Why study statistics?

Builds character? There is a better answer. On this page we have started a collection of papers that consider different ways in which statistical thinking can improve the world. From refuting an exaggerated claim to identifying biased rating scales to collaborating across disciplines to using statistical techniques to enhance critical thinking and / diversity in the workplace - it is worthwhile to examine the ways that we use statistics in our various fields of endeavor.

In this post-factual, post-truth age, it is important to have the skills to separate facts and truth from the distracting fictions that may surround it.

At the moment, the papers cited on this page are those written by your site authors, Profs L & H Friedman.
We will be happy to consider including other relevant research. Please email us with suggestions.

The knowledge economy and the digital age have changed the nature of work and management. More people than ever before need to master critical thinking skills to meet employers' demands. This paper proposes a set of 15 scenarios and where they may be used in the context of a one-semester course in statistics. Scenarios are chosen for their relevance, are from the fields of medicine, school admissions, institutional rankings, criminology, business, quality control and pharmaceutical research. Individuals who wish to sharpen their critical thinking skills may benefit from this paper.

You might be wondering, what is statistical thinking good for? Here are some attention-grabbing examples that demonstrate relevance of statistical thinking to all areas of the world we live in. Real-world examples in teaching, evidence-based research, health research, NNT (number needed to treat), happiness research, teacher cheating, attractiveness research, college rankings. The correlation coefficient helps us to evaluate research, i.e., the strength of the relationship reported.

The most exciting and valuable research is often done at the intersection of two or more disciplines. How can scholars in diverse disciplines using different scientific models, communicate for collaboration? Just think of the Manhattan Project which had experts for a variety of different countries, speaking different languages, and from such areas as math, physics, cryptography, game theory, and management.

Before introducing a new drug or other product to the market, companies must test for safety, efficacy and any other claim they might wish to make. Even the nutritional information must be substantiated. Of course, the competition (and government watchdogs) must examine these claims independently. Hypothesis tests might be conducted in order to substantiate a claim a company wants to make or might be used to refute a claim that a competitor is making about its product.

Rating scales are used to measure happiness, pain, customer satisfaction, teaching quality, job satisfaction, employee engagement, etcetera - in fact, just about everything. According to Peter Drucker, "What gets measured gets managed.” As informed consumers of statistical research we should be aware of the many ways that rating scales may be biased.

Some of the most interesting empirical research involves discrimination in the workplace. Some types of discrimination are illegal, others are immoral but not necessarily protected. Subtle forms of discrimination include not hiring people who are unattractive, overweight, short, have speech impediments, have strange names, and/or wear different kinds of clothing. Not all discrimination is overt or conscious. Those who wish to claim discrimination at work will likely have to provide statistical evidence to support their claims.