Posts filed under General (3152)

Team Ratings for Round 6

The basic method is described on my Department home page.

Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Performance So Far

So far there have been 40 matches played, 21 of which were correctly predicted, a success rate of 52.5%.
Here are the predictions for last week’s games.

Predictions for Round 6

Here are the predictions for Round 6. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.

From the Herald:

US researchers have confirmed a strange link between touching rough surfaces and feeling for others, which could help charities raise more money.

Based on my usual complaints about this sort of claim, you might expect that the research didn’t look at donating money  or that it saw only a tiny difference. No.

There were five experiments, but only one that involved actual money. People were approached on the street and given a description of  a health-related charity, and asked to donate. One charity was real, working in a familiar disease; the other was fake, working in a real but obscure disease (all the money actually ended up with the real charity).  Half the participants were given the information and donation envelope on a clipboard with rough sandpaper on the back; the other half weren’t.

When asked to donate to the National Breast Cancer Foundation there was no difference between the rough and smooth clipboards (as you’d expect). When asked to donate to the National Sjögren’s Foundation, 10/34 with sandpaper-backed clipboards said yes compared to only 1/32 with smooth clipboards.

I’m going to go very slightly out on a limb here to say there is no way this ten-fold increase is a real and generalisable phenomenon.  So, what went wrong?  Part of the problem is what Andrew Gelman calls the ‘garden of forking paths’, after the Jorge Luis Borges story — there are many, many possible analyses and they don’t all show this dramatic difference.

For example, there wasn’t a difference in donation probability with the familiar charity. This was consistent with the researchers’ theory, but I’m pretty sure if there had been a difference the researchers wouldn’t have considered it as evidence refuting the theory. Also, the researchers note that they didn’t see a difference in donation amount with the sandpaper, just in donation probability.

Also, if you assume the ten-fold increase was overestimated even a bit, you then get into the problem of sample size. Suppose that the effect was only a two-fold increase rather than ten-fold. That still seems implausibly large to me, but the comparison would then be something like 2/34 vs 1/32 and would be completely unimpressive.  You’d need a sample size something like ten times larger.  And that’s if a bit of sandpaper on the back of a clipboard doubled the number of people who donated.

Still, these findings could have “significant implications for less well-known charities”, as the researchers suggest. If I got approached by a charity using sandpaper on the back of their clipboards, I would tend to think they were (a) poor at evaluating evidence, and (b) not all that honest. I could see that having an impact.

Q: Did you see vegetarian diets cause cancer now?

A: No.

Q: The Herald site front page: headline Vegetarianism can lead to cancer?

A: No

Q: The teaser: “Scientists have found there can be long-term health risks associated with a vegetarian diet, that could outweigh the benefits.”?

A: Well, it depends on what you mean by ‘long-term’, for a start.

Q: How long-term?

A: Centuries, perhaps thousands of years.

Q: How did they find people who were thousands of years old? And why isn’t that the headline?

A: Not people.

Q: I refuse to believe in century-old lab mice.

A: Human populations.

Q: Ok, so if we click through to the story (from the Telegraph) it seems they’re saying your great-grandparents eating lettuce gives you harmful mutations?

A: That’s what the story says, but it’s not what the research says. The research suggests that a mutation that with a modern diet might increase cancer risk arose randomly a long time in the past and became common in a South Asian population where vegetarian diets have been common.

Q: How did the mutation become common?

A: Because it wasn’t true that the long-term health risks outweighed the benefits — there’s genetic evidence of  ‘selection’ in the evolutionary sense, meaning that people with the mutation had more descendants on average.

Q: How much health risk did they find?

A: They weren’t looking at health risks

Q: But “long-term health risks” and “can lead to cancer”?

A: Sadly, yes.

Q: Ok, what were they looking at?

A: They were looking at enzymes that turns one type of fatty acid into another. The mutation makes it easier for the body to synthesis long polyunsaturated acids

Q: Aren’t they good?

A: Some of them, like the DHA and EPA also found in fish, are thought to reduce inflammation and heart disease. But arachidonic acid is thought to increase inflammation, though the American Heart Association isn’t convinced

Q: That’s heart disease. What about cancer?

A: The only links to cancer are pretty speculative — that the mutation could reinforce effects of modern diet in increasing cancer risk.  The contribution of arachidonic acid to that is controversial. But it could be real.

Q: Is there actually a higher cancer rate where they got their vegetarian population from, compared to the control population?

A: No.

Q: That ‘arachidonic acid’ thing. Why does that make me think of spiders?

A: Yes, me too. It’s a false cognate: Latin ‘arachis‘, ‘peanut’, not the mythic Greek technologist Aράχνη that arachnids were named for.

Team Ratings for Round 6

The basic method is described on my Department home page.

Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Performance So Far

So far there have been 41 matches played, 27 of which were correctly predicted, a success rate of 65.9%.
Here are the predictions for last week’s games.

Predictions for Round 6

Here are the predictions for Round 6. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.

Team Ratings for Round 5

The basic method is described on my Department home page.

Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Performance So Far

So far there have been 32 matches played, 17 of which were correctly predicted, a success rate of 53.1%.
Here are the predictions for last week’s games.

Predictions for Round 5

Here are the predictions for Round 5. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.

