Data Visualization Tips and Tricks

How can we deal with data better? How can we teach kids to deal with data better? Sarah Morris from the nonprofit Nuclear Learning Network has some answers. She educates students and the general population on better data visualization, or dataviz, and how to use modern tools for dataviz.

Related to this episode:

  • The Nucleus Learning Network: www.nucleuslearningnetwork.org/
  • Art Science Gallery in Austin, TX: www.artsciencegallery.com/
  • DataViz example at The Nucleus Learning Network: www.nucleuslearningnetwork.org/datavis/
  • Social Explorer (mapping tool): www.socialexplorer.com/
  • RAW (chart-building tool): rawgraphs.io/
  • The Law & Order Database: www.overthinkingit.com/2012/11/13/th…l-20-seasons/
  • Tips on using the Excel chart recommender: www.k2e.com/tech-update/tips/689-tips-using-recommended-charts-in-excel
  • Periodic Table of Visualization: www.visual-literacy.org/periodic_table/periodic_table.html
  • Chart Juice: labs.juiceanalytics.com/chartchooser/index.html
  • The Guardian data visualization: www.theguardian.com/technology/data-visualisation

Our closing music is from "Late for School" by Bleeptor, used under a Creative Commons Attribution License: creativecommons.org/licenses/by/4.0. Subscribe and find more podcast information at: www.k12engineering.net. The K12 Engineering Education Podcast is a production of Pios Labs: www.pioslabs.com.

Check out the book and ebook “Engineer’s Guide to Improv and Art Games” by Pius Wong, on Amazon, Apple iBooks, and other retailers: www.pioslabs.com/improv4design.html

TRANSCRIPT

Transcript of:

The K12 Engineering Education Podcast

Episode:

Data Visualization Tips and Tricks

Release Date:

3/6/17

 

[Pius Wong] Thank you to listener Victoria Villareal. She’s made this podcast possible by pledging to the show at the Engineer level at patreon.com/pioslabs. On behalf of everyone listening, thank you, Victoria.

[opening music fades in]

[Pius Wong]  It’s The K12 Engineering Education Podcast for March 6, 2017.

[Pius] Welcome to the podcast. I’m Pius Wong. The nature of data is evolving. We’re collecting more of it, and we’re getting more and more tools to try to understand it. To help make sense of how to teach this new data landscape to future engineers, I spoke to my guest this episode Sarah Morris.

[closing music fades out]

[Pius] Sarah Morris is the cofounder of an education nonprofit called The Nucleus Learning Network, where she specializes in teaching modern data visualization techniques, or dataviz for short. I met up with Sarah on the outdoor patio of Pacha café here in Austin to talk.

[Sarah Morris] So I’m a librarian at The University of Texas, and then as a side gig I run a nonprofit with a friend of mine, Emily Weerts. We founded it about a year ago, and we focus on educator training and just supporting local educators with a variety of topics, usually STEM education and digital literacy especially, which is a big interest of mine coming from the library world.

[Pius] Yeah, I was going to say, if you work at the university in the library, you must be used to lots and lots of data.

[Sarah] Absolutely. Yeah, I teach mostly classes for first-year students, and it’s a lot of digital literacy, information literacy, dealing with information, how to find information and use it effectively. So we’re just trying to empower them to succeed at the university, and then whatever their next endeavor is going to be once they graduate.

[Pius] Cool. And what do you do then with data and data visualization at The Nucleus Learning Network?

[Sarah] We offer workshops for educators. We’ve done a couple at the Art Science Gallery, which is a cool organization. Drop-in workshops, where we can have educators come, but also people in industry, artists. Just a whole array of individuals coming in. And that work sort of stemmed from work I’d done in the library world, so at UT, and previously at Loyola University in Chicago where I was for a couple of years.

[Pius] Awesome. I’m from Chicago.

[Sarah] Fantastic. Hey, connection. It was a good time. Yeah, so when I was at Loyola, I started teaching infographic data visualization workshops, because a lot of professors we were working with were starting to include that as class projects that they could do, just as alternatives to the traditional research paper or a supplement to a research paper. I just thought it was such a fun topic, and it’s a great way to pull in a lot of concepts, visual literacy, dealing with data, communicating ideas in different media formats. And it kind of snowballed from there. So when we started Nucleus, it just seemed like a nice fit and a good segue into being able to work with educators in that capacity.

