What’s Going on With This Graph? | Stats + Stories Episode 77 / by Stats Stories

Hessney Pic.jpg

Sharon Hessney is a veteran K-12 math teacher and writer, moderator, and coordinator of What's Going On With this Graph? WGOWTG is a free, weekly online feature, is a partnership of the New York Times Learning Network and the American Statistical Association.

+ Full Transcript

(Background music plays)

Rosemary Pennington: Data literacy and data visualization are issues we discuss with some frequency both on this program as well as outside of it. When done well, data visualizations can make complicated material easier to understand. When done poorly, they can distort or misrepresent research findings. Understanding how visualizations like graphs work and communicate data is the focus of this episode of Stats and Stories, where we explore the statistics behind the stories and the stories behind the statistics. I'm Rosemary Pennington. Stats and Stories is a production of Miami University's departments of statistics and journalism and film as well as the American Statistical Association. Joining me in the studio, our regular panelist John Baylor, Chair of Miami Statistics department and Richard Campbell, the former chair of the department of Media Journalism and Film. Our guest today is Sharon Hessney. Hessney is a writer, an educator and the coordinator of The New York Times’ learning network’s What’s Going on With This Graph feature, where once a week, students can take part in a live discussion of a graph that was published somewhere on the New York Times website. Sharon, thank you, thank you so much for being here today.

Sharon Hessney: Thank you Rosemary it's a pleasure.

Pennington: Would you mind, just to get us started, giving us a little bit of background about where or what is going on the…with this graph sort of came from?

Hessney: Sure. About twenty years ago the New York Times started a new department called the learning network and the learning network’s objective was to show teachers how to use the resources of the New York Times in their classrooms. It could be K through 12, it could be college, you could even be grad school. And one of the first features they had was something called - What's going on with this picture? Where they would show you a picture and then you try to guess what it was. Well about a year ago they came to the American Statistical association and they said, you know, we’d really like to have something that has to do quantitative and so they thought of the idea of why don't we do what's going on with a graph. So we take a graph that was in the New York Times in the past and we asked the students, what do you notice, what do you wonder, what story is this graph telling and write us a very brief headline that gets to the main idea. So that's where it came from. We've been up and running for a year. Last year we were once a month and this year we will be most Wednesdays through the school year.

Pennington: Oh wow!

John Bailer: That's awesome. So how do you pick the graphs? What are some of the criteria used to select these graphs?

Hessney: Well Rosemary gave me a list of the things I do for what's going on with the graph. The most fun part is like curate the graph. So I have been addicted for decades to reading the New York Times. I collect the graphs in the New York Times that I think are appropriate and then I go through them looking for ones to use for What’s Going on With This Graph. So there are a couple criteria I use. The first and the most important is the viewer, the student, needs to be able to see themselves in the graph. They really do not want to look at a graph of the growth of G.N.P. in Russia which was in The Times today, but they may be interested in a graph that shows how e-commerce is taking over Christmas sales, OK which we did last Christmas. So I go through them and I try to find ones that the students can see themselves in. The next thing is I want something that has a context of interest and is very timely. We vary the graph from month to month. So in the first year, we had heat maps, scatter plots, time series, box plots, bubble chart, histogram - all kinds of graphs and we try to see if from week to week we can have different ones. Finally what makes a graph really interesting is when you compare things in it. So not one set of data but two sets of data. So for example we had a very interesting one that we thought wouldn't probably get much response but got enormous response with. It was the labor participation rate in different countries in the world and so you could see, geez, the labor participation rate was going up in a lot of the countries. But why was it going down in the United States and so that comparison is what led to a lot of students wondering things about what was going on in that graph.

Richard Campbell: Sharon, one of the things that is a challenge and I have to say I really kind of admire your commitment to, particularly, high school education is the students that we run into that have math phobia or numbers phobia and what do you do when you sort of come up against that and I know we run a journalism program here and a lot of our students are just…struggle with this. And it seems to me like you know what you're doing is kind of an entry way into understanding that’s at a level that is not so daunting.

Hessney: You know, this is part of the genesis of the whole program. You know, what people often do, they will see an article and they'll see it has the graphs in it and totally skip it…just go, you know, to the picture, the headline and maybe read the text. What the people…those of us who use graphs all the time know that, if a picture's worth a thousand words, a graph is probably worth, you know, a million numbers. And so, we really want to kind of break down that intimidation factor in graphs.

