SRCCON 2018 • June 28 & 29 in MPLS Support OpenNews!

Session Transcript:
Visualizing by hand

Session facilitator(s): Amelia McNamara

Day & Time: Friday, 4:15-5:30pm

Room: Johnson

AMELIA: Well, good afternoon, folks. Thank you for coming to visualizing data by hand with me. I think it’s going to be fun. For those I don’t know, I’m Amelia McNamara. I’m in sort of a career transition right now. I was teaching data science and statistics at Smith College in Massachusetts. And as of July 1st I’m at the University of St. Thomas in their department of computer and information science. I sort of think about teaching and making statistics easier for everyone. I am @AmeliaMN, I live in Minnesota again, so it’s a double entendre.

I brought a ton of craft supplies. There’s wire and pom poms and googly eyes. And different stickers and balloons. We can do all kinds of fun stuff. I just want you to kind of be kind of thinking about that as I’m setting the stage. And then most of the time it’s going to be for us to produce something and have fun and hopefully unwind from the conference. So, I have sort of like three people or projects that really inspired me in this session. And the first one is Jose Duarte, And you can buy it if you want to do an exercise like this. I just had craft supplies at my house.

But his thing is you can make data visualization tangible and maybe it will be more engaging to people if it’s something that they can interact with kind of physically. So, he does these murals which are out on walls that have data about the community that they’re in. Like I said, he has this kit that you can get. So, that’s why I have balloons. He thinks that balloons are a good piece of your data visualization toolkit. And he has some other things, you know, to represent like one item or to create bars and things like that physically.

And I met him I was at the data literacy conference in France last whenever this was September, I guess. So, we did a workshop where my group decided to do this data visualization of the world happiness index on a set of stairs using colors and then it had like a key that you could you can read and you could sort of interact with it by walking on. So, I think that that’s one kind of I mean, it’s like handmade, but it’s also physical in some way.

And I think that if you wanted to create something like that here today, you could. Obviously we’re at the end of the conference and I don’t know if you want something to persist. But you could build something kind of big and grand if you wanted to.

I’m also really inspired by Mona Chalabi. I’m sure most of you are familiar with her work. She’s been a number of different places including let me see The Guardian and I think FiveThirtyEight. And I’m forgetting where she is right now. But she had this like Dear Mona advice column where people would write in with questions about the world like how much pee is a lot of pee, and she would find the data to answer their question. The way she would answer it, there would be some writing, but also a hand drawn visualization.

She has a talk from OpenVis Conf linked on the slides. The slides are on the Etherpad and you can get the links there. This is an example from her Instagram. Which I recommend following if you don’t already where she is visualizing the amount of holiday candy sold in 2015. And she’s using like Hershey’s chocolate bars to be the units of measure. So, it’s sort of making the data more playful and also kind of like making it physical in some way.

And then she I really love that she does not shy away from maybe sensitive topics and like gross stuff. So, she did this one about the peak month of flu viruses that has snot hanging down. And she has this one about discharge where she has like the underpants and sort of with small multiples thing where she’s done it, the data visualization. So, she’s just making the data more like inviting by doing it in this hand drawn style.

And so, both Jose and Mona I think tend to visual really small data sets. And if you’re going to visualize something by hand, it’s usually easier to do small data than big data. But my last piece of inspiration is Giorgia Lupi and Stefani Posavec and their Dear Data Project where they decided to communicate through data visualization over 52 weeks of postcards.

They would choose a topic for the week, collect data about the topic and then visualize it by hand on postcards and then mail the postcards across the ocean. So, I think this is about time. And I’m forgetting well, I guess the one, if it says Giorgia’s, it’s sent to her, so, it’s Stefanie’s. And the one that says Stefanie is sent and it’s looking at clocks and the way that they chose to visualize. But the data is different. The back is always the key. There’s church clock is the filled in triangle and the watch is this just sort of like upward carat. And there’s different symbols for different things. They did this hand drawn data visualization project and they were inventing new visualization forms.

