Mastering the Art of Quanti-Manipulation

Bookstore shelves are filled with books that tell you how to charm, impress, manipulate, seduce, sway and stage manage other people.

Lucio Buffalmano, a self-described entrepreneurial social scientist, has written a useful guide to the many books that purport to describe and analyze the psychological and emotional dynamics of manipulation. Here are a few that he mentions:

• Venkatesh Rao’s The Gervais Principle, which describes how business organizations exploit middle managers who are highly conscientious and clueless simpletons.
• George K. Simon’s In Sheep’s Clothing, which examines various forms of covert aggression used by social manipulators “so that they can always deny their aggression, retreat, save face, and then probably blame you for your own overreaction.”
• Robert Greene’s The 48 Laws of Power, which is explicitly Machiavellian, and advises readers to conceal their intentions, court attention, keep people dependent, cultivate unpredictability, disarm victims with selective honesty and generosity, play a sucker to catch a sucker, and pose as a friend while working as an enemy.
• Harriet B. Braiker’s Who’s Pulling Your Strings, which looks at the psychological weaknesses that manipulators exploit, including the need for approval and acceptance by others, fear of conflict and confrontation, a lack of assertiveness, and low self-respect.

Most Popular Stories

Most Popular

If any single theme can be said to run through these books it is, to use the rude, crude vernacular, to “mindfuck” others: to deliberately inflict psychological damage, for example, through gaslighting (using psychological methods to force someone into questioning their own sanity); guilt tripping (creating feelings of fear, guilt or obligation); cajolery (persuading through flattery or false promises); grooming (methodically building trust to lower someone’s inhibitions); shaming (playing on a person’s insecurities and feelings of inadequacy); emotional blackmail (pressuring someone to do something by exploiting their secrets or vulnerabilities); infantilization (making someone feel weak and powerless); and bullying& (intimidation through threats, insults or aggressive behavior).

To this list we might add some other forms of emotional manipulation: playing the victim, downplaying someone’s concerns, silently sulking (and other forms of passive aggression), pushing buttons, twisting the truth, and, perhaps most important of all, exploiting someone’s trust or emotional neediness.

Of course, the classic text on manipulation is Darrell Huff’s 1954 How to Lie with Statistics, which explains how marketers and politicians and even news outlets misrepresent data and exploit the public’s trust in numbers, graphs and charts.

Why do experts and authorities and advertisers mislead and lie with statistics? Huff’s answer: “To sensationalize, inflate, confuse, and oversimplify.” And, of course, they profit or benefit in some other way.

The book may be dated and lack substantive statistical analysis. But it does do a very effective job of describing the techniques that data manipulators use to mislead, misrepresent and bamboozle the guileless and gullible and lead them to draw unsubstantiated conclusions:

• Cherry-pick (or omit or fudge) data.
• Engage in survey or sampling bias.
• Hide inequalities.
• Manipulate charts and graphs.
• Ignore the difference between mean, mode and median when reporting averages.
• Ignore intervening variables.
• Use correlation to imply causation.

Among my favorites: Using the words “up to” as a way to exaggerate savings or impact or improvement. Making statistics look more precise by including decimals. Using percentages to hide raw numbers. Omitting statistical qualifiers such as degrees of significance or comparison groups. Truncating or mislabeling graphs.

(For a wonderfully witty update of How to Lie with Statistics, see UCSD’s Rick Ord’s “How to Lie, Cheat, Manipulate, and Mislead using Statistics and Graphical Displays.”

One of Higher Ed Gamma’s overarching arguments is that the college curricula ought to better reflect our learning objectives. Like Derek Bok, I don’t believe that the current approach to gen ed—to require a course or two in a particular area, such as diversity, cultural difference or cross-cultural understanding—is satisfactory.& Also, like President Bok, I believe our institutions need to prioritize teaching, adopt evidence-based reforms, and strive to educate the whole person and promote growth along every vector: cognitive, social, emotional and ethical.

