This video is an interview with Ben Jones who started a company called Data Literacy focused on data literacy training. Learn about what he thinks is the biggest challenge we face in terms of data literacy and how solving it will make the world a better place.
- So I'm very happy now to welcome Ben Jones to the microphone today and Ben spent a bunch of years, six years I believe, at Tableau Software, where he was doing marketing and outreach and evangelism and he left the company in 2019 to start his own new company, funnily enough called Data Literacy LLC, which is a very good thing because we're here to talk about data literacy today, so Ben, welcome, thank you very much for joining me to have this conversation. - Thanks Bill, yeah great to be on the show here with you. - So we are talking about data literacy, I know you spent a lot of time at Tableau doing, as I said, marketing, outreach, evangelism, which means trying to bring software to people, to get that product to people, but I know it's more than that, I know that you did things like events like the fantastic Tapestry Conference that I love and went to a bunch of times. And other things, which really goes beyond marketing software, it's really fostering a community around data visualization and communications, et cetera, and so it feels to me like a very natural move to be talking and focusing on data literacy, which is sort of this broader, bigger topic, which is, you know, very much at the heart of that community. So very long way of getting to, what made you passionate about data literacy and made you want to spend the next however many years thinking and focusing on that subject? - Yeah, great question, so I guess there's really a few things that I can list. Well, number one, Tableau was having lead the Tableau Public Platform, which is their free offering for students or journalists, bloggers, anyone all around the world who's telling a data story online and so that's why the community aspect came into things, that's why, you know, training people to be able to, you know, use data and create interactive and engaging data stories was sort of what I was all about there and so then, also at the same time, I was teaching at the University of Washington. I had written a book from O'Reilly called Communicating Data with Tableau and because of that I was able to start teaching a class at the University of Washington and just sort of got hooked on that experience that you see someone have where the light bulb goes on, they have that eureka moment and they realize, you know, they're actually capable of doing a lot more than they thought they ever were with data. And so, I started to realize also, that I would say the third thing that came about in my, especially in the last year or two when I was at Tableau is that I started to hear a lot of people say that as amazing as the software is, and it really is, there are still a lot of people that don't really feel like that's sort of their language, it's still maybe, while it may be easy for me to make a box plot or a histogram, there's a huge percentage of the people out there that, you know, really don't feel comfortable reading and interpreting a graphic like that and so, I realized that there was just this massive education gap and it's kind of a opportunity of a lifetime would be able to help people get there and so that's what I've been up to since then. And but yeah, in a way, it very much is a natural transition from the sorts of things that Tableau was already having me do. - Yeah, that's great. So it's funny, I did a workshop actually focusing on data literacy and analytics for a big federal agency and the overwhelming response was they are so hungry to learn this stuff and these are people who work with data, this is their job, but they're not data analysts, but they're working with data all the time like we all do and yeah, and it still seems like there's a huge gap for sure. So you wrote, sorry go ahead. - Well, obviously not very many of us got a formal education that prepared us for the sorts of things that everyday jobs, not even an analyst job, is requiring us to do now and so, you know, the question is how do we get there, how do we close that gap? And I think there's a lot of great innovation in the space right now, a lot of people who are spending their time and energy to try and help people get there and I'm happy to be apart of that. - Yeah, so you said you wrote a book, you also wrote I know an ebook called the 17 Key Traits of Data Literacy, which is really great, it's perfect, and I recommend people to download it and check it out. So if you could, maybe, let's start off with a little definition, what is data literacy? It's a very, you know, hot term right now, sort of in the zeitgeist, but a lot of people don't really know what that means, so if you could define that and then maybe also share some of those key traits with us. - Sure. Yeah, so like data science, or any other term, there might be different people with different ideas about what it means. To me, I actually really like the Wikipedia definition for data literacy, which today says you know, the ability to read, understand, create, and communicate data as information. So, read, understand, communicate, create. Those four things to me are what makes someone highly data literate. And so certainly, there's apart of it, which is just reading and interpreting visuals that other people have made, you know, there's a large component of it which is also related to being able to deal with raw data, convert it, but what I tried to do in that 17 Key Traits book, and it's a very quick read, it's only about 20 pages or so, is I tried to list other things like attitudes and behaviors that come along with data literate individuals. I tried to think back on my time in the business intelligence industry as well as working for a Fortune 100 company prior to that and as well as time in academia, asking myself what made someone highly data literate? And at the end of the day, it was also about their attitudes and behaviors, you know, things like are they inclusive when they think about data in terms of do they bring in other people into the conversation? Also, are they looking to continuously improve upon the state of data in their environment? It's not perfect, it's never perfect, but what can be done