This isn’t a standard bar chart. Mona is doing the snot dripping down, but it’s using the snot as the length of the bar. These are new forms that you wouldn’t have thought of on the computer. That’s one thing. And then the other thing is they’re not afraid to use kind of big data. They’re like using a lot of marks. And there’s there’s many, many pieces of data that are being represented.

So, this was another week. Oh, my gosh. I think this one was about laughing and you can’t read it here in the images. But, again, color means something and then there’s also, you know, these star bursts coming out and the connections and all of the different pieces of the visualization have some meaning in the visualization which they’ve chosen. And so, they’re kind of like breaking away from the constraints of what they might ordinarily think about as visualization and building these beautiful things.

And then just my last piece of inspiration is this post by Nathan Yau, one dataset in many ways. And essentially he just took a dataset which I believe is about life expectancy and he tried to think of as many ways as possible. And I think he got 25 different ways. And you can see some of them, right? So there’s like, broken out into small multiple histograms, this kind of cyclical graph, filled in densities, a scatter plot. He tried to have as many ways to visualize data as possible. He’s, of course, using a computer, it’s not by hand.

He’s using mostly familiar forms. And so, that leads me to our first exercise where we’re going to try and do some iteration and practice. This is, again, from Dear Data. Stefanie at the beginning of the year said I’m terrible at drawing. And her husband said, well, you’ll be better after a year. The more that you practice it, the better you get and just like the more comfortable you are. So, what I want us to do is to start with an exercise where you’re going to take there should be 8.5x11 sheets of paper on the table. Take it and fold it in half both ways so you get quadrants.

And then you’re going to have sort of four spaces on either side. So, you have space for eight ideas. And I want you to try to visualize this small dataset since we’re in Minneapolis. I picked data about the lakes. And it’s mostly numeric data. So, I tried to throw on here like a categorical one in case that was interesting to you. You don’t have to visualize all the variables if you don’t want to. Or you can offer the observations. I just wanted to give a little bit more room.

So, I want you to try and think of eight different ways to visualize this same data. And I’m going to give you five or six minutes. So, that’s like 30 45 seconds an idea. And I want you to like kind of sketch out the idea. If you need like pens or pencils I have a bunch of stuff in the bag up here. There’s more pens here in this box. Yeah. And you can always take whatever supplies. If you need googly eyes right now, you can get those. But I think mostly let’s see. Some pens. All right. And I’m going to set a timer.

If you need to start with googly eyes, start with googly eyes. I don’t know if they I think I have tape. I wasn’t thinking about that.

AUDIENCE: You had them. That’s all that matters.

[Working on visualizations]

AMELIA: Okay. It’s been about a minute. You should have at least one idea down. Move on to another one. Some of them are going to be terrible. That’s okay. And it doesn’t have to be either. We’re just trying to get an idea.

You’re halfway done. Is this too stressful?

AUDIENCE: Like a Zach Morris time out.

[Working on visualizations]

AMELIA: Okay. I’ll give you one more minute. So try to think of at least one more idea. Remember, it could be like a note to yourself about how you use pom poms or wire or, you know, some some supply that I have here. Balloons.

Okay. Time. So, now what I would like you to do is we’re all really intently focused and super quiet is talk to the people who are sitting at your table. And there’s a few people who are like maybe you could consolidate if you want is someone to talk to. But just talk about like maybe pick your favorite of the eight or two or four or however many you came up with and maybe show it to your table. And also talk like the experience of making the visualizations. Like, was it hard? Is your favorite one the first one? The last one? Like, how is the process of coming up with all of those? And I’ll give you some time to go around and introduce yourselves and talk about what you made.

[Group work]

Okay. So, I do want to take like just a couple minutes because I wasn’t part of your conversations. I would love to hear what has been like coming out. Like, so, can anyone from any of the groups say something that was like an experience that maybe several people had? Or a common theme that came up in what you were working on? Yeah?

