io9: “I Fooled Millions Into Thinking Chocolate Helps Weight Loss. Here’s How.”

Peer-reviewed publications is the best way that we’ve figured out for vetting scientific experiments and disseminating scientific knowledge. But that doesn’t mean that the system can’t be abused, either consciously or unconsciously.

The eye-opening article http://io9.com/i-fooled-millions-into-thinking-chocolate-helps-weight-1707251800 describes how the author published flimsy data that any discerning statistician should have seen through and even managed to get his “results” spread in the popular press. Some money quotes:

Here’s a dirty little science secret: If you measure a large number of things about a small number of people, you are almost guaranteed to get a “statistically significant” result. Our study included 18 different measurements—weight, cholesterol, sodium, blood protein levels, sleep quality, well-being, etc.—from 15 people. (One subject was dropped.) That study design is a recipe for false positives.

Think of the measurements as lottery tickets. Each one has a small chance of paying off in the form of a “significant” result that we can spin a story around and sell to the media. The more tickets you buy, the more likely you are to win. We didn’t know exactly what would pan out—the headline could have been that chocolate improves sleep or lowers blood pressure—but we knew our chances of getting at least one “statistically significant” result were pretty good.

And:

With the paper out, it was time to make some noise. I called a friend of a friend who works in scientific PR. She walked me through some of the dirty tricks for grabbing headlines. It was eerie to hear the other side of something I experience every day.

The key is to exploit journalists’ incredible laziness. If you lay out the information just right, you can shape the story that emerges in the media almost like you were writing those stories yourself. In fact, that’s literally what you’re doing, since many reporters just copied and pasted our text.

And:

The only problem with the diet science beat is that it’s science. You have to know how to read a scientific paper—and actually bother to do it. For far too long, the people who cover this beat have treated it like gossip, echoing whatever they find in press releases. Hopefully our little experiment will make reporters and readers alike more skeptical.

If a study doesn’t even list how many people took part in it, or makes a bold diet claim that’s “statistically significant” but doesn’t say how big the effect size is, you should wonder why. But for the most part, we don’t. Which is a pity, because journalists are becoming the de facto peer review system. And when we fail, the world is awash in junk science.

Who was kissing in the famous VJ Day picture?

We are approaching the 70th anniversary of VJ Day (August 14, 1945), which marked the end of World War II. And perhaps the iconic photograph of that day is the picture of two anonymous strangers kissing in New York City’s Times Square celebrating the end of the war.

And a question that is still unresolved after 70 years is: Who are they?

The short answer is, Nobody knows for certain. But in a clever bit of geometric and astronomical forensics, physicists at Texas State University (Donald Olson and Russell Doescher) and Iowa State University (Steven D. Kawaler) recently pinpointed the exact time that the photograph was taken: 5:51 pm, or about an hour before President Truman formally announced that the war was over. From the press release:

Overlooked in the right hand background of the photo is the Bond Clothes clock. The minute hand of this clock is clear, but the oblique angle of view and the clock’s unusually short hour hand makes a definitive reading of the time difficult. The clock might show a time near 4:50, 5:50, or 6:50 p.m. A prominent shadow falls across the Loew’s Building just beyond the clock, however, and this shadow could potentially give just as accurate a time reading as the clock.

Every tall building in Manhattan acts as a sundial, its cast shadow moving predictably as the sun traverses the sky. In this case, the Texas State team studied hundreds of photographs and maps from the 1940s to identify the source of the shadow, considering, in turn, the Paramount Building, the Hotel Lincoln and the Times Building. The breakthrough came when a photograph of the Astor Hotel revealed a large sign shaped like an inverted L that advertised the Astor Roof garden.

Calculations showed that only the Astor Roof sign could have cast the shadow, but to be certain, Olson and Doescher built a scale model of the Times Square buildings with a mirror to project the sun’s rays. The location, size and shape of the shadow on the model exactly matched the shadow in Eisenstaedt’s kiss photographs.

So who are the kissers? Again from the press release:

Over the years, dozens of men and women have come forward claiming to be the persons in the photograph. All have different stories, but the one thing they share in common is kissing a stranger in Times Square that fateful day.

“All those people have said they were there and identify themselves in the photograph,” Olson said. “Who’s telling the truth? They all could be telling the truth about kissing someone. They were probably all there, and kisses were common in Times Square on VJ Day.

“I can tell you some things about the picture, and I can rule some people out based on the time of day,” he said. “We can show that some of the accounts are entirely inconsistent with the astronomical evidence”…

“Astronomy alone can’t positively identify the participants, but we can tell you the precise moment of the photograph,” Olson said. “Some of the accounts are inconsistent with the astronomical evidence, and we can rule people out based on the position of the sun. The shadows were the key to unlocking some of the secrets of the iconic VJ Day images–we know when the famous kiss happened, and that gives us some idea of who might or might not have been in the picture.”

From a news report:

“There are probably 50 or 60 sailors who have come forward and say, ‘That’s me! I’m the guy in the photograph.’ Fewer women, maybe five or six women, have said they’re the woman in white. There are articles all over the internet advocating for one [or] the other,” Olson said.

Olson can’t say who is correct, but he can rule out a few.

“What we can do is calculate the precise time, 5:51 p.m., when the photograph was taken. That does appear to rule out some of the widely accepted candidates,” he said.

