Back in fall 2006, I had barely started my career in the IR/IE field, and our oldest child was starting her first year in Governor’s School in Virginia. One of her assigned books was How to Lie with Statistics by Darrell Huff. Although it was originally written in 1954, I had never heard of this book yet found the title intriguing. So many of the points and conversations she and I had that year on what she learned from this book mirrored the same conversations that had transpired in my very first statistics class, which wasn’t until I was a grad student!

Helping our constituents read and understand data—grasping that all-important compelling story of our institution—is in our DNA as IR/IE professionals. And yet, we regularly witness the misuse and misunderstanding of data, whether well-intended or intentionally misleading.

How to Lie with Statistics is even more relevant today than it was over 67 years ago when it was written. Shannon Vallor in An Introduction to Data Ethics states, “The combination of increasingly powerful but also potentially misleading or misused data analytics, a data-saturated and poorly regulated commercial environment, and the absence of widespread, well-designed standards for data practice in industry, university, non-profit, and government sectors has created a ‘perfect storm’ of ethical risks.” To further drive this home, a recent EdSurge article discussed how disinformation is affecting our U.S. democracy and the role colleges can play in rectifying this before it is too late. I would argue waiting until college IS too late; building this ability to think critically about information needs to begin much earlier, and reading How to Lie with Statistics is a great starting point, before entering higher education or the work force.

How to Lie with Statistics is 142 pages, easy-to-read, and filled with fun sketched illustrations by Irving Geis, removes the stigma and intimidation of learning to read, understand, competently use, and make decisions based on data and information. Huff introduces readers to basic skills and understandings that will help them become informed and discriminating data consumers. With the overarching outcome of teaching readers to always question the source whenever given so-called facts and opinions, each chapter breaks this concept down into easily understandable subcomponents.

One of my favorite chapters, Chapter 9: How to Statisticulate, drives home the real-world use of the book’s title. Huff provides example after example of how corporations, the media, and others present accurate data using graphics, but with unintended consequences. The Darkening Shadow is one of the first eye-opening graphics, using a U.S. map to show how much of the national income was being taken and spent by the federal government. Both the western-style and the eastern-style show the exact same data representation. However, the western style uses the states with the largest area AND the smallest populations AND comparatively smaller incomes. Huff goes on to present example after example to assist the reader in understanding exactly how data can be presented both correctly and deceptively at the same time.

Huff skillfully leads the reader through all sorts of concepts like this, from studies selecting the average that best suits their needs (mean, median, or mode) to making the change in data over time appear more substantial by scaling down the chart axis.

As IR/IE professionals, we tend to be keenly aware of the importance of data ethics and data integrity, and how data (quantitative) literacy goes hand-in-hand with these. Our AIR Statement of Ethical Principles includes very specific statements relating to data integrity and data ethics. How to Lie with Statistics is a great starting point in helping all data consumers with the data literacy part, building a better appreciation of both the art and science behind our work. I encourage you to join me in talking with your local school districts about including data literacy in their curricula. The earlier we can build data literacy, the more informed we can be as consumers, and the less likely we will be to succumb to misinformation.

P.S., the book makes a great stocking stuffer.…

References

AIR Statement of Ethical Principals (2019 September 13). https://www.airweb.org/ir-data-professional-overview/statement-of-ethical-principles/principles 

Huff, D. (1993). How to lie with statistics. W. W. Norton.

Koenig, R. (2021, October 7). American Democracy Is Sick. Can Colleges Be Part of the Cure? https://www.edsurge.com/news/2021-10-07-american-democracy-is-sick-can-colleges-be-part-of-the-cure 

Vallor, S. (n.d.) An introduction to data ethics module author: Retrieved October 19, 2021, from https://www.scu.edu/media/ethics-center/technology-ethics/IntroToDataEthics.pdf.