AUDIENCE: We all struggled to encode every variable into each visualization, so we chose two or one to focus on.

AMELIA: Yeah, I see heads nodding.

AUDIENCE: There were a couple of fried egg diagrams.


AUDIENCE: Fried egg. The water shed area and the lake area.

AMELIA: Okay. Cool.

AUDIENCE: What’s wrong with that?

AUDIENCE: It’s delicious.


AUDIENCE: And where in the name is the word “Lake” if it’s the first word, the last word or not one of the words.

AMELIA: I love that. Seriously.

AUDIENCE: We did a lot of narratives amongst the data that was represented by the lakes. But also how to represent that as a in a visual way that was simple but compelling. And also how to narrate life around the lakes and sort of thinking about the qualities and visualizing different kinds of ecologies. The layers of the lake, under the lake, the depth of the lake. Maybe industries that surround the lake. So, things like that might be also sharing the data. What’s found. It’s not part of the dataset. But also the area ecology.

AMELIA: I heard some additional interesting from this table as well. I picked because Wikipedia had this information. But if you were to create something compelling with this data, you might look for more stuff to fill it out and make it more interesting. Anything else? Was your favorite one the first one? The last one? Did anyone actually do it in that amount of time?

AUDIENCE: I did one because I’m insubordinate. And I just want to let everyone know that.

[ Laughter ]

AMELIA: Great.

AUDIENCE: Mine were like most creative to least creative.

AMELIA: You had a great idea, and uh oh, I need to fill it out, so bar chart.

AUDIENCE: The last one was definitely that.

AUDIENCE: Mine was opposite. How to encode as much as possible in a compact way.

AMELIA: Okay. Cool. Anything else about the experience?

AUDIENCE: Yeah. I had an idea and drew it out and realized there was a better way to do something very similar. And so, my next chart was a better version of the previous one.

AMELIA: Cool. Yeah. I think people have different experiences of this like being forced to come up with a lot of ideas. Some people are insubordinate and they just want to do one really good one. But I think that it is like it forces you to be creative to have that structure and constraint and thinking more. You have to iterate. It has to be completely different. You know, one can be just sort of like a nicer version of the previous one.

Okay. So, another thing about the Etherpad is that in the Etherpad, I put a link to my slides. My slides are editable. If you want to put images into the slides, you can do that. If you have more inspiration of people who do handmade data visualization, you can add some more inspiration slides. If you wanted to put in something that you made here, like add a slide and put a photograph in that you made, that would be awesome. But I don’t know if that’s really reasonable.

And then what I want us to do for the bulk of the time in this session is to to work on, like, putting together one cool visualization. And you might have like some different drafts. You might I didn’t put design here. But you might use the dataset we were just using or pick something else. But I think that one distinguishing factor is if you want to work with other people or if you want to work alone. So, if you would like to have a group and work together like you could form a group at a table if you wanted to regroup and do that. You can also do that once we start working.

But I thought I would give you a second to think about that. And then what I want you to do is either pick one of your sketches or pivot to a different dataset and do something real. I am not going to structure the draft process. But that’s part of it. In the Etherpad I have put some links to data. So, let’s see if I can reconnect here.

So, I love the FiveThirtyEight data repository. I don’t know if anyone is here from FiveThirtyEight. But you probably know they have all of their underlying data on GitHub. And I was trying to figure out how to share that with you and then this morning I went to the dataset session and I would really recommend using that interface you want to look at a FiveThirtyEight dataset. So, I have a couple links to my highly recommended ones. So, the candy power ranking from the ultimate Halloween candy power ranking story. This is nice because it’s essentially a CSV but you can click and reorder.

