Big Data, Plainly Spoken (aka Numbers Rule Your World)

Neat reads 9

Two recent articles worth reading:

Decoding your email personality, NYT: (link)

This describes the field called "forensic linguistics", the use of statistics to link (or delink) writings from their writers. Mark Zuckerberg is being sued again, this time by one Paul Ceglia who has produced emails (saved as Word documents) that show that he may own half of Facebook. Mark hired a linguist who argues that he can show that those emails are "unlikely" to have been written by Mark. Language Log has some additional insights here.

This is another fascinating example of statistics used in real life. I find it hard to believe that in this case, it would be possible to say with acceptable confidence whether Mark is the real author or not.

Counting calories? Your weight-loss plan may be outdated, NYT (link)

Some new conclusions from a long-term study that tracks people for decades to try to understand the link between food consumption and weight gain/loss. I'm going to need to find time to read the underlying study but certain aspects of this class of health studies have always bothered me.

First, the results are often expressed as: milk is good, cookies are bad, butter is ok, etc. This presentation gives the impression that all components of a diet are independent of each other. If milk will rise my weight by 0.2 percent and cookies by 0.3 percent, then if I take both, my weight would go up by 0.5 percent. Unfortunately, things are not so simple. In this example, cookies and milk are often consumed together; someone who chooses to eat fewer cookies is likely to also drink less milk.

Second, these studies report averages for the entire population but surely there are subgroups that differ from the average person.

 

 

Neat reads 8

The New York Times reported on a bit of controversy about the use of statistical techniques to prove the existence of ESP. Here's a profile of the researchers who claimed to have proven ESP, and here's an overview of the statisticians who disagree. The statistical discussion is rather technical, and Andrew Gelman has a good post on it. My view is similar to Andrew's: the real issue is how to use statistical methods to measure effects that are really small. Indeed, is this a solvable problem? If the "noise" in the data is much stronger than the signal we're trying to detect, is it possible to detect the signal with a reasonable level of accuracy?

***

You are not alone in noticing that Google search results have been invaded by spam. It is now often the case that the first-page results consist of nothing but commercial websites, spam websites, and unreliable content created by several companies set up specifically to game Google's algorithms. Google is either losing the battle with spammers or content to make money from whomever.

See Paul Kedrosky's pieces here and here. Unlike Paul, I don't think curation is a standalone solution; it is not scalable to the multitude of possible searches. Smart algorithms with human input is our only option. This is in its essence the same problem as the email spam problem.

Here is a summary of three articles on this subject. 

In Chapter 2 of Numbers Rule Your World, I briefly mentioned why it is a horrible idea for regulators to force banks to reveal their credit-scoring algorithms. Doing so would lead to the kind of gaming that has harmed Google's search algorithms.    

Neat reads 6

The illusion of control, Bad Science (link)

Must-read. This area of research sounds truly amusing. Points again to the importance of measuring the placebo effect.

Mystery: what happens when neuroscientists, psychiatrists, and business executives get together in a room? (link)

Someone at Wired is obsessed with why expensive medicines don't work. Last year, Steve Silberman proposed that placebo effect is getting stronger and in the way of meds. I made some dissenting comments here. Now, Jonah Lehrer tells us there is a New Yorker article claiming that "disturbing" false positive results threaten the "scientific process".  My frustration with this whole storm in a teacup is that the participants spend all their time on secondary issues (such as the placebo effect being "too strong", "our facts losing their truth", "claims enshrined in textbooks suddenly unprovable", too much natural variability) while ignoring the most obvious hypotheses (such as medicines not good enough, effect sizes too small to be measured properly, clinical trials standards generate too many false positives, clinical trials poorly designed, conflict of interest leading to insufficient attention to side effects, fast track and other politically-driven standards that usurp the scientific process, use of medicines beyond the originally targeted groups).

The quoted passage ends with this truism:

...experiments are essential because they "put nature to the question". But it appears that nature often gives us different answers.

This natural variability is posed as a challenge to the principle of "replication" in science. If this is the mainstream view, we are back in the Stone Age. It is precisely because of variability that replication is necessary. If every experiment would yield the same outcome, we only need one measurement. If pharmas spend billions of dollars to develop medicines that have effects so tiny they can't be measured without extraordinary efforts, then no amount of truth-finding can hide these bad investment decisions.

Statistics invade the humanities, according to the New York Times (link)

Can such analyses yield generalizable results?

getstats: a 10-year plan to improve statistical literacy (link)

I applaud.

 The Joy of Statistics, a BBC documentary (BBC link, Open University link)

Related to the getstats campaign. Hans Rosling, creator of Gapminder (absorbed by Google) is the star here. Looks like the documentary focuses on how to find insights from big databases of public data. Hopefully, it will cover how such insights can directly improve our lives. As I have been saying in my talks, it is not sufficient for statistics to be fun, it needs to be impactful.

 

Neat reads 4

Here are some stuff I've read recently:

Is your laptop cooking your ...?, Reuters

As I'm writing this, I'm doing a very bad thing to my fertility -- this is if you believe story time by the people involved in this study. How should you read this article? No control group. 29 total participants (tiny sample). No mention of how they were selected, or why "young" was a necessary criterion for selection. Most important of all, the study measured the increase in temperature, but the link between temperature increase and fertility is unproven fiction. Such a link would be complicated to establish as I can imagine that some sustained level of use may be required to kill sperms.

How the Internet correctly predicted 36 of 37 Senate races, Randall Lane in Business Insider

In which Mr. Lane from the Daily Beast boasted about its "election oracle." What's good: innovative use of online text mining to help make election predictions; nice application of statistics. What's bad: the headline claim of "the Internet" predicting is badly misleading; scientists used data collected from the Internet to make predictions. Should you believe this? Accuracy rate has a denominator: as he discussed, it depends on whether you count all races or only close races. But what about races so close that the results are within the errors of measurement? What does it mean to have a predictive algorithm that predicts with "high accuracy" such essentially tied races?