to continue to improve it? Someone who's highly data literate understands that there is an evolution about the data that they're working with and so they advocate for continuous improvement to that, as well as effectively using it, as well as being able to present to an audience what they found through a data discovery session. So those are a handful of, so I broke 'em up into those four categories I guess, 17 traits is a lot, but it's really about knowledge and skills, those are the hard skills people typically think of when they think of data literacy, but then also the attitudes and behaviors, those are the other two categories that traits fall into, just to be able to help someone really evaluate, you know, how much they're contributing to a data dialogue in their work, but not just in their work, it could be their community, it could be their own home, you know, finances or fitness or health, these are all aspects of our life in which data has a key role to play and so, in which, you know, data literacy can be effectively utilized I guess is a way to think about it. But yeah, it was a fun little, I really appreciate the four quotes that I got from some very smart individuals in this space, Alberto Cairo down at the University of Miami, R.J. Andrews who wrote the book Info We Trust recently, Giorgia Lupi, who is now at Pentagram, who just released a clothing line of data, as well as-- - I saw that. - Yeah it was really neat, I think it's just today or yesterday actually, Cheryl Phillips, who used to be the data editor here in Seattle at the Seattle Times where I live, she know teaches data journalism down at Stanford. Those were each individuals that gave me a quote for every one of those four sections and I really thought about who would be best to comment on those four aspects of data literacy. So, it's been a great way to just get the conversation going with people and again, like you said, you know, answer that first question, which is what does data literacy even mean? - Yeah, so I do, I love the four sections and I did particularly love the attitudes section because you know, yeah, we think about these hard skills and even behaviors, it's a thing you do, it's an action, and people do think of data literacy as being oh, I need to work with data, I need to do stuff with data and it's definitely part of that, but I love that section because being confident, alert, and ethical are really important, you have to do that to be data literate, to do well, especially to have integrity, which of course we want to do, especially when we work with data 'cause people think data is truth and of course we know that it can be manipulated and you can do bad things with it if you're not careful. So, I also really love the fact that you, and you just touched on it a little bit, that data literacy is about far more than creating charts and graphs and doing data visualization, it's about communications, absolutely, it's also even about the ability to read and understand data, which people forget and most people, all people will have data that they have to deal with like their social security, you know, details, whatever, all these things, mortgages, and we're not as data literate as we need to be. So, as a good segue from there, I want to just ask you why is this so important, you know, and I think the answer's fairly obvious based on what I just said, but go beyond that, like what makes this so important and where, how data illiterate are we generally as a society do you think? - So two parts to that question, one, how important is it? I mean now, more than ever, and it's only getting more and more important. Why? Because there's more and more data available, the tools at our disposal allow us to access and make use of that data like never before. I think about when I was a kid, I grew up in Southern California. In the early '90s, there was a huge earth quake that woke me up at 4:30 a.m. and I woke up, you know, out there on the lawn with my family, waiting for the shaking to stop and everything. So back then, if I wanted to study earthquake data and understand the nature of earthquakes and what not, I don't know what I would've done. I maybe would've gone to my hometown library in Thousand Oaks and I would've tried to find a book with tables in the back maybe or I would've looked a phone number and tried to get a hold of somebody in the government agency to get a hold of some kind of tables, graphs, but you can go to USGS right now, the United States Geological Survey website, and you can download over 100 years of earthquake data. Where it occurred, the magnitude, you know, all of those aspects of this rich history and so, you know, well back in the '90s maybe I would say, you know, my opportunities as a citizen to try and understand more about earthquakes in my part of the world was very limited, but now it's massive and it's wide open. And so, to the degree that I can, you know, leverage that and make use of that, now I give myself an advantage to be able to understand my environment in ways that maybe I could never before, or it would've been very difficult before, and which, unfortunately, what I'm finding is, you know, very few people still can do that. And that was something that surprised me, I got to be honest, as I started to kind of move into an evangelism role at Tableau, was really coming to the understanding that there were still so many people feeling like they're completely left out, that they really don't have the basic skills and tools they need to, you know, and they kind of almost like feel like this data revolution is passing them by. And people are afraid about that, you know, and they look and they say well, they see some kind of dashboard and their first reaction is I don't know what this is telling me and then they start to feel afraid about that and they start to feel like there's something wrong with them, like they're, you know, they don't have the necessary intelligence, and they do, it's just a question of being taught how to look more closely, being taught for what sorts of things to look for, what kind of questions to ask. What does it mean when we see data expressed in a certain way and how does our brain interpret that? And it's almost just becoming more aware of themselves and how they process information visually and so yeah, I mean that was