• “Unlike other projects that map cities by sound, Chatty Maps isn’t measuring volume or ranking neighborhoods as noisy or quiet. Instead, it shows the city across a spectrum of different sounds—as well as the emotions we associate them with.” I’m not convinced, but it’s interesting to look at. (via @teh_aimee)

• “World Cup fans not responsible for the Zika outbreak”. (Scientific American blog, open-access research paper)   I think ‘responsible’ is the wrong word, but in any case, looking at the genomes of Zika virus specimens suggests that the current virus has been circulating in the Americas since 2013 at least. Also, the three samples from microcephaly cases don’t share any relevant mutation, so the more-severe disease in the current outbreak probably isn’t due to a change in the virus.  You can do a lot with genetics.

• “Can an algorithm be wrong?” from limn.it

• “Exposing algorithms” from the Tow Center for Digital Journalism: a summary from a session at the National Institute for Computer-Assisted Reporting conference

• Also via the Tow Center, “The Curious Journalist’s Guide to Data” by Jonathan Stray. A book for journalists, but also relevant to consumers of journalism.

Team Ratings for Round 5

The basic method is described on my Department home page.

Here are the team ratings prior to this week’s games, along with the ratings at the start of the season.

Performance So Far

So far there have been 33 matches played, 19 of which were correctly predicted, a success rate of 57.6%.
Here are the predictions for last week’s games.

Predictions for Round 5

Here are the predictions for Round 5. The prediction is my estimated expected points difference with a positive margin being a win to the home team, and a negative margin a win to the away team.

• Many people have hypothesized, plausibly, that giving people risk estimates for disease based on genetics would encourage them to behave more healthily. The available evidence isn’t supportive.

• The Herald, for the second time in a week or so,  treats a bogus poll as supporting evidence for something.

• Expensive but extremely effective hepatitis C drugs: an example of the sort of thing Pharmac might well want to fund ahead of Keytruda.

• Early newspaper data graphics for the 1849 cholera epidemic

• Barfoot & Thompson surveyed a bunch of young non-home-owning people in Auckland to ask them what they expected to pay for a house. I would have thought knowing more about housing prices than your customers was part of the point of having real estate agents. The Herald put it on the front page, too.

• In the US, doctors who get more money from drug and device companies tend to prescribe expensive drugs more.  Note: not only is this correlation, but conceivably it could be a good thing — perhaps they are just more knowledgeable.  Or, of course, not.

Q: Did you see snake venom can cure Alzheimer’s Disease now?

A: I saw the story in Stuff.

Q: Do you have to get bitten by the pretty green snake?

A: No, you don’t, though you’d probably need the compound from the venom injected into your spine. And the snake isn’t green.

Q: So it’s mice? It looks green

A: The photo is of the wrong snake, and the research isn’t even in mice. The last sentence of the story says “The treatment will now be trialled in mice before it can be considered as a viable treatment for humans.”

Q: But it could work?

A: It could, though so far treatments that try to dismantle amyloid plaques have ranged from ineffective to actively harmful in treating the actual disease. There’s even a respectable hypothesis that amyloid starts off as a protective response against lurking bacteria or viruses that activate in the brain in old age. It’s all very unclear and depressing.

Q: But there seem to be lots of natural products that are promising cures. It’s not just the snakes, look at the ‘Read More’ links from the story:

A: The chocolate link isn’t about Alzheimer’s at all. The maple syrup story talks about compounds in maple syrup, and the need for ‘further animal trials’. The ‘cheap pill’ trial “did not investigate whether resveratrol has any effect on memory or improving other symptoms of Alzheimer’s disease.”  The blueberries had a small effect in one tiny, short-term trial and a different small effect in another tiny, short-term trial.  And the sleep disruption isn’t something you could do much about, even if it is more than correlation.

Q: So why are there all these unreliable or preliminary reports being published?

A: Because there isn’t any other good news. If you have a really hard problem, nearly all the people who say they have solutions will be wrong, and won’t have checked their solutions thoroughly.

Q: I see. It’s like aliens.

A: <blinks> Huh?

Q: Lots of serious researchers are looking for alien life, but all the people who say they’ve actually found it are talking about crop circles and UFOs, and everything else is just like “we’ve found a planet that’s not as far from being habitable as the previous ones were”

A: Pretty much.

From the ‘Parenting’ section of the Stuff Life & Style page:

That’s both wrong and implausible.  If Dravet syndrome, a serious epileptic condition, was about as common as, say, autism spectrum disorder, you’d have heard of it already.

The actual rate is about 1 in 20,000, two hundred times lower than the teaser says. If you click through to the story and read it carefully you’ll see that Dravet Syndrome is responsible for about 1% of childhood epilepsy.

So, how did the numbers get so badly messed up? Well, one contributing factor is probably that the story was taken from The Conversation, and whoever did the editing job didn’t read it carefully enough.  As seems to often happen with pieces taken from The Conversation, there’s no attribution either to the original publisher or the authors, and all but two of the nine links in the original have been scrubbed.

The Conversation encourages republication of the pieces they publish, but the Creative Commons license they use requires that republishers attribute the piece and indicate if changes have been made.  I don’t know if the NZ news sites have negotiated an alternative deal, but I can’t see why lack of attribution would be desirable — I thought the by-line was as sacred to journalists as to academics.