Skip to main content
Tulane Home Tulane Home

Focus on Faculty: Scott McKinley

Publication date

January 17, 2018 3:45 PM


Jake Ward

Scott McKinley, Associate Professor of Mathematics at Tulane.

Scott McKinley, Associate Professor of Mathematics at Tulane.


Scott McKinley, Associate Professor of Mathematics at Tulane, brought his research passion into the classroom in his first-year Honors Colloquium, “Lies, Damned Lies and Big Data.”  The course addressed fundamental questions of truth in academia and the world at large, aiming to provide students with tools from his field to critically assess statistical claims.

The story of how Professor McKinley came to teach his Colloquium begins as all good stories do: with scrolling through the comments on an online article about climate change.  Via this, he began reading about climate change denial, and resolved to teach a course that would help students recognize lies and misuse of statistics.

The course name is a reference to the quote, popularized by Mark Twain but likely originating earlier—“there are three kinds of lies: there’s lies, damned lies and statistics.”  Twain himself credited the line to ex-British Prime Minister Benjamin Disraeli, though this is also disputed.  Even the mystery behind these words reflects the elusive nature of the distinction between truth and falsehood.

Professor McKinley calls a lie with statistics “a special kind of lie.”  It carries a weight that other fabrications do not.  Most of us feel underprepared to assess its veracity.  His 21st century update to the quote replaces “statistics” with a new buzzword: “big data.”  In our increasingly technologically savvy world, the power of numbers is only growing.

The Colloquium was split into three parts.  The first taught students “how to lie with statistics.”  Professor McKinley wanted to start here because, as the maxim goes, “the best way to catch a liar is to learn how to lie.”  Different strategies of lying with statistics were discussed—for example, the use of anomalies in the data to suggest different trends than the overall data set.  Introduced to these strategies, the hope is that students will pay more attention to their “red flags” in the future.

The second section of the course posed the more complicated question, “So what is truth?,” to students.  In doing so, it reviewed the varying standards of truth in different academic fields.  Truth to a mathematician is simple; it usually relies on straightforward deductive logic: “if this, then this.”  The standard for a statistician is a little less rigorous, but still clear: a causation is true if there is a less than five percent chance of it happening by random chance alone.  In the social sciences, meanwhile, truth is difficult to define, since no controlled replication is possible.  In the humanities, coming up with a definition is even more challenging.  The intention of this part of the course was to demonstrate to students that the standard of “truth” is not always the same; yet, it does have meaning.

The final component of the course asked students to pick a truth and tell it.  Working in small groups, they had to decide on the best way to present a certain truth to people who disagree—they chose a medium, an argument, and anticipated doubts of the viewer to refute.  The key issue was one of trust: How could they assure people that they were not using statistics dishonestly, and that they were utilizing a real standard of truth?

“Fun, but a lot of work” is how Professor McKinley describes the experience of teaching this course.  Ultimately, he says, it’s the engagement and thoughtfulness of Honors students that made it worth it.  They were unafraid to follow the themes of the class to new places, and he was able to leave his comfort zone as well, inviting guest speakers to help him explore topics outside his area of expertise.

Professor McKinley ultimately wants to help his students distinguish between fact and fiction in the real world, and present their own facts in a trustworthy manner.  This aim is probably even more salient now than it was when he first submitted his syllabus to the Honors Program a year ago.  As a statistician, McKinley wants his students, and others, to understand that there is a rigorous standard for truth in academia, even if it differs among fields.  Honors students were excited to learn about the tools he had to offer.

All faculty interviewed recommend a book that they find compelling and important.

Professor McKinley’s first recommendation is The Signal and the Noise, by Nate Silver.  It offers a rigorous presentation of a statistician’s perspective on the world, providing real world applications “better than anyone I know.”  McKinley also recommends Carl Sagan’s The Demon Haunted World, a book that was incredibly important to him during his time as an undergraduate (at Tulane).  This work is less about statistics, but more about being skeptical about potentially unfounded claims.  Sagan, a big inspiration for Professor McKinley, is a great example of an academic who reached out to people outside his field, attempting to present the scientific standard of truth.