So, oftentimes if you’re going to do something by hand you want to know, you know, what was the most or the least or you want to filter and you can say like chocolate equals 1. And apply. And then just have the chocolate candies. So, this gives you a little bit of an interface to data without having to do any data work. So, that one is cool. If you want to do something heavy, this hate crimes dataset is, of course, really interesting and has a sort of spatial component which I always think you have to make a hand drawn map or are you going to make a bar chart or something like that. And then just because drugs are interesting. And I think this one might have been a Mona Chalabi story. I could be wrong. About the different age groups and the frequency that they use these different kinds of drugs.

So, what I’m saying is you can use any dataset. I gave you those three ideas because I think they are smallish and they’re possible to visualize by hand. You could use the lake data and dive in and make something really finished off. And so, just like like we’re just going to do it. Like I said, there’s tons of supplies up here. I have rulers if you want it to be more exact. I have more things and stencils and stickers.

I won’t mention the other items because I’ve said them several times. There’s colored construction paper. You could try working on something for a while and see how it turns out. And then what we’re going to do is have maybe 20 minutes of time to work and then we’ll hopefully share some stuff and do some takeaways. So, have at it.

[Working on visualizations]

Okay. As a time check, it’s 5:06. The session ends at 5:30 and I would love for us to be able to look at each other’s work. I’m going to check back in maybe ten minutes. See how things are going.

[Working on visualizations]

Yeah, let’s try and wrap up by 5:20 so we can look at things. So, like seven minutes.

[Working on visualizations]

AUDIENCE: Excuse me, has anyone in a group and kind of board and wants to write some copy for my candy personality test?

AMELIA: You have four minutes. You could write some good copy.

AUDIENCE: Could you add that to the Etherpad?

AMELIA: What? Oh, sure.

[Working on visualizations]

Okay. Two minutes and then we’re going to talk about what you guys made. And if you’re still working, maybe you’ll be one of the later people to talk.

All right. It is 5:20. I want us to look at some of the stuff you produced. And it’s okay if you’re still working. I’m going to bring one of these portable mics so maybe it will be easier for our transcriptionist to hear what you have to say. Who is bold and wants to go first and talk about what they were working on?


AMELIA: I don’t know if the mic is live.

AUDIENCE: This is the earth at three different times of the year. The solstice, showing Minneapolis, the sunset time and the angle of how much of the earth is dark at sunset. Here is summer where there’s like more sun in the north. And here’s September. And it’s equal. And here is December where the sunset is at 4:30. And it’s north.

AMELIA: Great, thank you. Who is bold enough to go next? Yeah.

AUDIENCE: I can’t lift mine up. But it is a chart of every not every it’s a chart of every episode over the course of 14 seasons of Gray’s Anatomy plotting out which episode is the deadliest episode for a character who has died.

AMELIA: And what’s the understand?

AUDIENCE: Unclear because I didn’t get the data really. But probably episode 10. 12.

AMELIA: Okay. Cool. Are you ready?

AUDIENCE: We made a bar chart. Unlike most bar charts, there’s no bars. It’s what you would find in your bar. Cocktails by volume and liquor. And garnishes as an additional variable.

AMELIA: I love it. Cool. Yeah? Let’s go.

AUDIENCE: So, I created one of the maps I originally scheduled out of the lakes here in Minneapolis. I have the water shed area on one axis, the area on the other and the size of the circles is the depth.


AUDIENCE: Sure, I’ll take the mic. I visualized the I did eight different teas. And the brew time and how many tablespoons of tea should be in each cup and the brew temperature. I regret some of the design and color choices. There’s lots of variables.

AMELIA: Yeah. Wow. That’s got a lot of information. Cool. Do you want to talk? You don’t have to. You’re going pass. Anyone who is bold? Who wants to okay. Yeah. I have been watching this one.

AUDIENCE: I just decided to show the area and depth of the Minnesota lakes with wire and made them into bowls.

AMELIA: Cool. Etsy project. Great. Anyone else? Bold? Yeah. Yeah.

AUDIENCE: So, I have been working on a story about New York City crash data and where the projects have been according to the crash data. This is highly inaccurate. But we used pom poms. And theoretically the sites would be how many people died in the intersection and the string represents the roads.