not part of the curriculum growing up, it just wasn't. You know, even in, I did an Engineering degree at UCLA, wasn't part of the curriculum you know. So, I think it's starting to become more and more common in universities, I think Berkeley a couple years ago or maybe last year just launched a data science undergraduate program, so these sorts of, and also when I was at Tableau, I had a chance to lead the academic programs team, which was also about, you know, helping professors, teachers, as well as students make use of data in the classroom and we just started to see this explosion of data programs in all kinds of schools, not just CS or business, also, you know, science, arts schools some times, were teaching data design and infographics and so, I think we're seeing a massive wave right now and this desire to try to get to a place where we are all so competent with data and I think it's really important because how many of the world's problems do we need to think about the data angle? You know, whether it's about we're talking about major issues in society or about the planet we live on, I mean these have a data drive component to the solution and to the degree that we can learn those skills and apply them, I think we have a chance to help build a better tomorrow. I think it's apart of the equation, it's not the only part of the equation, but I think it's a big part of it. - Yeah, you know, it's interesting, one of Tableau's competitors, Click, did a survey, I think about a year ago, within the past year I believe, and it asked essentially, are you data literate? And, 20, only 24% of decision makers in business define themselves as data literate. So 76% of decision makers, these are people higher up in the food chain, don't think they're data literate and that was shocking to me. So taking that as the reality and people are struggling and people do feel less than, I think that's a great point. What are some sort of the low hanging fruit to improve, in addition to obviously hiring new, or somebody else to teach them in you know a workshop, or read your book, whatever the case may be, what are one or two tips or tricks you can share with them now that might help them sort of get a leg up? - Yeah, I think it's looking for data that matters to you, I think that that's helpful and a good starting point, whether that's something that you're very closely tracking like maybe a fitness goal or at work you are a team and your team has a metric to hit, and it's really focusing there because it's already in an area of your life that you care about and so you're more likely to pay close attention and that's going to be also part of something you get an immediate benefit out of as opposed to something maybe more abstract. But, that being said, I do think there are a lot of amazing online resources today to get better. You know, part of the reason why I went on such a a binge of working with data was because Tableau Public was a free platform, you know, and I was able to download that and get started and try my own fun little pet projects and whether you want to watch videos and that's your way of learning or you want to sign up for an online class and get involved with a group that's also learning like you, that's another great way. I think it's just, you know, be clear with yourself about what your learning style is and then seek out opportunities within that sort of, that set of learning kind of experiences that you want to choose from. And yeah, it's just question of you know, also just looking closely, right, be very diligent about looking closely at charts and graphs and looking more than just what you at first see, but look beyond that, learn to see very clearly what's there. So part of it is just learning about yourself and what you do. There's some great books out there now, I mentioned Alberto Cairo, he has a book called How Charts Lie. The whole first chapter is dedicated to well, how does it work well, how do charts tell the truth, you know? Oh, or R.J. Andrews in his book Info We Trust, if you're a book person, you can read those books, I think that's a great way. Also, Storytelling with Data by Cole Knaflic is a great book for someone who's thinking about it from the lens of design and how to focus an audience's attention. So there's just a wealth of resources. Getting involved in a community, you know, whether it's a data community, if it's centered around a tool, with some amazing meet up groups like Women in Data that are promoting, you know, this field, data visualization as a discipline to women, to girls in STEM and trying to get that kind of gender gap closed there, you know, so there's some groups if that's your thing, and you're willing to get involved with a group of people. Those are a handful of ways, I think that, you know, it's a sport where the more you participate, the better you get and so, it isn't really a sideline kind of a sport, I think you got to get involved and I know when I first got involved, I wrote a blog, I started putting my whole little thoughts and tutorials out there 'cause I wanted to remember things and that was a good way to do that, but that was a way that, you know, and there's a lot of fear associated with that first, you know, step of putting yourself out there. I think that those who have done that have found that there's just a very welcoming group of people that are not going to bite your head off, you know, they're going to welcome you, they're going to give you some feedback if you ask for it, they'll take their own time and do that, so I'm a big believer ever since, you know, kind of I got welcomed into the Tableau community that that's a way to build your own skills and to meet other people who are going to round out your skills because they do things maybe that you don't do. So, those are a handful of tips that I think about, yeah. - That's great. I wonder if in your travels, either during Tableau days or more recently, and since you're doing data literacy training, whether you've come across any organizations who were struggling with this and then you were able to see that transition happen, and some success happen, and anything you can share about maybe what brought that success to them. Either, you know, it can be on a smaller level, just a team within an organization or an entire company, whatever you've seen. - Yeah, so