forked from dgmiller/datavis_resources
-
Notifications
You must be signed in to change notification settings - Fork 7
/
Copy pathREADME.md~
200 lines (119 loc) · 8.38 KB
/
README.md~
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
# Data Visualization
[Thursday Portfolio Reviews Google Sheet](https://docs.google.com/spreadsheets/d/1DSqLDCsrzKlF4G4GbHl_ICxONLsGzOTRkiB9G3NK9iI/edit?usp=sharing)
## Portfolio Projects
+ Redesign an ACME lab
+ Complete a visual data analysis project and write up on Medium
### March 21: Guest Lecture: Doug Thomas
Meet in the HFAC room A460
### March 14: Presentations
Read Tufte's *The Cognitive Style of Powerpoint*
Basically, handouts are better than slides 90% of the time.
**Homework**: keep working on your projects
### March 7: Data Narratives
Introduction, Conflict, Resolution
[Science Isn't Broken](https://fivethirtyeight.com/features/science-isnt-broken/)
[2016 Election Coverage](https://pudding.cool/2017/03/elections/)
Stories provide context. They are a mental model that helps us remember information. Facts are hard to memorize. Stories stick.
### February 28: Color
Start by either speed drawing or blind contour drawing an image.
Watch [this](https://bids.github.io/colormap/) color video
Show the person next to you what you are working on. Talk about the color.
Sign up for individual meeting times before the end of the semester.
[Great podcast about color](http://datastori.es/episode-7-color-feat-gregor-aisch/)
**Homework**: Create a github account and make a repo called Portfolio. Start pushing your iterations up to this portfolio.
### February 14: Workday
Practice makes perfect.
Since I was unavailable last Thursday, it's a work-in-class day. That way, I can come around and meet with those who I wasn't able to meet with. My apologies.
### February 7: Learning to See
Training Your Eyes to See
*Anyone who can't communicate their ideas is on the same plane as someone who has no ideas.* -- Pericles
ANNOUNCEMENT: There are 4 Tufte books and another book called Dataclysm in the 150 computer lab.
You are free to use them to get ideas and learn more about data vis. Please keep them in the room
and don't write in them. I recommend starting with *The Visual Display of Quantitative Information*.
Activities:
+ Watch [How to Doodle](https://www.youtube.com/watch?v=6gSmQNmC-5c) and doodle in class.
+ Learn how to see by drawing without looking at your paper. This is an exercise called [Blind Contour Drawing](https://en.wikipedia.org/wiki/Blind_contour_drawing).
+ Learn what is important in a sketch by drawing it in under a minute. I call this one-minute sketching.
Designers have a saying: get it right in black and white. If it doesn't make sense in black and white, it will make less sense with color and style.
**Homework**: Have a finished redesign of an ACME lab to show me on Thursday. This is a redesign of the solutions file and plots that you turned in to complete the lab, NOT a redesign of the written pdf of the lab. I recommend using a jupyter notebook and include some comments. I expect a minimum of an hour for this project. Do not spend more than three hours on this assignment. It is meant to show you how to use basic visual variables in matplotlib to improve a visualization.
### January 31: Exploration vs Communication
*The greatest value of a picture is when it forces us to notice what we never expected to see.* -- John Tukey
Talk about the difference between exploring a data set and communicating an insight in a data set.
What are the differences in the visualization methods? What should you focus on in each? What skills
or abilities do you need for each? Pros and Cons of both?
+ Tufte Cholera Example
+ Tufte Challenger PowerPoint
Examples of Medium data vis blog post
+ [Rogue Train Caught with Data](https://blog.data.gov.sg/how-we-caught-the-circle-line-rogue-train-with-data-79405c86ab6a#.4zgj3hs33)
+ [Germany's U-boats and Data Visualization](https://medium.com/@kadenhendron/germany-s-u-boats-data-visualization-6e018c6c174#.9u14mvypt)
+ [Dialogue in Classic Literature](https://medium.com/@thesarahkay/how-often-do-classic-characters-chat-67525d0e5008#.oyi1te6vz)
Be careful about using exploratory analysis to confirm your own beliefs. Some good blog posts by Andrew Gelman at Columbia:
+ [Exploratory and Confirmatory Studies](http://andrewgelman.com/2010/02/16/exploratory_and/)
+ [Thinking more seriously about exploratory studies](http://andrewgelman.com/2016/11/17/thinking-more-seriously-about-the-design-of-exploratory-studies/)
**Homework**: Sign up for a time to meet with me on Thursday. Bring either some visualizations for your ACME lab or stuff for your Data Report project.