[Pius] Very cool. And for anyone who doesn’t know, then, what is an infographic, and what are the other big terminology that people should know?

[Sarah] Absolutely. So an infographic is a type of data visualization, and it relies on graphics and a lot of images. They’ve become very popular in a lot of media outlets now. So you’ll see them in the New York Times, the Wallstreet Journal, The Guardian, wherever people are having a graphical representation to talk about either a lot of data or a lot of information they want to cover. Data visualization is kind of a catch-all term. It can be used in a lot of different ways. Basically it can refer to everything from a map that’s looking at information to a basic chart or graph that you could even do in Excel. That’s still a form of visualizing data. So there’s a lot of different outlets that can take, and an infographic is just sort of one type of that, that really relies heavily on imagery and, you know, color, and kind of almost poster-type design, to present a topic that you want to cover.

[Pius] And so a lot of the people listening are involved in STEM education in particular, and engineering education, hence the name of this podcast. And so they deal with a ton of data, as well. It might be different data from what The New York Times deals with. So what kinds of data visualizations do you think are more unique for those types of fields?

[Sarah] Yeah, that’s a really good question. I think the fun thing is you see a lot of overlap in different fields, even. I’ve seen recently, a lot of people are doing maps for the election in the fall. Everyone is doing polling data and that kind of thing. But I’ve also seen maps where people were maybe looking at earthquake data and plotting it on a map, or historic maps where people have looked at, you know, where Enlightenment philosophers were living and sending letters from, and people were mapping the trip of the letters and how they were like traveling around and connecting with people. So yeah, you can use the same form for a lot of different fields, which I think can be really fun for engaging with students, because no matter how they’re coming at it, they’ll find something of interest.

[musical interlude]

[Pius] And so Sarah, I understand that you have different strategies in the classroom that you could probably pass on to other teachers. What are they?

[Sarah] Yeah, definitely. I’ve been fortunate to have experience working kind of different ranges of experience with dataviz. Sometimes I’ll have students who have more experience, and it becomes more, just: let’s go through different tools together, let’s have some hands-on time, you can explore different stuff. And I’ve definitely had students who have no idea what the term even means and are just completely from scratch, which will probably be the case in high school. So I often start showing different examples to let people see kind of what’s out there and sort of talk through what’s happening, what seems to be working, what doesn’t seem to be working. And then I will often do activities where I’ll just even give them a really basic sample data set and have them work in pairs to think about how they would represent that information. And this can even be done on pencil, paper, or on a whiteboard, wherever, and I think it’s always fun to see what the different groups come up with, because everyone will have wildly different ways of doing something, and you can sort of show them the actual graph or chart that I got it from at the end. OK, that’s how they chose to view it, but you’re also exploring different avenues there. One other thing is, I like storyboarding, as well. So having them think, if you have a guided project you’re doing, think about who is the audience for this visualization. Who are you communicating with? What’s your main point? What data do you want to make sure is included? And kind of plan that out before you dive in. I think with especially infographics, you can have the tendency to: I’m going to dump everything I’ve ever thought into this thing, and it becomes kind of incoherent and a little overwhelming.

[Pius] So teachers should do a little planning before they just throw a bunch of data at their kids.

[Sarah] I would definitely recommend that.

[musical interlude]

[Pius] I know for a fact that the engineering teachers I’ve worked with, they use Excel a lot. Some students love it. Some students do not.

[Sarah] I’ve gotten mixed reactions from students, but I am a proponent of good old Excel.

[Pius] If you are guiding someone who doesn’t like these tools, what do you do in teaching them?

[Sarah] I think I just try to find some element that they’re going to enjoy, whether it’s – and there are so many different tools out there. I think, sometimes, Excel might seem sort of basic, because there are a lot of new flashy things that have come out in recent years, but I always just tell them, find the tool that fits what you want to do, and just because it does exist doesn’t means that you have to use it. Kind of like PowerPoint back in the day, when you could animate the letters and have them all spin on the screen. You don’t necessarily have to do that every single slide. You can use that judiciously. Yeah, figure out what your point is and what you’re trying to do, and finding a tool that fits, I think can help people sometimes. Because if the tool is going to fit, if you know what you’re doing, what story you’re trying to tell with your data, then it kind of can flow a little better, and you’re not sort of forcing something, or it doesn’t become this awkward fit.