The full article has been published in the August 2015 issue of Sky and Telescope magazine (sorry, you’ll have to buy a copy in you want to read the article). I also recommend clicking through the photographs in the press release; the captions of the photographs give many details of how the time of 5:51 pm was pinpointed.

15-year-old catches math error at the Boston Museum of Science

See the full article here: http://www.boston.com/news/local/massachusetts/2015/07/06/year-old-catches-math-error-the-museum-science/awhREamdn1KRg7nz2gGyPO/story.html?p1=Must_Reads

Arrangements of Stars on the American Flag

Reasonable star patterns on the American flag correspond to special factorizations; the density of such factorizations is less than the density of values in a multiplication table; Paul Erdös showed this density asymptotically approaches zero by considering the average number of prime factors of an integer. – See more at: http://www.maa.org/programs/maa-awards/writing-awards/lester-r-ford-awards/arrangements-of-stars-on-the-american-flag#sthash.e9PHpilF.dpuf

Read the article here: http://www.maa.org/programs/maa-awards/writing-awards/lester-r-ford-awards/arrangements-of-stars-on-the-american-flag

Issues when conducting political polls (Part 2)

The classic application of confidence intervals is political polling: the science of sampling relatively few people to predict the opinions of a large population. However, in the 2010s, the art of political polling — constructing representative samples from a large population — has become more and more difficult.

The Washington Post recently ran a feature about how the prevalence of cellphones was once feared as a potential cause of bias when conducting a political poll… and what’s happened ever since: http://www.washingtonpost.com/blogs/the-switch/wp/2014/11/03/pollsters-used-to-worry-that-cellphone-users-would-skew-results-these-days-not-so-much/.

Previous post: Issues when conducting political polls.

Why Do We Pay Pure Mathematicians?

I really enjoyed this article from Math With Bad Drawings: http://mathwithbaddrawings.com/2015/02/24/why-do-we-pay-mathematicians/

Value Added Meets the Schools: The Effects of Using Test-Based Teacher Evaluation on the Work of Teachers and Leaders

The March 2015 issue of Educational Researcher was devoted to the perceived usefulness/uselessness (depending on the perceiver) of high-stakes testing. The issue contains multiple perspectives from teachers, principals, and education researchers. The abstract from the journal’s editors sets the tone for the issue:

Teacher accountability based on teacher value-added measures could have far-reaching effects on classroom instruction

and student learning, for good and for ill. To date, however, research has focused almost entirely on the statistical

properties of the measures. While a useful starting point, the validity and reliability of the measures tell us very little

about the effects on teaching and learning that come from embedding value added into policies like teacher evaluation,

tenure, and compensation. We pose dozens of unanswered questions, not only about the net effects of these policies on

measurable student outcomes, but about the numerous, often indirect ways in which these and less easily observed effects

might arise. Drawing in part on other articles in the special issue, we consider perspectives from labor economics, sociology

of organizations, and psychology. Some of the pathways of these policy effects directly influence teaching and learning

and in intentional ways, while other pathways are indirect and unintentional. While research is just beginning to answer the

key questions, a key initial theme of recent research is that both the opponents and advocates are partly correct about the

influence of these policies.

Math anxiety

I recently read the following interesting articles on math anxiety from the perspective of cognitive scientist: http://www.danielwillingham.com/daniel-willingham-science-and-education-blog/math-anxiety-what-it-does-and-what-can-be-done and http://www.aft.org/pdfs/americaneducator/summer2014/Beilock.pdf. I don’t have a particular opinion on this, as my own expertise is on post-secondary students, but this definitely gave me some food for thought.

Talking about math to Congress

From MAA Focus:

If you think it’s hard to distill research results into a 15-minute conference presentation, try this: Choose a subject like matrix factorizations or recent progress on the twin prime conjecture. Figure out how to make a nonexpert audience—members of Congress, say—if not fully understand the chosen topic, at least appreciate its significance. Do this in a minute. The clock is ticking.

Jerry McNerney of California’s ninth congressional district has risen to such a challenge more than 10 times in the U.S. House of Representatives, where he has served since 2007. The only current member of the House or Senate to hold a doctorate in mathematics (University of New Mexico, 1981), McNerney has read into the congressional record one-minute expositions of such abstruse subjects as vector bundles, synesthesia, and the Large Synoptic Survey Telescope…

Boiling down complex material into a minute of talking is tricky, McNerney concedes, but he has been pleased with the results. As a member of the Public Face of Mathematics panel at the 2014 Joint Mathematics Meetings, McNerney told listeners that being coaxed into thinking about math has a positive effect on his congressional fellows.

“Instead of all the usual bickering that you get on the House floor, everyone smiles,” he reported. “They say, ‘This is really fun.’”

Source: http://mcnerney.house.gov/media-center/in-the-news/math-by-the-minute-on-capitol-hill

Gender Stereotypes in STEM

From the article Gender Stereotypes in STEM:

The study, which compared white and black women’s participation in and perception of STEM fields, found that black women were more likely than white women to show an interest in studying STEM disciplines when they enter college.

The research also shows that African Americans were less likely than white Americans to view STEM programs as masculine, which may help explain why the participation levels vary between the two ethnic groups.

The authors argue that race and ethnicity influence the gender stereotypes that women hold, which in turn influence their interest in the sciences, said Laurie O’Brien, an associate professor of psychology at Tulane University and one of the article’s lead authors.

Despite the findings of higher initial interest reflected in the journal article, other data show black women are underrepresented in the number of STEM bachelor’s degrees actually earned, according to the paper.