Posts filed under General (3156)

XKCD, reminding us of the difference

Dr Mark Apperley, from Uni of Waikato, with Phil O’Brien

“Big Data” isn’t just big, it’s unstructured.

(via Kate Hannah)

An NZ wine maker has produced wines with high levels of the chemical resveratrol. In Stuff’s story

“We can’t say that alcohol is healthy, but we can say resveratrol is,” co-director Kathleen Corsbie said.

Actually, the evidence for health benefits of lowish levels of alcohol consumption compared to none is much, much stronger than the evidence for any benefit of resveratrol, because resveratrol is almost unstudied in human disease.

Stuff’s story is well done, with useful background and expert commentary, eg,

The French, who eat fatty foods such as cheese and pate and also drink a lot of red wine, are the most commonly used example, Holt said.

“They have one of the lower rates of heart disease.”

But saying the wine will protect drinkers against cancer and other diseases is an exaggeration, Holt said.

It’s a respectable, though far from proven, hypothesis that resveratrol has some part in lower French rates of heart disease, but as he points out, that’s not true for cancer (the French actually don’t have low cancer rates compared to the rest of Europe, for example).

If you go to the Southern Wines website, you find more detail about their claim

The human body uses many complex biochemical pathways and reactions to function; but these reactions result in waste products such as free radicals (molecular compounds that contain an extra unpaired electron). These free radicals are the body’s terrorists and cause biological havoc which helps contribute to our degenerative ongoing diseases such as cancer, dementia, diabetes, vascular disease (heart attack and stroke), macular degeneration (most common cause of blindness in people over 65) and arthritis.

So, they are claiming a reduction in a large collection of aging-related diseases, based on anti-oxidant effects.  In this context it’s useful to note that while there have been no large studies of resveratrol in humans, there have been large randomised trials of anti-oxidant vitamins (beta-carotene, vitamin C, vitamin E) that have, by and large, not found benefits.  For cancer, they have sometimes even found harm, perhaps because free radicals are an important part of the immune system’s armory.

This is the sort of case that makes setting advertising standards tricky.  There’s no question that resveratrol is an antioxidant.  Presumably the manufacturer knows how much resveratrol is in their wine, though saying it’s “40 times higher” as they do is harder to defend given the more-than-20-fold range in resveratrol concentrations in regular red wines (PDF).  There’s moderate evidence that resveratrol is good for lab animals. In humans, almost nothing is known directly — there’s slight evidence for reductions in some heart disease risk factors — but there’s strong evidence that some other antioxidants don’t have large benefits.

Southern Wines co-director is quoted as saying

“One glass of our Balancing Act wine is equivalent to consuming the resveratrol contained in around 40 glasses of normal wine, so why would you drink anything else?”

As with chocolate, the StatsChat advice would be that if you are drinking wine primarily for the health benefit, you’re doing it wrong.

• If you’re going to write a story about one of Satoshi Kanazawa’s publications, you might try Googling him first. (The fact that he lived in ChCh for a year is not the most salient piece of his background)
  
• The Daily Mail has been publishing a few interesting science stories recently. Apparently not writing them, though.
• Nate Silver talked to thousands of statisticians at the Joint Statistical Meetings in Montreal this week, about principles for data-based journalism

Some South African readers have asked for Currie Cup predictions and since the methodology should work, I have decided to provide them. Interestingly, the home ground advantage in past games appears to be large, and the predictability of games has varied enormously. Last year’s competition was very hard to predict it seems.

Team Ratings for Round 1

Here are the team ratings prior to Round 1, along with the ratings at the start of the season. I have created a brief description of the method I use for predicting rugby games. Go to my Department home page to see this.

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

Predictions for Round 1

Here are the predictions for Round 1. 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 22

Here are the team ratings prior to Round 22, along with the ratings at the start of the season. I have created a brief description of the method I use for predicting rugby games. Go to my Department home page to see this.

Performance So Far

So far there have been 152 matches played, 92 of which were correctly predicted, a success rate of 60.53%.

Here are the predictions for last week’s games.

Predictions for Round 22

Here are the predictions for Round 22. 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.

• A good story in the Herald about a survey of health risk behaviour in NZ high-school students. Things are improving, as the story says. There’s even a link to the real research.  Pity about the headline.
• Short free-access piece in the Listener on estimating and understanding risks of death.  There’s a longer cover story in the paid-access or squashed-tree versions, also by Mark Broatch.
• The NZ Science Media Centre has a new edition of its Desktop Guide for Science Reporters. It’s also valuable for people reading science stories — especially those by journalists who haven’t read the guide.

What are randomised trials and why are they important: one of a set of videos made for the 100th anniversary of the UK Medical Research Council

Also, a post about public randomisation ceremonies as a way of making the process transparent and obviously fair in trials where whole communities rather than individuals are randomised. In this case it was in Cameroon, but the PRISM trial of improved support for new mothers in Victoria, Australia used the same approach.

• “Drawing conclusions from data”: a lecture and examples from Jonathan Stray at a Science Immersion Workshop for journalists
• Why we believe stuff — in this case high-dose vitamins.  Paul Offit in the Atlantic blames it mostly on Linus Pauling
• Using Big Data to oppress a subset of the population (vampires, via Keith Ng)
• A Herald story on meth-lab damage mentioned meth alarms, which are apparently a thing. The vendor’s website says “A silent alarm system guaranteed to detect the manufacture of methamphetamine (‘P’) in your property.” That could be valuable, but unfortunately there’s no information given on the false positive or false negative error rates, and if you read the detailed terms and conditions, the alarm system doesn’t actually seem to be guaranteed to do anything except bill you monthly.

According to The Conversation, that’s the name of a group petitioning Queensland radio station Triple J to include at least 20 seconds of science news in their news updates.  They’ve got support from one of the great science journalists, Robyn Williams, and from an ANU survey that found science news was regarded as more interesting than politics or sports news