If we want to produce ethical graduates, we need to embed moral reasoning within particular majors or pre-professional degree pathways. If we want to prepare students to participate in a globally interconnected world, then graduates must acquire a competency in at least one of the following areas: global problems, international relations, foreign languages and literature, comparative and regional studies, and global problems.

If we want to nurture interpersonal and intrapersonal skills—teamwork, resilience, perseverance and creativity—we need to integrate collaborative problem solving into our classes, increase participation in community service, athletics and other extracurricular activities, offer mind-set training, and incentivize students to participate in innovation labs and maker spaces and other forms of experiential learning.

If we hope to cultivate graduates who are culturally literate or who have an appropriate level of knowledge about social science theory and methods or the frontiers of the natural sciences, we need to design our courses accordingly.

In today’s data-informed age, we need to do a better job of instilling statistical literacy.

As the macroeconomist Hannes Malmberg and the population geographer Bo Malmberg have recently written in an important article entitled “How Mathematics Built the Modern World”:

“Geometric calculations led to breakthroughs in painting, astronomy, cartography, surveying, and physics. The introduction of mathematics in human affairs led to advancements in accounting, finance, fiscal affairs, demography, and economics—a kind of social mathematics. All reflect an underlying ‘calculating paradigm’—the idea that measurement, calculation, and mathematics can be successfully applied to virtually every domain … It was this paradigm, more than science itself, that drove progress. It was this mathematical revolution that created modernity.”

In this article, the authors describe the adoption of Hindu-Arabic numerals, the popularization of decimal notation, the introduction of logarithms, the invention of the slide rule, and the appearance of the first printed mathematical tables, which “were crucial for computation before the advent of electronic calculators.”

The authors also show how the geometrical concepts of points, lines, planes, and surfaces revolutionized painting, surveying, cartography, astronomy, open sea navigation, mechanics and even warfare through the redesign of forts and emergence of new ideas about ballistics and trajectory. Then, they also describe the impact of Arabic algebra on European business, banking and state financial administration, providing the basis for double-entry bookkeeping and improvements in calculating interest and discount rates, and of the contribution of new theories of probability to the birth of actuarial science, demography, economics, modern engineering and epidemiology, as well as the emergence of the insurance industry.

Especially striking is the authors’ description of the diffusion of “the calculating paradigm” less through universities, with their classical curriculum emphasizing grammar, logic and rhetoric, but through Abacus and military schools and various private academies, popular applied mathematical textbooks, and the educational programs of Philip Melanchthon, Johannes Stöffler and Petrus Ramus. The authors do an impressive job of showing how basic mathematics and geometry underlaid the development of precision engineering, including the invention of modern clocks, locks, steam engine cylinders, and the many instruments and tools described in recent books by Simon Winchester and Henry Petroski.

However important statistical representation, manipulation and inference was in the past, it is at least as valuable today. We need to ask ourselves: How can colleges do a better job of familiarizing all college graduates with the art and science of collecting, analyzing, interpreting and drawing conclusions from data and helping them make data-informed decisions?

How, in short, can we ensure that our graduates achieve an appropriate level of competence with:

• Descriptive statistics: Describing samples and populations and understanding frequency distributions, central tendencies and data variability.
• Inferential statistics: Using appropriate statistical techniques to make predictions, draw generalizations or conclusions, test a hypothesis based on a sample or estimate how a change in one variable will alter another.

When I directed Columbia University’s Graduate School of Arts & Sciences Teaching Center, I had the opportunity to invite Andrew Gelman, one of the real giants in statistical modeling, causal inference and Bayesian data analysis, to lead a session on how to teach probability and statistics.

He began the workshop with two “tricks.” The first had one group of students flip a coin a hundred times and write the results on the blackboard; the second group pretends to toss coins, and write heads or tails on the board in any order they wished. Then, Professor Gelman identified the group that faked the results: It wasn’t sufficiently random; it didn’t contain enough streaks of heads or tails.

The second trick was to estimate the value of coins in a pickle jar.