[Pius] Sure, like there’s a natural reason for them to want to use the tool.

[Sarah] Exactly, yeah. At least in workshops that I’ve done working with educators, we try to cover a wide variety of tools that can work in a lot of different scenarios, and I think sometimes students are just as excited to see the different options that are out there, whether it’s making a nice-looking chart in Excel to maybe finding something where you can make a map, or playing with Google Maps, or whatever it might be, just to have fun with it.

[Pius] So can you talk a little bit about the content of the workshops that you do related to this? What kind of tools do you specifically go over, and how do you actually teach someone this?

[Sarah] Yeah. That’s a really good question. At least with student workshops that I’ve done – so for college freshmen, teens, age-range typically – we’ll sort of break it down by type of tool, often. We’ll maybe pick one sort of mapping tool that we like and demonstrate it, and then give them time to do hands-on exploration, which is I find very important, to let them bang around and explore. The nice thing is, a lot of tools that have come out, for instance a mapping one that I like is called Social Explorer. They pull US Census data, and they have a lot of built-in stuff. So data sets are already in there, and it walks through what you need to do. It can be a good way to learn how to do things and see what your options are in a more guided environment. There’s a tool called RAW, R-A-W, and it’s another sort of chart-building tool, but they have sample data sets within the environment. So you can play with one of theirs and just get used to things and see what types of visualizations you can do. So that can often be a way I’ll approach this. Demonstrating and letting them get some hands-on experience with whatever type of tool you might be interested in having them learn.

[Pius] I see. So it sounds like you use a lot of existing big data sets that are out there. And so, that’s funny, I use the phrase “big data” because a lot of computer science teachers have told me that’s a big, big trend now. What do you think about the idea of big data? Should the average person really care about that?

[Sarah] I do think the average person should care about it, because it’s becoming such a big thing, so to speak, pun intended. I think it’s becoming ubiquitous in a way. There’s so much heightened access to data sets. There are so many people putting free data sets out there. Even the government, the state of Texas has a data portal. The city of Austin has a data portal where you can just grab a data set and go to town with it, which is really cool, but I think it’s becoming more readily accessible. You’re seeing news outlets engaging with it and putting that out there. Scientists who are making their stuff widely available are making use of it. So I think it’s good to understand, you know, what it is and where it’s coming from and how people are engaging with it and the challenges that come from working with it. Even if it’s just a more superficial level, I think everybody, even if you’re not a scientist, engineer, can still benefit from knowing what that is.

[Pius] So you remind me of some work I was doing with The University of Texas a little while ago, as well, where we were trying to have a one-week module about big data. And it struck me that it’s really hard to find engineering-related data sets that are, like, real and relevant to kids, like something that kids would care about, rather than, I don’t know, financial stock market data. So what are your suggestions of maybe some more interesting data sets out there that teachers could play with?

[Sarah] Yeah. That is a really good question, because I’ve run into that, too, because half the time the free ones are just weather data, or something that’s kind of a snoozefest. So I’ve kind of set out to find borderline ludicrous data sets you can use. A couple – I’ve found one that was the box office returns for all the Fast and Furious movies in the franchise. It was a nice little Excel spreadsheet. It had the box office returns, and you could compare all the movies. It was simple enough that you could easily build a chart or graph with it. I had some kids once that made an infographic out of it, and it had little car icons. I mean, it’s kind of ridiculous, but it was basic enough that you could play with it and not have to worry about cleaning it up or that it was too overwhelming. There’s this resource called The Law and Order database that overthinkingit.com put together. I hope it’s still up. This was a while ago. I think it’s still around, but it is every episode of Law and Order and the verdict of the case. So whether it was guilty or not guilty, but it was just a massive spreadsheet of Law and Order episodes that you can play with.

[Pius] Are you a fan?

[Sarah] I am a fan of Law and Order. Yeah, I was trying to find more pop culture, just sort of more silly ones that let you play with it and aren’t too overwhelming. Because you’re right. Kids don’t get too engaged with stock market data necessarily. So I had a friend of mine who found the Audubon Society had some sort of bird data set you could use, so yeah, there’s sort of hopefully accessible topics or subjects that you can just use to play.