The way we do it is a technique that Annie Fetter from the Math Forum, which is now part of the National Council of Teachers of Mathematics does, which is what do you notice, what do you wonder. So when you look at a graph, everyone can notice something. There are dots on it, there are…you know, an X axis and a Y axis. It's about different countries. You can always come up with something that you notice, and as you dig more deeply, you'll come up with additional noticing. After noticing, you say then, what do I wonder? What questions haven't been asked? We think the best practices to do What’s Going on With This Graph, is first for people discuss this either individually or in small groups and then have a whole group. Because by hearing other peoples’ noticings and wonderings, you get more and more ideas. After you've done this for a short period of time, it's really pretty exciting and simple to come up with a short catchy headline about what the story is in the graph. So that's the way we do it. And it's amazing how students are so excited about contributing to this and coming up with things they never thought of. In addition, when they walk out of the room I certainly expect that they'll probably still be thinking about the graph and the issues in it.

Bailer: So what graph has gotten the most interest of all the ones that you’ve had so far?

Hessney: (Laughs out loud) Out of over two thousand and seventeen, the food graph!

(Collective laughter and giggles)

Campbell: I’ve seen that.

Hessney: So you want to know why? I wonder why that much of interest! Well in addition to the fact that it's about food and they were great little pictures of food, it was a scatterplot and it compared stat with nutritionist, whether nutritionists feel a particular food is healthy or not, to what the general public says about it. And it had each of these fifty foods graphed on a scatterplot and the really interesting thing about it was, there was this line that went from the lower left hand corner to the upper right hand corner. Now when you have a scatter plot that line which is halfway between the two axes that's probably the Y and X axis, Y=X line, and it was where nutritionists and the public agreed. So those foods above the line, nutritionists thought were healthier than the public and those below, the public thought was healthier than the nutritionist. And that had a lot of discussion about what was going on with this graph and you know the story behind it.

Pennington: What kind of classrooms are you imagining this is being used in and is this something that is just for math classes or are there other classes where you think this might be an appropriate or a useful tool?

Hessney: I think it's useful in any class. If the content relates to the class, so we had how men and women express love in literature…

Pennington: Oh!

Hessney: And you would think that would be great for an English class but the math behind that graph would also be appropriate in a math class and I'm sure would have the students smiling when they saw that that was the graph they were going to be talking about. But because students talk about the graphs, because they have critical thinking about the graph, because they're engaged in current events and timely topics, and because they end up writing about what they see, what they notice and what they wonder, it's really appropriate for all courses and it doesn't even have to be in a classroom setting. I'm part of a volunteer group that observed trees in the Boston area. And they're going to be looking at a graph because we have them in our volunteer program and just having them get used to how you read these different graphs that have to do with climate, temperature, time of day when the leaves come out, when the fruit comes out…it can be all kinds of situations where the graphs…after all, you know it's not just textbooks that we see graphs in. We see graphs on Facebook, we see graphs on YouTube, we see graphs on other…different other websites you know, and learning how to see understand graphs is really a twenty first century skill.

Bailer: I'm curious what got you to this point? You know, how did your career evolve to the point that you're curating this really cool exercise?

Hessney: Well be careful what you do when you're young because you could become addicted.

(Collective laughter)

Hessney: I told you, I'm addicted to reading newspapers. I tend to read them at 6 every morning and my students for twenty years and beyond in the Boston public school would know, I would walk into class, I would put a graph under the….what then was the overhead because that’s the technology that we had and I would say, What’s Going on With This Graph? And it was always something that intrigued me…I mean, I didn't do it every day. And I have to be honest, I didn't think the quality of the answers that I got back was that great. And I often ended up telling the students what was going on with the graph which is not the best way to proceed and so when with the New York Times we came up with the idea of using what do you notice, what do you wonder, then we really saw an increase in the quality of responses and that's really the key of exploring because it's those two questions that kind of put together the pieces of the puzzle to figure out what the story is. And we became very successful with that, so that's how it started.

Campbell: So as a longtime reader of newspapers, can you talk a little bit about what you…just as a general reader, what you've noticed about what journalists get right and what journalists get wrong, what they maybe need to know and...it would be helpful probably to have you as a teacher as well.

