One word: statistics

In another sign of the growing interest in the field, an estimated 6,400 people are attending the statistics profession’s annual conference in Washington this week, up from around 5,400 in recent years, according to the American Statistical Association.

The attendees, men and women, young and graying, looked much like any other crowd of tourists in the nation’s capital. But their rapt exchanges were filled with talk of randomization, parameters, regressions and data clusters. The data surge is elevating a profession that traditionally tackled less visible and less lucrative work, like figuring out life expectancy rates for insurance companies.

That is the New York Times’ idea of a sales pitch for more undergraduate statistics.

Even so, the number one piece of advice I give my undergraduate advisees: don’t graduate without (at least) a solid grounding in multivariate regression and experience manipulating data. This advice is needed; with the exception of economics, not a single social science degree at Yale requires a stats credit.

This is not simply self-serving advice. Whether you want to understand the latest political poll, medical discovery, or business proposition, a basic grasp of stats is essential. If you want to understand why most polls, medical analyses and business decisons are flawed, then you need to go even further.

Don’t believe me? Here is Peter Orszag (Obama’s budget director) on why stats is essential for good public policy. And here is Tim Harford musing on China’s latest morale-boosting campaign dubbed “Statistical Feelings”.

Possibly those feelings get a little too gushy. China’s Bureau of Statistics even offers a love poem to the discipline, Love the Homeland, Love Statistics:

Life
Some mock me for doing statistics
Some loathe me and statistics
Some don’t understand what statistics are
Why is it that statistics
Put a calm smile on my face?

Personally, I eschew the stats poetry and buy all my research assistants a copy of this econometrics introduction. Any recommendations for equally user-friendly intro books?

13 thoughts on “One word: statistics

  1. sometimes I find myself hugging Wooldridge’s undergrad text and weeping with gratitude.

    I think I shall submit this post anonymously

  2. tell you what I wouldn’t recommend: Basic Econometrics by d. Gujarati. The best way to put anyone off econometrics.

  3. Mostly Harmless Econometrics is my latest purchase. Not for beginners but for those staring down the barrel of a life sentence of applied research, it supposed to be an excellent read. (This is presumably known as a predictive comment, rather than a descriptive one.)

    Plus the book has lots of Douglas Adams quotations, if all else fails.

  4. The focus in education (high school and college), at least in the US, seems to be on calculus… I’m not sure why. I can’t imagine what the average non-math/science student gets out of calculus I. (If that’s even required – at my university, most majors can get away with taking only ‘finite math.’) To me, it would make far more sense to instead emphasize probability and statistics.

  5. Actually, the psychology major at Yale requires its students to take statistics in order to graduate.

  6. Could someone comment on the prior knowledge required to understand the Kennedy book? Is it possible to pick up the math on the way, or to learn it from the appendices?

  7. I personally don’t like the Kennedy book. I think the undergrad Wooldridge book is fantastic.

  8. Wooldridge’s undergraduate text, Introduction to Econometrics, is probably the clearest primer on the basics of econometrics– it covers a wide-range of the bread and butter topics without using matrix algebra. There are tons of useful examples and online data sets that you can use to reproduce results. The appendixes also provide useful background to most basic mathematical and statistical tools econometricians use…

  9. Hear, hear! Do encourage people to study this more. I earned my BA and MA in history; statistics was optional and I didn’t take it. Almost nobody did. I regret it regularly and wonder how I can develop the skills… I think I need to be shown some exercises where I can answer questions – even simple ones – about the world (maybe using some free census data to start with).

  10. what about a book anda a paper. the book: “statistics for social data analysis”, by knoke et al.. it covers from basics [descriptive stats] until loglinear and structured equation models. the paper: “how not to lie without statistics”, by king and powell. it offers a good discussion about common misunderstandings in best practices in qualitative research.