[Pius] What you’re saying reminds me – I don’t know if I was poking around on the Nucleus Learning website or something else, but I saw a data set of someone looking at Star Trek episodes, and whenever there was like a red alert or something or a yellow alert – That was your website?

[Sarah] That was our website, yeah it was. The same people who did the Law and Order database also did Star Trek – the Next Generation I should say. They tracked the number of red alerts and yellow alerts per episode, and they actually charted them by season, so by Season 4, they had more red alerts than yellow alerts for some reason.

[Pius] That’s cool, because you can correlate that with, did the ratings go up or down? I think that’s what you said for the Law and Order seasons. That’s what you were doing.

[Sarah] Exactly. It’s kind of just a basic environment where you can teach some basic statistics concepts or look at cleaning up data, making sure everything is formatted correctly.

[musical interlude]

[Pius] Then when it comes to the actual visualization and understanding and parsing of all that data, what are some of the, like, top five, top ten things that you think that a kid, or an adult, really, should know how to do? Is it formatting a chart, or what is it?

[Sarah] A few things. One on the end of producing it yourself, and it’s definitely just – I think Excel basics can go a long way, even as a little refresher for people, just making sure if you get away from Excel for a while and you get back to it – how do I do this again? What do I do? Just making sure the kids are, for what you want to do in a classroom, making sure everyone is on the same page. Like if you want to make sure, hey, do we all understand how to convert these two percentages? Or do you know how to enter this formula so it adds the row up for you? Just making sure everyone’s cool with that level of stuff. Yeah, and I like – Excel has this sort of chart recommender now that’s built in to Excel. So even when you first format it, having them run the chart recommender just to make sure you’re not getting anything wacky when you produce a chart. Because sometimes you might have it in there, and you think it looks fine, and the x and y axis got flipped around without you realizing it. Oh nevermind, that’s not what I intended. So it’s a maybe cool test before you move into something else. Yeah, so just some basic Excel stuff and making sure it’s OK. Because the thing I always tell people is, there are all these great tools out there you can use, like free things online, whatever, but the tool isn’t that smart at the end of the day. Whatever you put in, it’s just going to spit something back out to you, and it will not correct something if you have something wacky – even if it’s something as simple as you forgot to label something, it’s not going to realize it should label it for you. Just be careful of that. The other thing I really emphasize, at least in my workshops of my students, is the visual literacy component. Being able to read a data visualization is something I think can go a long way. I often just do examples at the top, both good and really horrible examples, and just – we talk through it as a group. What is going on here? Because, just to make sure we’re keeping best practices in mind, like how you want to represent something. I’ll show them deliberately awful examples. Maybe someone used a pie chart unnecessarily. OK, that doesn’t make sense. How could you represent this in a way that makes more sense to people? And you sort of problem solve the dataviz. I use real-world examples with them, too. I was at this conference once, and I won’t name names, but somebody had a sort of infographic they were presenting, and the title said one thing, and then something – The first piece of information right under the title was basically the opposite, and so everyone got confused as to what – It was a tiny thing, but it flipped the whole thing around, and people thought they were trying to say A with their infographic, but really B was their point, so to speak. So it caused confusion at the end of the day. So even just your header at the top, just make sure you’re clear about what you want people to get out of something, because it can get really misinterpreted if you’re not careful.

[musical interlude]

[Pius] In your experience, is there a particular topic related to data visualization that’s really hard or the hardest for people?

[Sarah] Honestly, I feel the hardest, sometimes, is – back to what I was saying earlier, there are so many options out there. People want to get really elaborate and do a lot of stuff, and sometimes a simple solution can be just as well, and that can be hard for people to kind of grapple with, sometimes. Because sometimes the more elaborate solution might be muddying the waters or confusing your issue, and you could have unintended – unintended what it’s saying. So yeah, that can be tricky as we’re working with kids on the outset. You know, select the best visualization option for what they need to do, and making sure that’s all clear.

[Pius] OK, that’s a good tip for our teachers. Hopefully they’ll teach it to their high school kids, so that by the time they get to you, they won’t have as many of those issues.