I'll think of a actually a government agency group, there was a small team within that and a time they were able to really just, you know, bring to the table lots of different skill levels, so everything from a data scientist to the person that, you know, really wasn't even comfortable say like, for example, you know, using Excel and just getting in the same room, talking about data, having a conversation about the challenges, I think those can be helpful experiences where you get on the same page about what about your environment today is helping versus hindering, you making progress or strides with the data that you have, and then working through some examples together using your own data, using data that's really similar to what it is that you're working with and trying to accomplish. And then, you know, again, it's really about, and so much of it is outside of the realm of skill building and it's about the organizational road blocks. Maybe there's not access to tools, maybe some of the data isn't available to certain individuals who need it, and then being able to have that conversation. So, you know, that's an organization I've just really starting to work with recently, so can't really talk about the success story about that yet as I think it's still a work in progress there. But that's every group I've noticed, really, you know, even Fortune 100 companies who are realizing they need a road map now for how to actually unlock, you know, growth based on the data that they've built over time, the tools they've paid to acquire, and so you know, what's their plan to actually make use of that, what's their plan to build a community around that, to incentivize people to do that, to actually, you know, share their knowledge, so that it's institutional as opposed to just maybe a one off? There's a new class form of products coming out now called data catalogs, which I think are useful for curating high value data sets, you know, having people add comments to them or flag them so that people can perhaps be alert to issues that they may have discovered with a given variable or field or a table somewhere. And just institutionalizing that whole, you know, process of taking data, turning it to information, turning it to knowledge, and then finally, wisdom. That whole flow is something that, you know, we kind of have to kind of figure out how to make that something we all contribute to, as opposed to pockets of success where someone might be advanced and off and running whereas there might be another corner of the team and the business that really just doesn't even feel like they can get out of the starting blocks. So what's the way you can kind of help those different groups to communicate with each other in a way that, you know, lifts the boats. So, I think that there's a lot of, every organization now, no matter how big or small they are, those are the sorts of things they're looking at from my experience. I mean, I'm leading a business of one, and so, same thing right? How am I trying to make use of the data that I have, what are my goals, how do I link them together to data that I'm actually putting to use for me as opposed to just having it compile somewhere and not actually tapping into the value that it contains. - Yeah. I spoke with Diana Yoo who's the head of data visualization at Capital One. In the conversation, we were talking about this one phrase, the democratization of data, which is exactly what you're talking about, it's everyone needs access to data, everyone needs the tools to work with data at whatever level they're at, and everyone needs access to the insights that the data has created, that someone has sort of pulled out and teased out of the data. And so yeah, data literacy is at the core of making that all possible. I think something else you said really resonated with me earlier, you know, and I can't remember exactly how you phrased it, but you know, you were talking about the ability to read data and understand data and how critical it is and how, you know, many of us struggle with that and for some reason, it triggered another interview that I had, a thought in my mind, when creating data visualizations, especially people who like hey I'm in Excel, I just need to make a chart and click, I made a chart, the software just makes it so easy to do that, but people just aren't thinking when they're doing that and you know, you made the point that when you see a chart, you're not really thinking. You just see it and hopefully you get, maybe you don't, if you don't, you probably just walk away and forget about it. And the phrase that I always use that when you're selecting charts and making charts, it's all about intentionality. Make a decision and do it for a reason and if you're intentional for all of the decisions you make in a chart, whether it's the scale, the color, whatever it is, as long as you're being intentional and doing it for a reason, then it's going to be better if you don't. And, the same is true, and what I think I hear you saying, when reading data. Be intentional, look at it, think about it, judge it, make decisions in your mind about what you're going to take away from it, and don't just let that immediate response be the only thing that happens, and especially if it's going to make you turn and walk away 'cause you're too intimidated by it. Is that a fair sort of summary? - Yeah, and that's what I love about that first chapter I mentioned in Alberto's book, 'cause he talks about like steps to take when you first encounter a graphic, like what is the axis say, is there an axis in there? Is there a legend that's telling you different colors, what's the legend? You know, kind of looking at those details, is there a title, are there annotations in there that are calling your attention to one or more data points? Is there a data source called out, do you know where this data is coming from, or is that completely mysterious? You know, and if so, maybe that should raise a question in your mind about how valid the data is. So, those are all things that I think we need to, in general, all of us need to learn to watch out for when it comes to reading and interpreting charts and graphs. That is that aspect of looking more closely. I think it is good to see something and have a knee jerk reaction, and then it is good to then go a step below