AMELIA: Cool. Thank you. Okay.

AUDIENCE: I have created a personality test tree. What candy are you? And that you would be willing to submit to a personality test. I only have all right. All right? Are you sweet or more low key?

AMELIA: Sweet.

AUDIENCE: Worth every penny or a bargain?

AMELIA: Worth every penny.

AUDIENCE: Crowd or niche.

AMELIA: Niche.

AUDIENCE: You are Sugar Babies, Sweet Tarts or Walkers.

AMELIA: I hate all of those. Cool. Thanks. Yeah. So, this is just a took some data from up to 21. The colors are different drugs. Green obviously marijuana. Yellow is alcohol. Blue is what was blue? It was cocaine. So, each quarter is just five. And the percentage in one year. Percentage of those in the age group that used it.

AMELIA: It’s like a pie chart up there. A bar chart?

AUDIENCE: More bar. Yeah, pieces. Can have AMELIA: Okay. A bar with pieces.

AUDIENCE: Compound. We like stickers.

AMELIA: Yeah. It looked great. Anyone else? Yeah? Okay. So, no mic.

AUDIENCE: I visualized all of the places that I lived by order. But also how much I liked them. Why I moved there and whether or not it was temporary or permanent. Temporary being like I was only subleasing someplace for a couple months. Yeah.

AMELIA: Great. Yeah.

AUDIENCE: I have a much less successful version of the lakes area and depth. I so, I like to cut out these are actually two not to scale. These aren’t correctly finished. Oh, but then I ran into the trouble of if I was just layering blue construction paper on top of each other, you wouldn’t be able to see the individual lakes. And so, I’m using pom poms to like pop them out from one another. But it’s very weird.

AMELIA: Weird is good.

AUDIENCE: Similar concept. I traced here. Let me hold it up. I traced all the lakes to the same scale and their actual shape. So, they’re interesting shapes. Like this one is shaped to like this. Whereas these ones are nice and round. And then I was trying to stick them on wire at the appropriate heights so you can kind of like a 3D scatterplot with the shapes of the lakes.

AMELIA: Cool. Anyone else we didn’t talk to who wants to talk. Yep. Okay.

AUDIENCE: So normally I get more visual with this. But we didn’t have a bunch of candies laying around. However, I am severely allergic to peanuts and I looked through the chart and looked for the cheapest ways to kill a person, Reese’s miniatures, Boston baked beans and peanut M & Ms. And most enjoyable, Reese’s peanut butter cups and snickers. And the best value, miniatures and peanut butter cups and snickers as well.

AMELIA: Cool. Anyone else want to talk about the work done? I wanted us to talk about takeaways from this session and I have two minutes. So, if anyone has anything that they’re going to takeaway, that would be great to hear. And if the answer is, like, something fun on Friday afternoon, I’m delighted by that. Yeah.

AUDIENCE: My biggest takeaway was actually using like your hands to make things helps you be more creative. Because I think when you approach visualizing just from your computer you immediately start thinking about what’s easiest to do. And then you the subset of things that you start exploring immediately shrink.

AUDIENCE: To build on that, it’s like sort of the canvas metaphor if you have a blank canvas. But our computers are kind of like a canvas with only certain kinds of tools that we’re most familiar with or hear about from other people. And this was like all kinds of things.

AMELIA: Cool. Yeah?

AUDIENCE: Something that I do this normally. I keep colored pencils at my desk and often ask people to draw pictures of what they’re thinking about. And what I’ve learned is often it takes ten iterations on something before you actually get there. And usually the final product is usually really well thought out. Like color, size, all that stuff usually is in the process and everybody is super thrilled when they’re done with that process.

AMELIA: That’s great. Cool. All right. Well, thank you guys so much for coming to this weird session. I had fun. I hope you had fun. And I think that’s it.

[ Applause ]