If you are unfamiliar with Professor Gelman’s pedagogical approach, pick up a copy of his Teaching Statistics: A Bag of Tricks, now in a second edition, co-authored with UC-Berkeley’s Deborah Nolan. The authors identify entertaining ways to engage students in virtually every topic taught in a probability and statistics course, from measures of statistical significance and data dredging to survey sampling, time series, confidence intervals, logarithmic transformations, Poisson and normal distributions, and probabilities of compound events.

Most of their examples—involving gambling, lotteries, election exit polls, sporting events or space shuttle disasters—are drawn from real life. These include estimating the value of a life, corner cutting in medical studies, the percentage of marriages that end in divorce, regression of income, height, and sex, the fairness of exams, how common a student’s name is, the effect of coaching on SAT scores, and evaluating the Surgeon General’s report on smoking and cancer.

The authors donate all royalties to nonprofit educational organizations.

We’re all familiar, I suspect, with various statistics-related witticisms. “Statistics don’t lie, but liars use statistics.” “There are lies, there are damned lies, and then there are statistics.” “Statistics are like a bikini—what they reveal is suggestive and what they conceal is vital.”

In today’s world, statistics lies at the heart of business, engineering, medicine, public policy and science. Statistics help us find patterns and trends and forecast future behavior. Statistical literacy is, in short, essential.

But, as Peter Schryvers, an urban planner, observes in his 2020 book, Bad Data: Why We Measure the Wrong Things and Often Miss the Metrics That Matter, data deception is rampant, information hubris is widespread and statistics are often twisted to advance various agendas.

Schryvers’s book “dissects the metrics we apply to health, worker productivity, our children’s education, the quality of our environment, the effectiveness of leaders, the dynamics of the economy, and the overall well-being of the planet.” He shows how the misuse of statistics has contributed to the growth of “corporate cultures that emphasize time spent on the job while overlooking key productivity measures, overreliance on standardized testing in education to the detriment of authentic learning, and a blinkered focus on carbon emissions, which underestimates the impact of industrial damage to our natural world.”

The book offers advice that all of us, regardless of our area of specialization, should take to heart: That statistics aren’t as objective as they seem. That focusing on select data points can mislead by masking variables—sociological, psychological, political or cultural—that we don’t understand. That we mustn’t confuse data trends with the outcomes that we actually desire.

The book, in short, reinforces some lessons that we should all recognize:

• “Not everything that can be counted counts and not everything that counts can be counted.” (From sociologist William Bruce Cameron)
• “When a measure becomes a target, it ceases to be a good measure.” (Goodhart’s Law)
• “Making a decision based solely on numbers, metrics, and quantifiable data inevitably results in flawed decisions.” (MacNamara’s Fallacy)
• "The more any quantitative social indicator is used for social decision-making, the more subject it will be to corruption pressures and the more apt it will be to distort and corrupt the social processes it is intended to monitor.” (Campbell’s Law)
• “For every data point that proves one point of view, there are two that refute it.” (Shevlin’s Law)

The most important takeaways are these: In today’s world, statistical and data literacy are as important as any other literacy. Nor are data interpretation or risk analysis or statistical reasoning just for researchers.

Even for those not going into a field directly related to statistics, having a foundational understanding of these concepts is valuable for navigating a data-driven world, making informed decisions and critically evaluating information presented in professional and personal contexts.

Every college graduate, regardless of their major or future career path, should have a basic understanding of statistics. They should understand basic statistical terms and concepts and recognize different kinds of data. They should understand how data are collected, including survey design and sampling methods, and potential biases in data collection. In addition, they should understand the basic principles of probability and statistical significance and be able to interpret data presented in various forms, be familiar with basic data analysis techniques and be able to apply statistics in everyday life and in various professional fields.

I wholeheartedly agree with Christine Franklin, the American Statistical Association’s K–12 ambassador: “Becoming statistically literate and data literate should be as core to our school-level curriculum as reading and writing.” If statistical and data fluency is important at the grade school level, it’s even more imperative for the nation’s college graduates and future professionals and decision-makers.