Hessney: Yeah. OK. Graphs are tough. They are very tough. You know, every…not every…most graphs are different, they take different approaches, they have different kinds of data. Even once you have the data, there are different ways of representing it. There are right ways and there are wrong ways to represent it. It is difficult and the only way you get good at interpreting graphs and asking the right questions is if you have a lot a lot a lot of practice OK. So I think graphs in general, in media, have got better because a lot of publications have been burned by really bad graphs which been published. And they now probably have on staff people that will help them out or at least look at the graphs to make sure that they're OK. In addition you've got to be able to get the text that goes with the graph to be consistent with what the graph says. And there we have a little component of What’s Going on With This Graph which may be helpful which is called Stat nuggets. So with each of the graphs we have two or three statistical terms that we define in layman's terms. And then after we do that we write up where you see it in the graph. And these are the knotty statistical terms that often are interpreted incorrectly in text. So the difference between number and proportion or net change and percentage change, or the one that is used most frequently which is the difference between association and correlation. So we include those in the instalments of What’s Going on With This Graph, where appropriate, and along the way you kind of end up learning how to approach the graph.

(Background music plays)

Pennington: You're listening to Stats and Stories where we discuss the statistics behind the stories and the stories behind the statistics. I'm Rosemary Pennington, with Miami University’s statistics department chair John Baylor and Media Journalism and Film former department chair Richard Campbell. Our guest today is Sharon Hessney, who works with the New York Times on the learning network, on the What’s Going on With This Graph feature. Sharon, before the break you mentioned there were some fairly well known examples of graphs gone bad in reporting. Do you have an example or two where you could, sort of you know, for the listeners kind of explain what you mean by a graph gone bad in reporting?

Hessney: (Chuckles) There are a whole different….well there's a whole bunch of graphs especially in newspapers which like to show how quantities have increased over time and so you know the amount of garbage collected in a city. And instead of having a nice little bar graph you know we collected fifty million tons, then one hundred million tons and then two hundred million tons and having the bars double in size each time, they'll draw pictures of garbage can. And the second garbage can is twice as tall as the first and the third one is three times as tall as the first. Garbage cans signify the volume, which is cute, OK? And you know lines, you know just increase by the length, OK? So it's a misrepresentation that looks like it's increasing a whole lot faster or a whole lot more than it really was and so that's kind of a silly one that often appears. There are plenty of other, kind of silly ones that appear. As far as when I talk about association and correlation, correlation should really only be used when there's a linear relationship and with either of them you shouldn't assume that there is a cause and effect. There's only a cause and effect if you do an experiment and so you should be careful with using those words, because that has special meaning.

Bailer: We tend to differentiate pretty strongly between correlation and causation as we discuss this in our classrooms as well. The issue of kind of general association is something that we often subsume in a discussion about correlation is a linear association with it is a special case. You know your comment about the idea of bad graphs and reporting in the new…the more recent ones being better, I strongly agree and I think back on this. I used to use the papers for the examples of all the bad graphs like that to show my students to not do, but I think now that things have really improved. You're not as likely to see the misrepresentations like you described. I still think there's often some room for aesthetics and improvement in design, but generally I really like how times have changed in terms of the quality of visual displays.

Hessney: You know, one of the other things is I'm hoping that publications are less shy about putting in graphs. So when you read an article that is just full of facts and figures, you know, in 2008 it was this, in 2018 it's now that, and they give it to you for five different cities and in different groups of people or something, it would be great if that could be represented in a graph. You can get a lot more understanding in a graph than in a text form.

Campbell: How…you mentioned before, the relationship between writing text to go with the graph, what's the challenge in that, especially if you're, you know, if you're trying to help an English teacher explain a graph and get the students to write about that graph, tell the story that's there? What kind of tips do you have for getting somebody who's really not trained in statistics or math to be able to tell that story in the best way?

Hessney: OK. So, the big one is practice. You know it's the same with the violin – how do you get to Carnegie Hall? You just practice, practice, practice and you get good at graphs if you practice. The other is, so what makes What’s Going on With This Graph so really special is we release it on Tuesday afternoon. On Wednesday from 9 am to 2 pm, we live moderate student responses. So it's very easy, everything is free. Students and teachers sign up with What’s Going on With This Graph, and the way you get to it is, just put in your browser, What’s Going on With This Graph, you'll come right to it, click on that one graph OK sign up all you have to do is put in a name and a location and they can be as simple as you know John from New York. And we ask students to write down what they notice and what they wonder. In some cases teachers will compile what the students are noticing and wondering. We have teachers who moderate this and reply to the students’ responses in real time. And the import of what we do is not just a good job but say, can you tell us where you saw that in the graph? What makes you say that? If you wondered about something, where could you find that answer? And so we try to encourage them to go to the next step in analyzing graph. Now this year for the very first time we're doing, in addition to what’s the story in this graph, to writing the headline and it's going to be very interesting what we come up with, with headlines. And I don't know what my comments will be for that but we're excited about seeing it.

