Tuesday, September 25, 2012

Statistics as a principled argument

These past few days I've been reading Statistics as Principled Argument, by Yale's Robert P. Abelson. I've not finished yet, but I can say already that it's an enjoyable and informative book. This is not your standard how-to manual. Rather, it's a candid discussion about why people in general are afraid of statistics —and about why they should not be, as they are crucial for rational thinking— based on many examples of good and bad practices.

Abelson made the case for statistics as a discipline related to rhetoric, narrative, reasoning, and even journalism (see quote below). In an article about him, authors Ira J. Roseman and Stephen J. Read wrote: "Statistics as Principled Argument situated the process of statistical analysis within the overall enterprise of making arguments that are appropriately based on data, and fleshed out general criteria (magnitude, articulation, generality, interestingness, and credibility) for the persuasiveness of empirical claims."

You may understand, then, why I am finding this book enticing. It connects with some of the main points in The Functional Art and the current growth of data journalism. It's true that you need to know a bit about statistics to read it (every now and then it's forcing me go back to a textbook I have in my Kindle to clarify certain concepts), but not that much. Here you have some paragraphs from the intro and the first chapter that can give you an idea of the book's unpretentious tone:

Data analysis should not be pointlessly formal. It should make an interesting claim; it should tell a story that an informed audience will care about. (...) Meaningful research tells a story with some point to it, and statistics can sharpen the story. Students are often not mindful of this. Ask a student the question, "If your study were reported in the newspaper, what would the headline be?" and you are likely to receive in response a rare exhibition of incoherent mumblings, as though such a question had never been remotely contemplated.
The typical statistics course does not deal very well, if at all, with the argumentative give-and-take nature of statistical claims. As a consequence, students tend to develop their own characteristic misperceptions of statistics. They seek certainty and exactitude, and emphasize calculations rather than points to be drawn from statistical analysis. They tend to state statistical conclusions mechanically, avoiding imaginative rhetoric.
The purpose of statistics is to organize a useful argument from quantitative evidence, using a form of principled rhetoric. The word principled is crucial. Just because rhetoric is unavoidable, indeed acceptable, in statistical presentations does not mean that you should say anything you please.
For many students, statistics is an island, separated from other aspects of the research enterprise. Statistics is viewed as an unpleasant obligation, to be dismissed as rapidly as possible so they can get on with the rest of their lives (...) Students become rule-bound, thinking of statistical practice as a medical or religious regimen. They ask questions such as, "Am I allowed to analyze my data with this method?" in the querulous manner of a patient or parishioner anxious to avoid sickness or sin, and they seem to want a prescriptive answer, such as "Run an analysis of variance, according to the directions of the computer package, get lots of sleep, and call me in the morning."
And my favorite: "A null hypothesis test is a ritualized exercise of devil's advocacy."