[Sarah] There’s also a wealth of resources out there, best practices, even tools where you can say the type of data you have – like you want to compare two things, and it says OK, these types of visualizations help you compare two things. So you can kind of start with that.

[Pius] What is that? I’ve never heard of that.

[Sarah] There’s one that’s done. It’s a Periodic Table of Visualization, and you can click on different types of charts, and it breaks down what it is. It says, this is a scatter-plot. This is when you might want to use this. And there’s one called Chart Juice, I believe, and it’s also another comparison one, and that’s also linked on the website.

[musical interlude]

[Pius] What’s the future of data visualization? Where is it going? Like, is it going to virtual reality? Is it going to 3D? What’s going on?

[Sarah] I hope it goes to virtual reality. I think that’d be kind of cool. Yeah, I feel like it’s just going to keep growing and becoming more and more of a communication method that’s being used. I think especially with how we communicate online, it’s become so hyper-visual. You can even see – I always thought it was interesting. We have the LiveJournal era and blogging, and now we’ve more of Tumblr. Oh yeah, I rocked the LiveJournal. I’m dating myself. But no, you know, Tumblr, Imgur. It’s so visual-heavy, the way we communicate. Emojis and everything have become the sort of shorthand. So I see even major news outlets – The Guardian has its own dataviz department. They actually have a blog where they talk about how they create their own visualizations and all this stuff. I feel like it’s going to keep growing, and it’s great for students whether they go to create them or not to be able to consume them and understand what they’re seeing and how the data is being used and where it’s coming from, because I just think it’s going to become increasingly a way that we share info. Especially with big data growing, as well. So we’re going to have to figure out a way to process that and make it understandable.

[musical interlude]

[Sarah] Especially within the sciences, computer science, being able to communicate out clearly what you’re working on and what you have and be able to share that with a wide audience is such an important skill that we see, and I think that’s just one tool in their toolbox. And I like dataviz, too, because I feel it connects really with the humanities, as well. It’s so cross-disciplinary. So, you know, you can have artists consulting on a project with you, or engineers providing the data, how you communicate that out to look appealing. Yeah, there’s a lot of different angles you can approach it from.

[Pius] That’s really awesome. So it’s clear that you love data visualization, Sarah. I’m wondering if teachers or anyone listening, professionals, want to know what you do, what your tips and tricks are, how can they reach you?

[Sarah] The best way to get in touch and see how resources are compiled is our website: www.nucleuslearningnetwork.org. We’ll make sure the link is up on the site, and on the show notes, so people can get there. Our contact email is on there, and if anyone wants to get in touch and chat dataviz, feel free to fire off an email. I’d love to hear from you.

[Pius] All right. Thank you so much, Sarah. I really appreciate your time.

[Sarah] Thank you so much. This was a lot of fun.

[closing music fades in]

[Pius] That was Sarah Morris of the Nucleus Learning Network. Links to the resources that we mention in this episode are in the show notes and on the podcast website: k12engineering.net. Don’t forget that the transcripts are coming out for Season 1 episodes, and you can find them linked on the podcast website. As always, please subscribe on iTunes, Stitcher, or your favorite podcast player. Share every episode your friends and colleagues, and follow the show on Facebook and Twitter. Thanks for listening, and tune in next time for more.

[Pius] Our closing music is from “Late for School” by Bleeptor, under a Creative Commons Attribution License. The K12 Engineering Education Podcast is a production of my production studio Pios Labs, and you can support Pios Labs at patreon.com/pioslabs.

[closing music fades out]

[chime]

[Pius] Hey, it’s the post-show message from Pius, your host. I just wanted to let you know that I am going to be at South by Southwest this March, and it’s the first I’m ever podcasting from it, but that’s what I’m going to do. So if you can’t make it, or even if you can make it, and you’re interested in seeing or hearing ideas from different parts of the conference, well, stay tuned for The K12 Engineering Education Podcast.

[Pius] And like I mentioned in the last episode, again, I have a new book out called Engineer’s Guide to Improv and Art Games, and because you’re listening, if you’d like 20% off on the ebook, you can go to smashwords.com, search for the book title or my name, and when you check out the book, use the promo code EH22M, and that’ll give you 20% off. That code will be good through the summertime. Thanks.

[chime]


EPISODES