### January 24: Visual Variables
ANNOUNCEMENT: Formal class will no longer be held on Thursdays to not conflict with the soft skills seminar. All are encouraged to
attend the soft skills seminar. Instead of Thursday meetings as a class, I will be holding personal reviews at a time of your choosing
on Thursday. Go to [this link](https://docs.google.com/spreadsheets/d/1DSqLDCsrzKlF4G4GbHl_ICxONLsGzOTRkiB9G3NK9iI/edit?usp=sharing)
to sign up for a 10 minute slot on Thursday. If none of these times work for you, contact me and schedule another time.
With a partner, critique the following visualizations:
+ [xkcd Earth Temperature](https://xkcd.com/1732/)
+ [Polygraph Film Dialogue](http://polygraph.cool/films/ `)
+ [FiveThirtyEight Gun Deaths](https://fivethirtyeight.com/features/gun-deaths/)
*Visual Variables*
+ position
+ size
+ hue
+ saturation
+ value
+ shape
More about Hue, Saturation, and Value (HSV) [here](http://learn.leighcotnoir.com/artspeak/elements-color/hue-value-saturation/).
*Drawing exercise*
In a group, draw as many representations of this data set: 25, 13. Yes there are only two numbers. Take two minutes to
come up with as many representations as you can. Draw them on paper. What did you learn? Discuss.
**Homework**: Schedule a Thursday portfolio review with me.
### January 19: Concept Visualization
In a group, critique the initial visualizations of your data set. Focus your critiques on what the creator should
do next with this visualization (choose better colors, make dots bigger, add more variables, etc).
Concepts can be visualized just like data. Discuss in a group what the following visualizations do to represent a concept:
[Explained Visually](http://setosa.io/ev/)
[Setosa Visualizations](http://setosa.io/#/)
**Due next time**: choose an ACME lab to redesign and show potential employers.
### January 17: Know Your Data
Today we are talking about [Dear Data](http://www.dear-data.com/theproject) and multidimensional visualizations.
[FiveThirtyEight Dear Data](https://fivethirtyeight.com/features/we-asked-you-to-visualize-your-podcast-listening-and-wow-did-you-deliver/)
Collecting your own data and visualizing it teaches you a lot about how to represent data effectively.
You learn that people make assumptions when they want to collect data to measure something.
Your visualizations should reflect the choices and assumptions made during the data collection process.
When you are ready to communicate insight you have found in a data set, always start with paper and pen or pencil.
Draw what you want the visualization to look like before you build it.
**Due next time**: Have a visualization to present.
### January 12: Data, Not Visualization
We talked about the [Gapminder](https://www.ted.com/talks/hans_rosling_shows_the_best_stats_you_ve_ever_seen) data vis.
Data is more interesting than the visualization.
Find interesting data first, then create a visualization.
Tools you could use:
+ Tableau (free for students)
+ matplotlib
+ [D3.js](https://d3js.org/) (if you know javascript)
+ Bokeh
+ [Processing](https://processing.org/)
I added some files in the repo to help you load your data set in Pandas (python).
**Due next time**: Get your hands on a data set you want to visualize.
### January 10: Introduction
We learned how to critique a data visualization.
[A Visual Introduction to Machine Learning](http://www.r2d3.us/visual-intro-to-machine-learning-part-1/)
Some questions to ask when evaluating representations of data:
+ What are your first impressions?
+ Do you like the visualization?
+ Does it help you understand the data?
+ What is the creator trying to demonstrate?
+ What is being compared?
+ How many variables are present?
+ Is there anything unnecessary or redundant?
+ Where does the data come from?
+ Is the data presented in context?
[How to learn data visualization](http://datastori.es/episode-5-how-to-learn-data-visualization-with-andy-kirk/)
## Other
[Cool Wind](http://hint.fm/wind/gallery/index.html)
[Feltron](http://feltron.com/)