that and say all right, what is that initial reaction I just had, is that something I can trust, or is there perhaps an issue with that? And so learning to, almost like look at yourself. Are you evaluating a chart? Then you're evaluating yourself evaluate the chart, it's like oh my gosh, that's so much evaluation, but it's actually good though, you know, you kind of see things and see what you're looking at and then learn to be aware of the sorts of things that you're seeing. And I think also, you know, when it comes, as you mentioned, when it comes to analyzing data, often times that first question we get, well what if we didn't even maybe profile the data set to begin with, maybe we ran ahead and answered a question that we had in our mind, but we didn't even realize the timeframe that the data applies to or maybe how it was measured. I have a funny little story about that when I was giving a presentation at a luncheon by a bridge over here in Seattle, and there's a little counter on the bridge that was designed to count how many bicycles cross the bridge, and so, long story short, at the luncheon, we're showing the data over time and there's this huge spike in the timeline and so, we all had this interesting idea, like well, where did this come from, you know, and people are saying, well, maybe it was a race, or maybe it was bike to work day, or I don't know, maybe there's a bike club in town or something. And so those were all the ideas we had about this big spike in the data as being all related to some massive amount of bicycles that actually crossed the bridge and then about 20, we didn't know so we moved on, you know, we just like moved on with the presentation, I had no idea, they had no idea. And maybe about 15, 20 minutes later, someone raises their hand, they're waving their phone in the back of the room and they said, I found out, I found out what it was. They said, oh, you know, tell us, what was it? And he said, it turns out the bicycle counter on the bridge malfunctioned that day and for some reason, the battery was low, they replaced the battery, and then the counts went back to normal, but in any case, nobody in the room thought to say well maybe this isn't real, maybe this is some glitch or something, you know, so those are the sorts of things I think and all about being alert there, which is one of those attitudes of data literacy. It's also being able to say, well, you know, maybe this is a gap between data and reality that I need to look at more closely, at that gap and understand why I'm getting something that I don't expect, that's surprising to me. It might be something real, it might be something artificial, it might be something erroneous, you know, there's all kinds of possibilities there, but being the kind of person that understands that those are the possibilities on the table here, you know, as opposed to like I and everyone else in the room that at first, was just assuming it was the number of bicycles. So I think that's all to me part of becoming data literate, is kind of appreciating those aspects of the overall space we're in, you know, and what it means to be effective within it. - That's great, that's a good story and it's actually funny, 'cause you know, that sounds like a data pitfall and I know that you have a new book coming up called Avoiding Data Pitfalls. You know, we're short on time, so maybe if you can just give us a brief overview of that book and then we will go on our way. - Sure, yeah, so that is actually, I think that's in the book right there and yeah, so Avoiding Data Pitfalls, I've been working on it for about five years and a lot of that's because of difficulties devoting time to it, but one of the good things about taking that long to write a book about pitfalls is you get to fall in a lot more of them and then put 'em in the book, right? And so, it's essentially, I like to think of it as like if I were to go back in time, I'd give myself that book and say hey, don't do all these things you know, but also then, I hope it's helpful for other people to avoid some of the common blunders we all make when working with data and the general theme of it is just that is what happens, you know, and so then, first of all, we just kind of need to embrace that our data's not perfect neither are we and so we're going to make mistakes along the way and then so can we build almost like an immune system to that? Can we kind of become more aware of the sorts of pitfalls, whether it's, you know, things like statistical slip ups, like understanding sampling and how that works, or doing calculations with rates or percentages, or what are some common ways people get that wrong? Even to the graphical side of things, maybe I can design something and show it to you in a way that completely confuses you, or even worse, misleads you, and just being able to be aware of the sorts of things that commonly kind of are part of that overall road we're on, you know, trying to get to this journey okay, and this like destination you want to reach, which is basically success, you know, with data, knowing that, you know, there are some things we might end up in the bottom of a pitfall along the way there. So, I just try to use like real examples, tell some stories in there from my own experience as well as some hands on examples of, you know, kind of showing people some of the kinds of mistakes I've seen made a lot and how we can build a little bit of an immune system to it, to ward off, and be more alert to those kinds of problems as well as preventing them from happening, avoiding them, you know, so that's that book. It should ship I think here pretty soon, but yeah, November 20th is when it should start to ship and really lookin' forward to seeing what people think about it. - That's great. So yeah, just for the audience, just so you all know, we're shooting this in November of 2019, so the book is coming out any moment now and so by the time this airs, it will be out and available for you to order and purchase. So Ben, thank you very much for being here today, talking about data literacy. It was a fun conversation, I'm sure that our audience will really have a lot of insights that they can take away from it and they can always go to your website and other places for more. So again, thank you very much. - Thank you Bill, appreciate it, bye everyone.