing Data
Interviewing Data
Asking targeted questions:
Imagine your data is a source, what questions do you have of it?
Acquiring Data
Understanding Data
Summarizing Data
Interviewing Data
Contextualizing The Answers
Data Visualization
(visualizing data for the reader)
Pair Programming Data Vis
Assignment
