To help freshmen grasp causation versus correlation, Dr. Samuel Richardson introduces a free data visualization tool—and a fun video project.
Associate Professor of the Practice of Economics, Chestnut Hill, MA
PhD in Health Policy, Economics Concentration; BA in Human Biology, Health Policy Concentration; minor in Economics
Dr. Samuel Richardson would not have predicted that a Swedish professor’s TED2006 presentation would change the way he teaches economics at Boston College. But after watching The Best Stats You’ve Ever Seen, he realized how easy it can be to help non-numbers people to understand and identify changing data trends—and how broad the impact can be when educators share edtech tools and teaching strategies with one another.
In the video, Hans Rosling, cofounder of the nonprofit Gapminder, used the data visualization tool “Bubbles” to animate data on worldwide trends in birthrates and lifespans over a 40-year period. As he speaks, the colored orbs (each representing one nation) actually migrate across the graph as the years tick by, allowing Rosling to dispel the widely held notion that Third World countries have more babies and shorter lives. In less than a minute, the video makes it impossible to ignore the global shift toward smaller families and longer lives.
For Richardson, the impact was off the charts. “During my own training in economics, we had looked at the typical graphs one examines to decipher trends, such as those on income and life expectancy,” says Richardson. “But those show only one point in time.”
Today, as an associate professor of practice in economics, Richardson’s aim is to help students in health economics courses like Human Disease: Health, the Economy, and Society learn how to think more critically about data. To do that, he is using tools like those on Gapminder to dispel some of the erroneous and statistically unsupported beliefs that his students hold about global health data—and teaching them how visual representations of data can help bring new connections to light.
See and share lecture notes, practice tests, and teaching materials.Get access now
“I tell students to find something interesting [about the data] and tell me about it [in a video]. Since my students are primarily freshman, they’re used to high school projects that are more clearly laid out for them. They can have trouble with the ambiguity in this project, but I think it’s really important that they encounter this ambiguity in college. In a job, they might be given more open-ended types of assignments, so students need experience of this kind to be prepared.”— Samuel Richardson, PhD
Course: ECON 1701 Human Disease: Health, the Economy, and Society
Course description: About 9% of the world’s economic resources (and 18% of the United States’ resources) are devoted to health care—the prevention and treatment of human disease. This course will explore the social consequences of and responses to infectious and chronic diseases. Economics can provide insight into why researchers focus more on some diseases than others, why some health care systems work better than others, and how health care resources might be deployed more efficiently. Students will develop the ability to analyze the broader context and consequences of human disease, with a focus on the economics of health care policy.
Gapminder video project: An overview
Gapminder World has been invaluable to Richardson, whose main goal is to help students in his health economics courses learn how to think critically about world health and economic data. (See the list of indicators in Gapminder World.)
By and large, Richardson’s students—who are primarily freshmen—come to him with only basic analytic skills. These students, like many adults, often hold mistaken and outdated beliefs about the state of the world. In fact, Rosling cofounded the Gapminder Foundation and developed the tools used by Richardson to address this very common concern.
Richardson says that Gapminder World specifically helps students more easily identify variables that are correlated in a data set so that they can identify potential causal relationships. After they do this, he walks them through how to seek evidence to support or refute the causal claim. Finally, students work in small groups of three or four to create a short video (six minutes maximum) in which they tell the story behind some aspect of a Gapminder World graph. The objective of the project is to have students explore country-level data and to think critically about relationships between different variables in the data, focusing specifically on topics in health. Students then use Gapminder and screen-recording software to create their videos.
Richardson assesses students on identification of interesting patterns in the data, originality in the story students tell about the data, clarity of presentation, and the strength and quantity of supporting evidence. He also asks them to provide alternative explanations for what they observed. In addition to their videos, students submit written transcripts of their video content.
Video project how-to: Telling true, compelling stories with data
Here are Richardson’s time-tested tips for introducing key concepts and Gapminder tools, and for helping students make videos that are as moving as Rosling’s:
Expand students’ understanding of correlation vs. causation
Early in the semester, Richardson teaches students the difference between correlation and causation with intriguing real-world examples. To do this, he first introduces students to some correlated data points that are clearly not causal and should therefore not be used to drive policy.
“If you look at a plot where you have higher CO2 emissions per capita, there is a correlation between higher CO2 emissions and longer life expectancies. But this is due to other factors—it is not as if we should take countries that have low life expectancies and emit more CO2,” Richardson explains.
Richardson also introduces students to Tyler Vigen’s website Spurious Correlations, which contains thousands of “humorous” graphs, such as one showing a 95.86% correlation between the per capita consumption of mozzarella cheese and the number of civil engineering doctorates awarded. Students quickly realize that a positive correlation between two variables is not enough to determine health policy recommendations—it is only a starting place for conducting more research.
Introduce students to Gapminder
The first homework assignment in Richardson’s course has students go to the Gapminder website and play around with any of the data tools there. They are welcome to do this individually or in groups. At the next class, they share whatever they found that piqued their interest. This gives students an early sense of efficacy with the tool and helps them identify classmates with similar interests for the larger video project.
Have them put it in writing first
As students work with Gapminder to determine the story they want to tell with the data they are exploring, they are drafting a transcript (about 1,000 words maximum) that will serve as the basis of their voiceovers for the video. This transcript should explain the overall story, as well as what is happening on the screen in terms of the data points within Gapminder. When they turn in their transcript with the video, Richardson requires that they also cite any facts or statements that are not common knowledge, including direct quotes and paraphrased content. (He offers students a link to “Citing with Integrity“ from Boston College Libraries to help guide them.)
Direct them to easy-to-use screen recorders
Though Richardson allows students to use any screen recorder to create their videos, he recommends that Mac owners use the built-in screen recording feature of QuickTime Player, and for PC he suggests Screencast-O-Matic. Both options are straightforward and include built-in voiceover features, making it easy for students to marry their narration with the moving pictures.
Provide additional support as needed
Richardson finds that the open-ended nature of this assignment can prove challenging for his class, which is comprised mostly of first-year college students. In order to provide in-person support, Richardson sets aside several course meetings to work with students on their projects and provide them with immediate feedback on their emerging impressions of the data and, later, on the preliminary drafts of their video transcripts. Richardson notes that he is not asked many tech questions, because generally someone in each group is tech-savvy and can coach the others.
Student work: Sample Gapminder videos
Richardson’s students post their completed videos on YouTube. Though he gives them the option of making them private, he hopes they will not, “so that I can share some of the best with future students (and perhaps Twitter followers),” as he says in his assignment. Making them public this encourages groups to take extra pride in the creation of their project—and it means that the videos can reach other people seeking a better understanding of data. (The first example below garnered 675 views in three years.)
- Effects of increased access to clean water on the health of the general population (February 2016)
- Why does Japan have the highest life expectancy? (February 2019)
One particularly rewarding outcome for Richardson has been seeing his students apply what they are learning in his course to delve more deeply into the course materials in their other college courses, particularly those outside of the Economics Department.
For example, Richardson had one pre-med student who completed the Gapminder project, creating a video focused on determinants of mortality. From his work in Richardson’s course, the student was able to challenge some of the assumptions put forth in his pre-med courses around the importance of high-tech advances in medicine. As a result of examining the evidence and potential causation, he was able to argue that other variables—such as vaccinations and access to safe water—are far more central to life expectancy.