(Slight voice overlap)

Bailer: I really like that you have that moderation built into that because one of the things I was thinking about as you were describing this, is what do teachers need to know to best facilitate discussions on this? I would imagine that being able to have you moderating or others that are reviewing this, given that real time feedback, would give the teacher some idea about how they might be able to do this, you know, offline, in their own classrooms.

Hessney: Yes So OK a couple things. First of all, all the releases are archived. So that’s kind of interesting because if one of your listeners says well, let's go to the food graph! You can just go find it there and you can have the students look at it and can respond. We will not reply but they can respond, OK? The other thing is the wave releases are written it tells it's actually written to the student which tells the teacher as well what to do. So it says, here's the graph. You answer these questions: What do you notice? What do you wonder? What’s Going on With This Graph? Give us a title…give us a caption. OK? So it tells you what to do and then we'll respond. So there's really no prep for teachers, or for anyone else who's going to be using that. Because it's all there.

Pennington: What's the response been from educators? I mean, you're going from once a month last year to what it sounds like a weekly schedule. I mean it sounds like the response must have been good, otherwise you wouldn’t be able to do that.

Hessney: You know we were getting hundreds of responses a month, we had over three thousand responses for the year. And the times that we have so many more graphs let's just go weekly.

Bailer: That's quite a commitment but there's certainly a lot of graphs that are being produced like you said in The Times.

Hessney: Yes. There's a lot of graphs. We have sixteen moderators from across the country who are teachers and one in Qatar who will actually be moderating tomorrow. And we rotate them through to get different perspectives.

Bailer: Oh that’s cool. How did you select the moderators or how did you recruit them?

Hessney: First of all, anyone who's interested. And. You know. Social media has all kinds of platforms where teachers talk to each other and we know which teachers are discussing statistics and graphs and coming up with really interesting lessons and strategies and those teachers are the ones who want to moderate it.

Campbell: So this is a sort of a more general question about…goes back to something we talked a little bit of earlier. So how do you inspire students who don't seem to show aptitude or interest in numbers and statistics? What have you done in the classroom over time to inspire them?

Hessney: Well you know I think it's as simple as when you put up the graph, they go, I want to do this. I want to do the food graph. This is fun! I want to see where my favorite foods are on this graph.

Campbell: So a lot of the example sometimes…

Hessney: Yeah, yeah, yeah which is why we never thought we were going to get much of a response from labor participation not exactly a hot topic for most people but it was so discouraging to see how the U.S. labor participation was declining in comparison to other countries, that people became engaged, the students became engaged. So it's you know it's just looking at the graph. I mean we have had a graph on the number of downloads by popular singers by parts of the country. OK and never was a negotiation as hard as when the New York Times and I negotiated with the four singers who would feature, and I went outside and I asked people and this is the one somebody wanted and I said what I went to the Boston Public Schools and this is what the kids wanted! So anyway, so you know, when you see that and you're really interested in one of the singers, you know, you really like Drake, OK, you're going to want to know, geez, where is he the most popular? Then the next thing is well let's assume you're not even interested in some graphics, not that exciting to you. If I just say to you, what do you notice? I don't give you a quadratic equation and say what’s the solution? OK, I say, what do you notice in this graph and when you say, you know, I see pictures of food on it, I say, great, and what does anyone else notice and you keep going. It keeps students interested.

Bailer: You know we had a guest many episodes ago that talked about numbers as plot elements and stories, you know. I think that what you're describing and what you're doing with this, is looking at graphs, this kind of more than you know, aggregates of numbers as part of the story that's…you know, I liked your comment earlier about you know, a picture is worth a thousand words but the idea of a graph worth a million numbers…there's the idea of this integration and collection of the story sounds really cool.

Pennington: Well that’s all the time we have unfortunately for this episode. Stats and Stories is a partnership between Miami University’s departments of Statistics and Media Journalism and Film as well as the American Statistical Association. You can follow us on Twitter, Apple podcast or other places you can find podcasts. If you'd like to share your thoughts on the program send your e-mail to statsandstories@miamioh.edu or check out our fancy website at statsandstories.net and be sure to listen for future editions of Stats and Stories, where we discuss the statistics behind the stories and the stories behind the statistics.