Interest payments are not profits, Dean Baker

Baker reacted to this ridiculous NYT article that showed up over the weekend, proclaiming profit-taking in the Fannie/Freddie bailout. Say: your brother has fallen on hard times, and comes to you for help; you agree to provide him a loan, and at the family rate; turns out you don't have excess cash lying around so you go to the bank to get a loan; your brother pays you interest; you think you've made a profit from this arrangement. So many things wrong here. You have not gotten your principal back yet. And the profitability of this arrangement was determined long ago: it is the differential between the rate you are paying the bank and the (family) rate your brother is paying you. Lost in all of this: the government does not have free money; there is a cost to money-printing.

 

Neat Reads 3

A chart that can save your life

Vitamin A is commonly added to sunscreens because of its supposed anti-aging effect but an FDA study from ten years ago showed that Vitamin A accelerates the growth of cancerous tumors in rats.

Moral hazard: people who buy high-SPH sunblocks tend to stay out in the sun longer because they think they are better protected.

Lab conditions versus reality: people who buy high-SPH sunblocks fool themselves in a different way; they apply only a quarter of the recommended amount, which means that the protective effect reported by the manufacturer is vastly overstated.

Comments on the graphs can be found on Junk Charts.

Why this British doctor has been banned, and what is he doing in the U.S.?

Using a Freakonomics-style argument, one can say that Dr. Andrew Wakefield may have endangered lots of children. He was the one who published discredited research that purportedly linked autism to the combined vaccine for measles, mumps, rubella (MMR).

As a result, vaccination rates have dropped (roughly from 90% to 80% in the U.K.), and measles have made a comeback in Western countries, with worrisome consequences (from under 100 cases to 1400 cases). But note that the 10-fold increase most likely came from the 10% who switched from the vaccinated to the unvaccinated category. There have, thankfully, only been a few deaths.

In the wake of the controversy, Dr. Wakefield moved to Texas but has recently left the clinic he founded.

Several attempts to replicate his research have failed. He also was found guilty of various counts of unethical conduct, including testing a new vaccine on a kid without permission, and taking blood samples from unsuspecting kids attending his son's birthday party (by offering 5 quid each).

The original Wakefield study had a sample size of 12.

Ben Goldacre of the Guardian did exemplary work in bringing attention to the MMR scare in the UK. He believed that the blame should be placed squarely on the media for promulgating Dr. Wakefield's "research" for years while ignoring available evidence to the contrary.

Martin Gardner, 1914-2010

Brian Hayes remembers a man who entertained many with mathematical puzzles.

Jacques Bertin, 1918-2010

Reader from France Bernard L sent in this note:

It is with great sadness I've learnt of the recent departure, early May at the age of 92, of Jacques Bertin, author of the Semiology of graphics

Through his work he laid down the foundation of information visualization.

I'll keep the fond memories of the time I've spent with him when he accepted to preface my book, of his wits and ever amused child gaze when we discussed the data visualisation topics. He left us for a new territory to charts and maps...

Neat Reads 2

How Privacy Vanishes Online, by Steve Lohr, New York Times, Mar 16 2010

I agree with everything he says, except this sentence: "So far, this type of powerful data mining, which relies on sophisticated statistical correlations, is mostly in the realm of university researchers, not identity thieves and marketers." Having worked for a credit card company, I know that many identity thieves are smart people with PhDs, so if the smart PhDs are doing this at universities, it is a certainty ...

Driveby Culture and the Endless Search for Wow, by Seth Godin, Seth's Blog, Mar 16 2010

This post confirms why I read Seth Godin; he's a marketer with brains. He asks: "Should I write blog posts that increase my traffic or that help change the way (a few) people think?"

ChatRoulette Explained in 357 Seconds, by Nicholas Carlson, Business Insider, Mar 17 2010

Seth's blog post introduced me to ChatRoulette. Well-produced video about this new craze, with some less-than-scientific survey research thrown in for fun.

21 Examples of Flash Journalism, by Mindy McAdams, Teaching Online Journalism, Mar 10 2010

Extremely rich post which will be my weekend reading. (Via Andrew Gelman and Aleks J)

Neat reads

Note: Once in a while, I will provide links to other blogs that contain relevant materials, with limited commentary. In so doing, I also hope to introduce some of the blogs I admire.

Is it okay to ignore results from people you don't trust? (Bad Science)

This is from Ben Goldacre's blog, Bad Science, which debunks charlatans and bad medical advice.  Goldacre is a UK-based writer, trained as a doctor. I'm not interested in his lede question but extremely grateful that he introduced the paper by Cataldo et. al. 

The paper reviews all available research on the link between smoking and Alzheimer's, for each author, determines whether he or she has links to the tobacco industry, and discovers that "the risks of Alzheimer's associated with smoking reported by [those researchers linked to tobacco] were on average about a third lower than those conducted by other researchers."

In a recent interview, I said: "It’s really not just about the numbers. It’s about the people who are using these numbers." This research demonstrates why we need to pay attention.

***

Eggs-actly what ARE the chances... (Daily Mail)

I can't believe I am linking to the Daily Mail. I found this on David Spiegelhalter's Understanding Uncertainty blog so that's my excuse. This report of someone discovering a box of six eggs all of which had double yolks understandably caused a stir.  In this article, someone who knows probability theory explains why this happens with a higher chance than we expect. David has a few more comments on his blog.

The lesson: in real life, the beloved mathematical assumption of independent events rarely holds.